Handbook of the Sociology of Education in the 21st Century

532 S. L. Morgan and D. T. Shackelford

Clotfelter, C. T., Ladd, H. F., Vigdor, J. L., & Diaz, R. A. Fisher, H. (1958). Teachers reading habits:A sign of profes-
(2004). Do school accountability systems make it sional interest. The Journal of Educational Sociology,
more difficult for low-performing schools to attract 32(3), 127–132. https://doi.org/10.2307/2264712.
and retain high-quality teachers? Journal of Policy
Analysis and Management, 23(2), 251–271. https:// Friedkin, N.  E., & Slater, M.  R. (1994). School leader-
doi.org/10.1002/pam.20003. ship and performance: A social network approach.
Sociology of Education, 67(2), 139–157. https://doi.
Coleman, J.  S., Campbell, E.  Q., Hobson, C.  J., org/10.2307/2112701.
McPartland, J., Mood, A.  M., Weinfeld, F.  D., &
York, R.  L. (1966). Equality of educational oppor- Fulbeck, E. S. (2014). Teacher mobility and financial incen-
tunity. Washington, DC: U.S.  Department of Health, tives: A descriptive analysis of Denver’s ProComp.
Education, and Welfare, Office of Education. Educational Evaluation and Policy Analysis, 36(1),
67–82. https://doi.org/10.3102/0162373713503185.
Cook, L.  A., Almack, R.  B., & Greenhoe, F. (1938).
Teacher and community relations. American Fulbeck, E. S., & Richards, M. P. (2015, September). The
Sociological Review, 3(2), 167–174. impact of school-based financial incentives on teach-
ers’ strategic moves: A descriptive analysis. Teachers
Cook, L. A., & Greenhoe, F. (1940). Community contacts College Record, 117, 1–36.
of 9,122 teachers. Social Forces, 19(1), 63–72. https://
doi.org/10.2307/2570843. Fuller, S. C., & Ladd, H. F. (2013). School-based account-
ability and the distribution of teacher quality across
Corcoran, S., & Goldhaber, D. (2013). Value added and grades in elementary school. Education Finance
its uses: Where you stand depends on where you sit. and Policy, 8(4), 528–559. https://doi.org/10.1162/
Education Finance and Policy, 8(3), 418–434. https:// EDFP_a_00112.
doi.org/10.1162/EDFP_a_00104.
Gamoran, A., Anderson, C. W., Secada, W. G., Williams,
Corcoran, S.  P. (2016). Potential pitfalls in the use of T., & Ashmann, S. (2003). Transforming teaching in
teacher value-added data. In J. A. Grissom & P. Youngs
(Eds.), Improving teacher evaluation systems: Making math and science: How schools and districts can sup-
the most of multiple measures (pp. 51–62). New York: port change. New York: Teachers College Press.
Teachers College Press. Gamoran, A., & Long, D.  A. (2007). Equality of edu-
cational opportunity: A 40 year retrospective. In
Crosnoe, R., Johnson, M.  K., & Elder, G.  H. (2004). R.  Teese, S.  Lamb, M.  Duru-Bellat, & S.  Helme
Intergenerational bonding in school: The behavioral (Eds.), International studies in educational inequality,
and contextual correlates of student–teacher relation- theory and policy (pp. 23–47). Dordrecht: Springer.
ships. Sociology of Education, 77(1), 60–81. Gamoran, A., Secada, W.  G., & Marrett, C.  B. (2000).
The organizational context of teaching and learning:
Edgar, D. E., & Warren, R. L. (1969). Power and auton- Changing theoretical perspectives. In M.  T. Hallinan
omy in teacher socialization. Sociology of Education, (Ed.), Handbook of the sociology of education
42(4), 386–399. https://doi.org/10.2307/2112132. (pp. 37–63). New York: Kluwer/Plenum.
Goldhaber, D., Choi, H.-J., & Cramer, L. (2007). A
Egalite, A.  J., Kisida, B., & Winters, M.  A. (2015). descriptive analysis of the distribution of NBPTS-­
Representation in the classroom: The effect of own-­ certified teachers in North Carolina. Economics
race teachers on student achievement. Economics of Education Review, 26(2), 160–172. https://doi.
of Education Review, 45, 44–52. https://doi. org/10.1016/j.econedurev.2005.09.003.
org/10.1016/j.econedurev.2015.01.007. Goldhaber, D., Destler, K., & Player, D. (2010). Teacher
labor markets and the perils of using hedonics to esti-
Ehrenberg, R.  G., & Brewer, D.  J. (1994). Do school mate compensating differentials in the public sector.
and teacher characteristics matter? Evidence Economics of Education Review, 29(1), 1–17. https://
from high school and beyond. Economics of doi.org/10.1016/j.econedurev.2009.07.010.
Education Review, 13(1), 1–17. https://doi. Goldhaber, D., Gross, B., & Player, D. (2011). Teacher
org/10.1016/0272-7757(94)90019-1. career paths, teacher quality, and persistence in the
classroom: Are public schools keeping their best?
Elfers, A.  M., Plecki, M.  L., & Knapp, M.  S. (2006). Journal of Policy Analysis and Management, 30(1),
Teacher mobility: Looking more closely at “the 57–87. https://doi.org/10.1002/pam.20549.
movers” within a state system. Peabody Journal of Gordon, C.  W. (1955). The role of the teacher in the
Education, 81(3), 94–127. social structure of the high school. The Journal of
Educational Sociology, 29(1), 21–29. https://doi.
Feng, L. (2010). Hire today, gone tomorrow: New org/10.2307/2263348.
teacher classroom assignments and teacher mobility. Greenberg, D., & McCall, J. (1974). Teacher mobility and
Education Finance and Policy, 5(3), 278–316. https:// allocation. The Journal of Human Resources, 9(4),
doi.org/10.1162/EDFP_a_00002. 480–502. https://doi.org/10.2307/144782.
Grissom, J.  A., & Youngs, P. (Eds.). (2016). Improving
Feng, L. (2014). Teacher placement, mobility, and occu-
pational choices after teaching. Education Economics, teacher evaluation systems: Making the most of mul-
22(1), 24–47. https://doi.org/10.1080/09645292.2010 tiple measures. New York: Teachers College Press.
.511841. Guarino, C.  M., Brown, A.  B., & Wyse, A.  E. (2011).
Can districts keep good teachers in the schools
Ferguson, R. F., & Hirsch, E. (2014). How working condi-
tions predict teaching quality and student outcomes.
In T.  J. Kane, K.  A. Kerr, & R.  C. Pianta (Eds.),

Designing teacher evaluation systems: New guidance
from the measures of effective teaching project (1st
ed., pp. 332–380). San Francisco: Jossey-Bass.

23  School and Teacher Effects 533

that need them most? Economics of Education Kalogrides, D., Loeb, S., & Béteille, T. (2013). Systematic
Review, 30(5), 962–979. https://doi.org/10.1016/j. sorting: Teacher characteristics and class assignments.
econedurev.2011.04.001. Sociology of Education, 86(2), 103–123. https://doi.
Guarino, C.  M., Reckase, M.  D., & Wooldridge, J.  M. org/10.1177/0038040712456555.
(2014). Can value-added measures of teacher per-
formance be trusted? Education Finance and Koedel, C. (2009). An empirical analysis of teacher
Policy, 10(1), 117–156. https://doi.org/10.1162/ spillover effects in secondary school. Economics
EDFP_a_00153. of Education Review, 28(6), 682–692. https://doi.
Hallinan, M.  T. (2008). Teacher influences on students’ org/10.1016/j.econedurev.2009.02.003.
attachment to school. Sociology of Education, 81(3),
271–283. Krei, M. S. (1998). Intensifying the barriers: The problem
Hanushek, E.  A., Kain, J.  F., & Rivkin, S.  G. (2004). of inequitable teacher allocation in low-income urban
Why public schools lose teachers. The Journal of schools. Urban Education, 33(1), 71–94. https://doi.
Human Resources, 39(2), 326–354. https://doi. org/10.1177/0042085998033001005.
org/10.2307/3559017.
Hedges, L.  V., & Schneider, B.  L. (Eds.). (2005). The Kukla-Acevedo, S. (2009). Do teacher characteristics
social organization of schooling. New  York: Russell matter? New results on the effects of teacher prepara-
Sage Foundation. tion on student achievement. Economics of Education
Horng, E.  L. (2009). Teacher tradeoffs: Disentangling Review, 28(1), 49–57. https://doi.org/10.1016/j.
teachers’ preferences for working conditions and econedurev.2007.10.007.
student demographics. American Educational
Research Journal, 46(3), 690–717. https://doi. Ladd, H.  F. (2008). Reflections on equity, adequacy,
org/10.3102/0002831208329599. and weighted student funding. Education Finance
Ingersoll, R. M. (1996). Teachers’ decision-making power and Policy, 3(4), 402–423. https://doi.org/10.1162/
and school conflict. Sociology of Education, 69(2), edfp.2008.3.4.402.
159–176. https://doi.org/10.2307/2112804.
Ingersoll, R.  M. (2005). The problem of underqualified Lankford, H., Loeb, S., & Wyckoff, J.  (2002).
teachers: A sociological perspective. Sociology of Teacher sorting and the plight of urban schools:
Education, 78(2), 175–178. A descriptive analysis. Educational Evaluation
Ingersoll, R.  M., & May, H. (2012). The magnitude, and Policy Analysis, 24(1), 37–62. https://doi.
destinations, and determinants of mathematics and org/10.3102/01623737024001037.
science teacher turnover. Educational Evaluation
and Policy Analysis, 34(4), 435–464. https://doi. Larkin, R.  W. (1973). Contextual influences on teacher
org/10.3102/0162373712454326. leadership styles. Sociology of Education, 46(4), 471–
Jackson, C.  K. (2009). Student demographics, teacher 479. https://doi.org/10.2307/2111900.
sorting, and teacher quality: Evidence from the end
of school desegregation. Journal of Labor Economics, Lee, V.  E., Dedrick, R.  F., & Smith, J.  B. (1991). The
27(2), 213–256. https://doi.org/10.1086/599334. effect of the social organization of schools on teach-
Jackson, C. K. (2013). Match quality, worker productivity, ers’ efficacy and satisfaction. Sociology of Education,
and worker mobility: Direct evidence from teachers. 64(3), 190–208. https://doi.org/10.2307/2112851.
Review of Economics and Statistics, 95(4), 1096–
1116. https://doi.org/10.1162/REST_a_00339. Lee, V.  E., & Smith, J.  B. (1993). Effects of school
Jackson, C.  K., & Bruegmann, E. (2009). Teaching stu- restructuring on the achievement and engagement of
dents and teaching each other: The importance of peer middle-grade students. Sociology of Education, 66(2),
learning for teachers. American Economic Journal: 164–187.
Applied Economics, 1(4), 85–108.
Jacob, B.  A., & Lefgren, L. (2007). What do parents Lee, V. E., & Smith, J. B. (1995). Effects of high school
value in education? An empirical investigation of par- restructuring and size on early gains in achievement
ents’ revealed preferences for teachers. The Quarterly and engagement. Sociology of Education, 68(1),
Journal of Economics, 122(4), 1603–1637. 241–270.
Jennings, J. L., & DiPrete, T. A. (2010). Teacher effects
on social and behavioral skills in early elementary Levinson, A.  M. (1988). Reexamining teacher pref-
school. Sociology of Education, 83(2), 135–159. erences and compensating wages. Economics
Jepsen, C., & Rivkin, S. (2009). Class size reduction and of Education Review, 7(3), 357–364. https://doi.
student achievement: The potential tradeoff between org/10.1016/0272-7757(88)90007-6.
teacher quality and class size. The Journal of Human
Resources, 44(1), 223–250. Loeb, S., Darling-Hammond, L., & Luczak, J.  (2005).
Jessup, D. K. (1978). Teacher unionization:A reassessment How teaching conditions predict teacher turn-
of rank and file motivations. Sociology of Education, over in California schools. Peabody Journal of
51(1), 44–55. https://doi.org/10.2307/2112281. Education, 80(3), 44–70. https://doi.org/10.1207/
s15327930pje8003_4.

Loeb, S., Kalogrides, D., & Béteille, T. (2012). Effective
schools: Teacher hiring, assignment, development,
and retention. Education Finance and Policy, 7(3),
269–304. https://doi.org/10.1162/EDFP_a_00068.

McCaffrey, D. F., Sass, T. R., Lockwood, J. R., & Mihaly,
K. (2009). The intertemporal variability of teacher
effect estimates. Education Finance and Policy, 4(4),
572–606. https://doi.org/10.1162/edfp.2009.4.4.572.

Moller, S., Mickelson, R. A., Stearns, E., Banerjee, N., &
Bottia, M.  C. (2013). Collective pedagogical teacher

534 S. L. Morgan and D. T. Shackelford

culture and mathematics achievement: Differences by unobservables. Education Finance and Policy, 4(4),
race, ethnicity, and socioeconomic status. Sociology of 537–571. https://doi.org/10.1162/edfp.2009.4.4.537.
Education, 86(2), 174–194. Rothstein, J.  (2010). Teacher quality in educational pro-
Morgan, S.  L., & Jung, S.  B. (2016). Still no effect of duction: Tracking, decay, and student achievement.
resources, even in the new gilded age? RSF: The The Quarterly Journal of Economics, 125(1), 175–214.
Rowan, B., Chiang, F.-S., & Miller, R.  J. (1997). Using
Russell Sage Foundation Journal of the Social research on employees’ performance to study
Sciences, 2(5), 83–116. the effects of teachers on students’ achievement.
Myers, A. F. (1934). Education of teachers for the schools Sociology of Education, 70(4), 256–284. https://doi.
of tomorrow. The Journal of Educational Sociology, org/10.2307/2673267.
7(9), 569–574. https://doi.org/10.2307/2961057. Rubenstein, R., Schwartz, A. E., Stiefel, L., & Amor, H. B.
Ost, B., & Schiman, J.  C. (2015). Grade-specific expe- H. (2007). From districts to schools: The distribution
rience, grade reassignments, and teacher turnover. of resources across schools in big city school districts.
Economics of Education Review, 46, 112–126. https:// Economics of Education Review, 26(5), 532–545.
doi.org/10.1016/j.econedurev.2015.03.004. https://doi.org/10.1016/j.econedurev.2006.08.002.
Player, D. (2009). Monetary returns to academic abil- Scafidi, B., Sjoquist, D. L., & Stinebrickner, T. R. (2007).
ity in the public teacher labor market. Economics Race, poverty, and teacher mobility. Economics
of Education Review, 28(2), 277–285. https://doi. of Education Review, 26(2), 145–159. https://doi.
org/10.1016/j.econedurev.2008.06.002. org/10.1016/j.econedurev.2005.08.006.
Raudenbush, S. W., Rowan, B., & Cheong, Y. F. (1992). Sieber, S.  D., & Wilder, D.  E. (1967). Teaching styles:
Contextual effects on the self-perceived efficacy of Parental preferences and professional role definitions.
high school teachers. Sociology of Education, 65(2), Sociology of Education, 40(4), 302–315. https://doi.
150–167. https://doi.org/10.2307/2112680. org/10.2307/2111938.
Ravitch, D. (1993). The Coleman reports and American Sørensen, A. B., & Morgan, S. L. (2000). School effects:
education. In A. B. Sørensen & S. Spilerman (Eds.), Theoretical and methodological issues. In M.  T.
Hallinan (Ed.), Handbook of the sociology of educa-
Social theory and social policy: Essays in honor of tion (pp. 137–160). New York: Kluwer/Plenum.
James S. Coleman (pp. 129–141). Westport: Praeger. Steele, J. L., Pepper, M. J., Springer, M. G., & Lockwood,
Reardon, S. F., & Raudenbush, S. W. (2009). Assumptions J. R. (2015). The distribution and mobility of effective
of value-added models for estimating school effects. teachers: Evidence from a large, urban school district.
Education Finance and Policy, 4(4), 492–519. https:// Economics of Education Review, 48, 86–101. https://
doi.org/10.1162/edfp.2009.4.4.492. doi.org/10.1016/j.econedurev.2015.05.009.
Rice, J. K. (2013). Learning from experience? Evidence Waller, W. (1932). The sociology of teaching. New York:
on the impact and distribution of teacher experience Wiley.
and the implications for teacher policy. Education Winters, M. A., Haight, R. C., Swaim, T. T., & Pickering,
Finance and Policy, 8(3), 332–348. https://doi. K.  A. (2013). The effect of same-gender teacher
org/10.1162/EDFP_a_00099. assignment on student achievement in the elemen-
Rosenholtz, S. J., & Simpson, C. (1990). Workplace con- tary and secondary grades: Evidence from panel data.
ditions and the rise and fall of teachers’ commitment. Economics of Education Review, 34, 69–75. https://
Sociology of Education, 63(4), 241–257. https://doi. doi.org/10.1016/j.econedurev.2013.01.007.
org/10.2307/2112873. Yasumoto, J.  Y., Uekawa, K., & Bidwell, C.  E. (2001).
Roth, L.  J. (1958). Occupational analysis and teacher The collegial focus and high school students’ achieve-
morale. The Journal of Educational Sociology, 32(4), ment. Sociology of Education, 74(3), 181–209. https://
145–151. https://doi.org/10.2307/2264177. doi.org/10.2307/2673274.
Rothstein, J.  (2009). Student sorting and bias in value-­
added estimation: Selection on observables and

Experimental Evidence 24
on Interventions to Improve
Educational Attainment
at Community Colleges

David Monaghan, Tammy Kolbe,
and Sara Goldrick-Rab

Abstract 24.1 Introduction
America’s community colleges play a major
role in increasing access to higher education Community colleges play a critical role in higher
and, as open access institutions, they are key education. Intended to provide accessible, flexi-
points of entry to postsecondary education for ble, and affordable opportunities for postsecond-
historically underrepresented populations. ary education and workforce participation, they
However, their students often fall short of have contributed to substantial increases in col-
completing degrees. Policymakers, scholars, lege participation. This is especially true for
and philanthropists are dedicating unprece- groups who are historically underrepresented in
dented attention and resources to identifying postsecondary education—including racial and
strategies to improve retention, academic per- ethnic minority, low-income, part-time, first-­
formance, and degree completion among generation, and adult students. Today, almost
community college students. This chapter 40% of all undergraduates—more than 6.6  mil-
reviews experimental evidence on their effec- lion Americans—attend community colleges
tiveness, finding that they often meet with lim- (Kena et  al. 2015). However, increased college
ited success because they typically target just enrollment does not consistently translate into
one or two aspects of students’ lives, are of program or degree completion. Completion rates
short duration, and fail to improve the institu- among community college students—as mea-
tional context. They also rarely address a seri- sured by earning a credential or transferring to a
ous structural constraint: limited resources. four-year institution—are less than 50% after
We discuss new directions for future interven- 6  years of enrollment and below 30% for low-­
tions, research and evaluation. income, Black, Latino, and Native American stu-
dents (Shapiro et  al. 2014). Fewer than two in
D. Monaghan (*) five community college students who enter with
Shippensburg University of Pennsylvania, the intent to earn some type of a degree do so
Shippensburg, PA, USA within six  years of initial enrollment (Shapiro
e-mail: [email protected] et  al. 2014) and only three in five enroll in any
college one  year later (National Student
T. Kolbe Clearinghouse Research Center 2015).
University of Vermont, Burlington, VT, USA
e-mail: [email protected] While even some college education appears to
benefit students, degree completion is essential,
S. Goldrick-Rab especially if students must accrue debt along the
Temple University, Philadelphia, PA, USA
e-mail: [email protected]

© Springer International Publishing AG, part of Springer Nature 2018 535
B. Schneider (ed.), Handbook of the Sociology of Education in the 21st Century, Handbooks
of Sociology and Social Research, https://doi.org/10.1007/978-3-319-76694-2_24

536 D. Monaghan et al.

way in order to cover college prices (Goldrick-­ 24.2 The Contexts of Community
College Education
Rab 2016). Low completion rates coupled with

substantial lag times between enrollment and

completion levy real economic and social costs Unlike other higher education institutions, com-

(Goldrick-Rab 2016; Bailey et  al. 2004). Since munity colleges were explicitly designed as open

they broaden access, community colleges appear entry-points into higher education, emphasizing

to substantially raise the educational attainment expanded opportunities for all rather than maxi-

of those otherwise unlikely to attend college at mizing outcomes for a few. In the aftermath of

all, while doing very little harm to students who World War II, the Truman Commission (1947)

might otherwise attend four-year colleges (Leigh called for action to democratize higher education,

and Gill 2003; Brand et al. 2012). Scholars, poli- postsecondary enrollments surged, and higher

cymakers, and philanthropic foundations are education leaders sought a means of satisfying

devoting unprecedented attention and resources popular pressure for access while protecting cur-

to identifying strategies to boost retention and ricular rigor at existing institutions (Brint and

degree completion among community college Karabel 1989; Trow 2007). In response, the

students (Bailey et  al. 2015; Grossman et  al. nation’s existing “junior colleges” were rechris-

2015; Sturgis 2014). These interventions address tened as “community colleges” and their ranks

a wide variety of conditions and contexts—at the dramatically expanded. Community colleges

individual, school, and system levels—believed would serve their purpose as “agents of democ-

to pose barriers to student success. Many efforts racy” by being both accessible and comprehen-

have been evaluated to assess their effectiveness. sive, within the bounds of their resource

In doing so, researchers have increasingly relied constraints.

upon randomized control trials (RCTs) to gener- Community colleges aim to minimize three

ate rigorous estimates of causal effects, providing barriers to college entry: price, academic require-

insights into “what works” to boost attainment ments, and distance. They are intended to be

among community college students. cheaper than four-year public colleges; open-­

This chapter reviews evidence from experi- enrollment, requiring that prospective students

mental evaluations of a range of interven- complete high school to gain admission; and geo-

tions—from financial aid to student graphically dispersed so they are within reason-

advisement to developmental education— able commuting distance for all Americans. As

with two main goals. First, we examine what public higher education institutions, community

the evidence from experimental studies colleges are primarily funded by state and local

reveals about the most promising interven- revenues. Historically, public funding sources

tions. It is evident that while sustained and have buoyed costs, keeping tuition non-existent or

multi-pronged strategies appear most effec- very low for students. Low-cost educational

tive at boosting completion, they are also opportunities reinforced community colleges’

uncommon. Second, we illustrate the role missions as open access portals for a broad range

research and evaluations incorporating exper- of students, including those who could not afford

imental design should play in future socio- higher education at other public or private institu-

logical research, especially when assessing tions. Community colleges are comprehensive in

the impact of education and social program their offerings, reflecting the range of needs and

interventions targeted at disadvantaged youth, interests of the community they served. They have

adults, and families. Effectively replicating or academic courses for students intending to transfer

scaling programs requires that future studies to four-year colleges, vocational training programs

more carefully document the context in which for students looking to upgrade skills or change

an intervention succeeded or failed, and the jobs, and general education courses for commu-

resources and costs involved. nity members interested in lifelong learning.

24  Experimental Evidence on Interventions to Improve Educational Attainment at Community Colleges 537

Aspects of the community college context— a holding area for students from the “educational

those related to accessibility and comprehensive- underclass” (Deil-Amen and DeLuca 2010).

ness in particular—may work at cross-purposes Part of accessibility is geographic dispersion.

with the goal of maximizing completion rates. Community colleges are, with few exceptions,

For example, open enrollment and a relatively commuter campuses. Sociological and education

low cost of attendance help attract a more hetero- research suggests that students who reside on

geneous mix of students, compared to those who campus are more likely to remain enrolled and to

attend four-year colleges and universities. eventually graduate, as such students spend more

Community college students are disproportion- time on campus and are far more likely to become

ately Black and Latino and are far more likely to socially and academically integrated into the life

be a first-generation college-goer or from a of the institution (Astin 1984; Pascarella and

lower-income household. Given open-­Terenzini 2005; Schudde 2011; Tinto 1987). In

admissions, community college students have on addition, part-time and part-year community col-

average lower levels of academic preparation and lege enrollment is the norm—fewer than half of

fewer resources than students attending four-year community college students enroll either full-­

public and non-profit colleges and universities time in both fall and spring semesters (Table 24.1).

(Table 24.1). Indeed, despite the constant charac- Indeed, many community college students are on

terization that they are “diverse” spaces, in fact campus only a few hours per week, giving the

community colleges are highly segregated colleges few opportunities to directly engage

(Goldrick-Rab and Kinsley 2013). them and build institutional loyalty or involve-

Stratification by student composition, and by ment. The low-intensity student enrollment pat-

extension aspirations and outcomes, translates terns reflect community college students’

into vastly different educational experiences and, “non-traditional” status. Most are older than 23

by extension, differences in opportunities and and, for many, college is negotiated along with

outcomes. Moreover, the effects of segregating full-time work and childcare responsibilities

students across institutional types may be exacer- (Table 24.1; Stuart et al. 2014). Students are fre-

bated by peer-effects. If having more uniformly quently not exclusively or even primarily ori-

poorer, less-prepared peers who are more likely ented towards college-going, and practically

to drop out of college impacts the social and speaking completing a degree is often not their

intellectual atmosphere and normalizes non-­ top priority. As a result, it is common for

completion, then community college students ­community college students to “stop out” for a

may be at a particular disadvantage (Century semester or two to attend to other responsibilities

Foundation 2013). or to transfer when another college is more con-

Because community colleges enroll many stu- venient (Bahr 2009; Crosta 2014).

dents without the skills needed to assimilate The effects of more limited opportunities for

college-­level material, remedial education has interaction with faculty are compounded by

been central to them since their inception (Cohen increasing student–staff ratios. Typically, the ratio

et  al. 2014). Remedial policies effectively bar of student support and other college staff is dou-

low-performing students from most classes ble that found at four-year institutions (Baum and

bestowing credit (Hughes and Scott-Clayton Kurose 2013). Staffing shortages are particularly

2011; Perin 2006), and the majority of students dire for student advisement and counseling; at

never complete the sequences of remedial courses community colleges, student-to-counselor ratios

to which they are assigned (Bailey et al. 2010). are frequently higher than 800: or 1000:1 (Park

Assignment to remediation substantially et al. 2013), resulting in inadequate, inconsistent,

increases the cost of a degree in terms of time and and often counterproductive academic and career

money (Melguizo et al. 2008). Critics allege that counseling (Deil-Amen and Rosenbaum 2002,

remedial regimes permit the colleges to maintain 2003; Grubb 2001, 2006; Rosenbaum et  al.

appearance of access, while effectively serving as 2006). Personalized counseling and advisement

538 D. Monaghan et al.

Table 24.1  Student characteristics, by sector (2012)

Female Community Public Private non-profit Private for-profit (two- or
White college four-year four-year four-year)
Black 55.7 53.9 56.6 64.1
Latino/a 55.8 62.2 65.1 48.5
Asian 16.4 12.8 13.4 25.6
First-generation 18.6 13.8 10.1 18.5
college
Income < 200% 5.0 6.9 6.9 2.9
povertya 46.3 31.7 28.3 58.1
HS GPA < 3.0a
HS math: Alg. 2/ 54.9 34.8 30.3 81.2
lessa
Didn’t take SAT/ 50.3 27.6 18.7 52.7
ACTa 61.3 26.5 21.3 71.0
Lowest quartile SATa
Age 24+ 29.7 3.8 2.0 38.7
Has children
Single parent 38.9 16.4 12.9 43.4
Living on-campus 50.8 30.4 28.8 68.4
Enrolled full-time 32.4 15.4 17.0 51.6
Enrolled full-year 17.9 32.7
Employed 7.3 8.1
Employed full-time 0.9 22.5 45.0 0.9
32.5 59.5 73.5 70.7
46.6 69.6 72.2 42.4
68.6 65.5 63.7 61.0
26.2 15.3 14.5 31.2

aAmong first time freshmen only; Sources: Beginning Postsecondary Students Longitudinal Survey 2012/14; National
Postsecondary Student Aid Survey 2012

is crucial to community college student success. courses, the diversity of programs also causes
Instead, community college counseling offices tri- problems for coordinating and scheduling
age counseling services according to student courses. This poses challenges for part-time and
needs, and devote a limited amount of time to working students to arrange classes in ways that
each student during heavy-use periods such as fit their schedule. For students attending commu-
registration. nity college as an entry point to a bachelor’s
degree, the “cafeteria self-service” course-taking
These challenges are complicated by the role model found at many community colleges also
that community colleges play as comprehensive can pose challenges for identifying a clear and
institutions that try to expand college enrollment efficient pathway to a four-year degree.
through a wide range of courses, degree, and cer- Oftentimes, incoming students do not know to
tificate programs. However, over time, at many which four-year institutions they will apply, or
community colleges, “comprehensiveness” has even the general requirements for transfer to a
translated into an array of often disconnected bachelor’s degree-granting institution. The
courses, programs, and support services that stu- absence of strong articulation policies that link
dents must navigate with relatively little guid- community colleges with four-year institutions—
ance (Bailey et al. 2015). The immense array of even public ones—means that the four-year col-
choices can overwhelm students, especially first- leges often differ in the courses required for
generation college students and others with lim- transfer and in the courses they will recognize by
ited experience with postsecondary education transferring credit. For community college stu-
(Rosenbaum et  al. 2006; Scott-Clayton 2015). dents who do manage to transfer, substantial loss
Since each program has its own set of required

24  Experimental Evidence on Interventions to Improve Educational Attainment at Community Colleges 539

of credits is a common occurrence (Monaghan resources is to spend less per student. Per-student
and Attewell 2015; Simone 2014). instructional spending at community colleges fell
by 12% between 2001 and 2011; on average,
Recently, researchers and outside experts have community colleges now spend 78% as much per
suggested that community colleges narrow their student on instruction as public bachelor’s col-
program structures in ways that faculty clearly leges and 56% as much as public research univer-
map out academic programs to create coherent sities, despite enrolling students with arguably
pathways that are aligned with requirements for greater academic challenges (Desrochers and
further education and career advancement (Bailey Kirshstein 2012).
et  al. 2015). This involves presenting students
with a small number of program options, devel- Lower spending may impact educational
oping clear course sequences leading to degree quality and output (Jenkins and Belfield 2014).
completion, arranging the courses so that they are The student–faculty ratio at community colleges
convenient (i.e., scheduled back-to-back), and was 22:1 in 2009, while at public four-year col-
providing personalized, mandatory counseling leges this ratio was 15:1 (Baum and Kurose
services. They contend that community colleges 2013). The higher ratio constrains the amount of
treat their clientele as if they were “traditional time faculty can devote to individual students
college students,” equipped with the motivation, and may affect instructional quality. This is par-
knowledge, and skills necessary to negotiate col- ticularly problematic in a commuter setting,
lege. Change the community colleges’ programs, where classroom time and faculty are the pri-
practices and resources, they argue, and one can mary opportunity for “socio-a­cademic integra-
improve student outcomes (Bailey et  al. 2015; tive moments” (Deil-Amen 2011). The impact
Rosenbaum et al. 2006; Scott-Clayton 2015). of higher faculty–student ratios is further exac-
erbated by community colleges’ other cost-sav-
But it is increasingly difficult to maintain both ing strategy: heavy reliance on part-time and
accessibility and comprehensiveness while also contingent faculty. At community colleges, two-
increasing completion rates as state governments thirds of faculty work part time, and only 18%
have reduced support on a per-student basis are tenured or tenure-track (Kezar and Maxey
(Goldrick-Rab 2016). While, on a per-student 2013). Exposure to part-time and adjunct fac-
basis, community colleges receive about as much ulty is negatively associated with degree com-
money from states as do public comprehensives pletion (Eagan and Jaeger 2009). Contingent
(Baum and Kurose 2013; College Board 2015), faculty may not have the institutional knowl-
they are far more dependent on state funding as a edge and skills to help students negotiate the
primary source of revenue. On average, commu- institution and contribute to short-term
nity colleges receive about 71% of their revenue faculty–student relationships that do not last
from state appropriations, compared to public beyond a semester.
four-year colleges’ 38% (Kena et al. 2015). This
makes community colleges particularly vulnera- Research diligently minimizing selection bias
ble to state cuts—particularly during recession- has consistently found negative impacts of initial
ary periods, which tend to couple funding community college enrollment, relative to four-­
reductions with enrollment surges (Betts and year college enrollment, on bachelor’s degree
McFarland 1995). Community colleges pass attainment (Brand et  al. 2012; Reynolds 2012;
some costs on to students. Between 2000 and but see Rouse 1995). But the 60% one-year reten-
2010 the percentage of revenue covered by state tion rate at community colleges is not apprecia-
appropriations fell from 57% to 47%. At the same bly different from that at non-selective public or
time, that met through tuition and fees rose from non-profit four-year colleges (62% and 61%
19% to 27% (Kirshstein and Hubert 2012). As respectively) (Kena et al. 2015). Monaghan and
community colleges’ capacity to increase tuition Attewell (2015), comparing community college
is constrained by their mandate to remain afford- students with those at non-selective four-year
able, their principal response to dwindling colleges, find that retention differences do not

540 D. Monaghan et al.

appear until the fifth semester, after adjusting for taking out loans, working long hours, or enrolling
student characteristics. part-time. Other interventions seek to improve
students’ “informational capital”—the knowl-
In summary, the community college sector edge required to select a major, choose the cor-
arose to accommodate demands to democratize rect classes that will enable them to complete the
access to higher education and offer a compre- major, or apply for financial aid (Rosenbaum
hensive battery of general education, vocational, et  al. 2006). Such efforts include informational
and academic options. The “imperious immedi- seminars, orientation courses, and counseling
acy of interest” (Merton 1936) in achieving these services. Similarly, institutions also intervene to
goals obscured the consideration of whether their address student academic shortfalls through man-
resulting organizational features might stymie datory remedial courses and through voluntarily-­
degree completion. Early critics alleged that accessed tutoring and writing centers. Finally,
community colleges “cooled out” the aspirations given that many community college students
of academically disinclined and/or lower-SES enroll part-time, interventions have been devel-
youth by tracking them into vocational programs oped that encourage full-time attendance or sum-
or permitting them to drop out altogether (Brint mer course-taking.
and Karabel 1989, Clark 1960). But it wasn’t
until the late 1990s that their low completion School-focused interventions attempt to
rates came to be collectively defined, in Blumer’s change how community colleges serve students
(1971) sense, as a social problem in need of a (e.g., guided pathways, course redesign, or struc-
solution. In response, policymakers, educational turing of support services). One frequent leverage
leaders and philanthropists have targeted their point has been the college counseling center:
efforts at new opportunities to restructure how assigning each student to a counselor, making
community colleges deliver education and sup- appointments mandatory, lowering student–coun-
port services, with an eye towards identifying selor ratios, and having counselors specialize by
reforms that improve both the effectiveness and degree program. Other interventions include forg-
efficiency with which they support not only ing social connections among students through
access to higher education but also completion, linked courses or “learning communities,” and by
for all students. building students’ connections to the institution
through providing in-class tutors or mentors.
24.3 Points of Intervention Skills assessment and remedial coursework also
has been an area for reform. Many community
Efforts to improve outcomes in community col- college students enter who are not “college ready,”
leges are focused on either the student, the insti- so remediation is widespread; it is estimated that
tution, or the system (Goldrick-Rab 2010). between 60% and 70% take at least one remedial
Student-focused interventions reduce financial course at some point (Crisp and Delgado 2014;
barriers, provide student support, or improve stu- Radford and Horn 2012).
dents’ academic skills (e.g., dual enrollment pro-
grams, financial aid, advising, or coaching). System-level interventions alter community
Financial aid is by far the most popular strategy. colleges’ incentive structures, the financial struc-
Community college tuition is relatively low; yet, tures that govern them, or the landscape in which
many students still struggle with paying for col- they operate. Statewide policies that provide free
lege as well as other living costs incurred while in or reduced tuition, such as “promise programs”
school. It is hoped that by putting in place pro- (e.g., Tennessee and Oregon) reorient the nature
grams and resources that alleviate material short- of community colleges in the higher education
ages and reducing stress, financial aid may enable system hierarchy (Miller-Adams 2015).
students to focus on academic work and to avoid Performance-based funding has been used by
potentially injurious alternative strategies such as states to encourage community colleges to orient
programs and resources toward specified goals
and metrics, oftentimes closely aligned with stu-

24  Experimental Evidence on Interventions to Improve Educational Attainment at Community Colleges 541

dent outcomes (Dougherty et  al. 2016; Hillman college reforms. To a great extent, their applica-

et  al. 2014). State articulation policies hold the tion has been in response to efforts on the part of

promise of smoothing transfer from community the U.S.  Department of Education’s Institute of

colleges into four-year institutions. Educational Sciences attention to causal research

and the corresponding shift in federal research

24.4 Evaluating Community dollars. At the same time, new institution-level
College Reforms data on student outcomes became readily avail-
able, drawing public attention to the considerable

gaps in college completion rates between com-

Until very recently, social scientists sought to munity college students and their peers at other

understand the community college through natu- higher education institutions (Bailey et al. 2015).

ralistic observation rather than measuring inter- As a result, the “College Completion Agenda”

vention impacts. However, such approaches are began to coalesce in the early years of the new

limited in their capacity to provide rigorous esti- decade, and philanthropic foundations added

mates of causal effects (Morgan and Winship their millions of private money to the public

2014). When participation in an intervention is money already earmarked for experimental eval-

voluntary, those who choose to participate tend to uation research.

differ in measureable and unmeasurable ways In this chapter, we catalogue randomized con-

from non-participants. As a result, it is difficult to trol trials in community college settings. Eligible

disentangle the intervention’s independent studies were identified by (1) searching Google

impacts from selection bias introduced through Scholar, the Web of Science, EconLit, Social

these baseline differences. Random assignment Sciences Full Text, Education Full Text, and the

to treatment ensures that differences between American Economic Association’s RCT Registry

treated and untreated individuals arise only from with combinations of keywords (experiment, ran-

chance and are unlikely to be considerable given domized control trial, community college, and

large enough samples (Rubin 1974). For this rea- two-year college); (2) scouring websites of eval-

son, randomized experiments permit unbiased, uation organizations such as MDRC and

internally valid, and truly causal estimates of Mathematica; (3) searching programs of research

treatment effects. conferences such as SREE, APPAM, and AEA;

But as with all methods, randomized control and (4) making inquiries among scholars knowl-

trials have limitations. Some questions cannot be edgeable in the field. Studies that met the follow-

answered by experimental evaluations, for rea- ing criteria were included:

sons of feasibility and ethics (Heckman 2005;

Lareau 2008). Additionally, unforeseen issues in • Subjects were assigned to intervention or con-

program implementation and participant behav- trol condition using random assignment;

ior after randomization can have substantial • Subjects were entering or presently enrolled at

impacts on treatment effects (Lareau 2008; community colleges, either exclusively or as a

Heckman and Smith 1995). Finally, experiments major sub-population; and

tell us little about why causes produce their • Interventions were aimed at improving aca-

effects, though additional non-causal evidence on demic outcomes such as retention, credit

mechanisms can be gathered using mixed-­ accumulation, academic performance, and

methods approaches (Grissmer et al. 2009; Harris degree completion.

and Goldrick-Rab 2012). They also tell us little

about how the context in which an intervention Given these criteria, we excluded observa-

occurred may have impacted its outcomes. tional studies, including those employing rigor-

Despite limitations, over the past 15  years, ous quasi-experimental designs, except to provide

experimental evaluations have increasingly been context for experimental interventions.

used to understand the impacts of community Interventions where subjects could not be ran-

542 D. Monaghan et al.

domized, such as those altering institutional or lege progress, naïve estimations of aid effects
policy frameworks, were excluded. Also excluded tend to be biased. Most research leverages “natu-
were interventions intending to impact whether ral experiments” such as aid cutoffs, program ter-
or where individuals choose to enroll in college. minations, and tuition reductions in order to
Given the fiscal constraints under which commu- identify causal effects (Alon 2011; Bettinger
nity colleges operate, knowing the cost of an 2004, 2015; Castleman and Long 2013; Denning
intervention is crucial for evaluating its realistic 2014; Dynarski 2003; Kane 2003; Singell 2004;
potential to be adopted at scale (Belfield et  al. Van der Klaauw 2002). Such studies have tended
2014; Belfield and Jenkins 2014; Schneider and to find that an increase in aid of $1000 increases
McDonald 2007a, b). Therefore, wherever possi- persistence by 2–4 percentage points, and degree
ble, a discussion of costs is included alongside completion by between 1.5 and 5 percentage
the assessment of impacts. However, to a large points (for reviews see Bettinger (2012), Deming
extent this information is notably missing from and Dynarski (2010), Dynarski and Scott-Clayton
extant research (Belfield 2015). (2013), and Goldrick-Rab et al. (2009)).

In total, we identified 30 studies of commu- To date there have been seven randomized
nity college interventions that met the selection experiments examining financial aid in commu-
criteria. In addition, we included seven in prog- nity college contexts. Two are evaluations of pri-
ress studies to give a sense of the future of this vately funded, need-based scholarships affecting
research. Next, the studies are discussed accord- both four-year and two-year students: Angrist
ing to their level of intervention. and associates’ (2015) evaluation of the Buffett
Scholarship, and Goldrick-Rab and associates’
24.5 S tudent-Level Interventions investigation of the Wisconsin Scholars Grant
(WSG) (N = 2641 and N = 12,722 respectively).
Student-level interventions work principally to Both scholarships targeted low- to moderate-­
augment the resources or change the behavior of income students, had high school GPA eligibility
individual community college students, while requirements, and were restricted to residents of
leaving the prevailing institutional environment a particular state (Nebraska and Wisconsin,
unchanged. As such, they seek to improve the respectively) who attended in-state public col-
capacity of individuals to navigate an environ- leges. The Buffett Scholarship is designed to
ment which is taken as given. Individual-level fully cover tuition and fees; two-year recipients
interventions are often, but need not necessarily were awarded as much as $5300 per year for up
be, prefaced on an assumption that individual to 5  years. In contrast, the WSG is designed to
deficits are at the root of outcomes deemed unac- reduce rather than eliminate tuition expenses; the
ceptable. We identified 14 such interventions. yearly award was $1800 for two-year students,
for up to 5 years. Another crucial difference is the
24.5.1 F inancial Aid timing of scholarship. The Buffett Scholarship is
awarded prior to enrollment, and thus can impact
The primary policy effort to raise community individuals’ choice of college, whereas the WSG
college completion rates is financial aid, and is awarded towards the end of the recipient’s first
nationwide governments spend about $57 billion semester.
on grant aid and another $96  billion on loans
(College Board 2015). But establishing the causal Both studies found measurable positive
impact of financial aid on college persistence and impacts for the full population of recipients, but
completion is not straightforward. Since the same null or negative results for initial two-year
trait—financial need—which renders a student enrollers. Anderson and Goldrick-Rab (2016)
eligible for financial aid also tends to disrupt col- estimate that the WSG increased one-year reten-
tion by 3.7 percentage points at University of
Wisconsin branch campuses and decreased it by
1.5 percentage points at Wisconsin Technical

24  Experimental Evidence on Interventions to Improve Educational Attainment at Community Colleges 543

Colleges, but neither result was statistically sig- enrolled at least half-time and earning a “C” aver-

nificant and there were no impacts on other indi- age, and the rest at the end of the semester contin-

cators of academic progress. The authors point gent on GPA.  At the end of the program year,

out that the WSG covered just 28% of the stu- treated students had earned 2.4 more credits, and

dents’ unmet financial need at two-year colleges, were 12 percentage points more likely to be

while it covered 39% at the four-year colleges retained into their second year. And one  year

and universities (and had sizable impacts on later, the credit advantage had grown to 3.5 cred-

degree completion—see Goldrick-Rab et  al. its4 (Barrow et al. 2014; Barrow and Rouse 2013;

2016). However, offering two-year students the Brock and Ritchburg-Hayes 2006; Ritchburg-­

grant did decrease their work hours, and particu- Hayes et al. 2009). But Hurricane Katrina brought

larly the odds of working the third-shift (Broton an end to the experiment. While the program’s

et al. 2016). Angrist et al. estimate a statistically evaluation points toward potentially promising

non-significant 1.9 percentage point lower one-­ effects, it also suggested that intervention’s costs

year retention rate for scholarship recipients who extended beyond the financial outlay for student

initially enrolled at community colleges. scholarships. Program implementation required

Importantly, Buffet scholarship recipients were 7 additional time on the part of counselors who

percentage points less likely to attend community monitored students’ enrollment and grades and

colleges in the first place than control students, were available to offer advice and referrals to

suggesting that the additional aid increased four-­ additional services. The program also required

year attendance among those who would other- additional personnel time to administer the aid

wise have opted for a community college to save program. That said, the evaluation falls short of

money.1 identifying the extent of additional time spent by

In 2004–2005, as part of its larger “Opening counselors and administrators, and did not

Doors” demonstration,2 MDRC evaluated a e­ stimate the costs associated with implementing

“performance-­based scholarship”3 (PBS) for the reform.

low-income, mostly female parents at two com- Encouraged by these results, in 2008 MDRC

munity colleges in the New Orleans area launched a larger PBS demonstration at commu-

(N = 1019). The scholarship provided $1000 per nity colleges in Ohio, New  York City, Arizona,

semester for up to two semesters, awarded incre- and Florida5 (N  =  2285; N  =  1502; N  =  1028;

mentally: $250 upon enrollment (at least 6 cred- N  =  1075). The scholarships all targeted low-­

its), $250 at midterm contingent on remaining income populations and made continued receipt

of aid contingent upon stipulated enrollment

1 Applicants to the Buffett Scholarship needed to specify a intensity and performance benchmarks (usually
“target” college in their initial application, but students part-time enrollment and earning at least a “C”).

were not bound to attend these colleges. The scholarships varied in terms of generosity

2 Opening Doors was a multi-site experimental demonstra- and additional behavioral requirements for parts
tion examining the impact of different sorts of interven- of the aid. In these RCTs, the experimental group
tions designed to improve college retention and experienced short-term gains of smaller size than
completion among lower-income students. These various in the Louisiana experiment. In only two were
interventions included learning communities, college

skills courses, intensive counseling, and performance-

based scholarships.

3 This name is something of a misnomer. Most scholar- 4 We are summarizing results for the first two study cohorts
ships and grants, need-based or otherwise, have perfor- (out of four) only, because four semesters of data are
mance and enrollment requirements for continued receipt. available for these cohorts. Program-semester effects for
Indeed, the specific performance requirements of the cohorts 3 and 4 are similar, though of smaller magnitude.

PBSs were substantially more lenient than those of the 5 A performance-based scholarship RCT was carried out at

WSG or Buffet scholarship. What distinguishes the PBSs the University of New Mexico, and another targeted low-

is the incremental disbursal of grants and the tying of income high school seniors in California (Cash for

these disbursals to the performance of specific behaviors, College), but these results fall outside the purvey of this

such as attending tutoring sessions. review.

544 D. Monaghan et al.

there impacts on retention: The Arizona scholar- on the notion that doing so will temper the “feast
ship improved one-year retention by between 2 or famine” dynamic occurring when aid is dis-
and 5 percentage points, and at one of the tributed in one lump sum. A pilot program was
New York sites the treatment group was retained conducted at three community colleges in 2010
at a 9 percentage point higher rate. The scholar- (Ware et al. 2013), and a large-scale randomized
ships consistently improved credit accumulation control trial is presently underway.
over the first year by between 0.9 and 1.7 credits,
and modestly improved academic performance, 24.5.2 F ree Computers
but these effects shrank to insignificance after the
end of the scholarship. Completion effects were Colleges—and even community colleges—tend
for the most part not yet available, but in Ohio the to assume that their students have access to the
treatment group was 3.3 percentage points more Internet. However, in 2010 only 66% of com-
likely to have earned an associate degree or cer- munity college students with household
tificate at the end of 2 years. incomes below $20,000 per year had home
computers with Internet access (Fairlie and
Collectively the five PBS experiments suggest Grunberg 2014). In 2006, a randomized control
that additional need-based aid can modestly trial at a community college in northern
boost retention and credit accumulation, but California tested the impact of providing stu-
seems to be more effective when paired with sup- dents with free computers. Researchers
port services such as tutoring and advisement. In recruited 286 students for the experiment, and
all experiments that incorporated such services half were given refurbished computers. Treated
(Louisiana, Arizona, and Florida) recipients sub- students were slightly more likely to take
stantially outpaced the control group in meeting courses which would transfer to a state four-
program-specified goals. However, in nearly all year college: Transfer-eligible courses made up
cases effects were observable only as long as 66% of all courses taken by treated students
scholarships were still operative (for results of and 61% of courses taken by untreated s­ tudents.
the Arizona RCT, see Patel and Valenzuela 2013; And in the first 2  years, treated students were
for Florida, Sommo et  al. 2014; for New  York, slightly more likely to take courses for a letter
Ritchburg-Hayes et al. 2011 and Patel and Rudd grade. But no impacts were found on passing
2012; for Ohio, Cha and Patel 2010 and Mayer courses, earning degrees or certificates, or
et al. 2015; for a summary of the demonstration, transferring to a four-year college (Fairlie and
see Patel et al. 2013). What is unclear, however, is Grunberg 2014; Fairlie and London 2012).
at what cost these gains were achieved. All of the
programs involved both financial investments in 24.5.3 F inancial Aid Information
scholarship payments to students as well as per-
sonnel time, particularly at community colleges, The financial aid system is complex and requires
to implement. This makes the cost effectiveness students to make weighty decisions, and many
of scholarship programs unclear, as well as what community college students negotiate it alone.
might be required of community colleges to Not surprisingly, this can lead to costly errors.
implement such programs. For instance, students who receive Pell grants
may not know that they need to reapply for them
There are at least two ongoing experiments annually. Nationally, 10% of Pell-eligible stu-
involving either financial aid itself or its method dents fail to re-apply for financial aid in their sec-
of disbursement. In 2014, the Wisconsin HOPE ond year of college, and the resulting loss of aid
Lab launched an RCT investigating the impact of is strongly predictive of dropping out (Bird and
need-based scholarships on low-income students Castleman 2014).
who indicate interest in STEM fields. And MDRC
is testing a program entitled “Aid Like a
Paycheck” that disburses financial aid in small
amounts regularly throughout the semester, based

24  Experimental Evidence on Interventions to Improve Educational Attainment at Community Colleges 545

There are two experiments that identify the on the notion that many students do not have the

impacts of providing students with financial aid requisite cultural capital to successfully negotiate

information. Castleman and Page (2015) con- higher education. They aim to impact skills in

ducted a randomized control trial among low-­ study habits, time management, organization,

income first-year college students in the Boston self-presentation, goal-setting, and negotiating

area in which the treatment group was sent text-­ the educational bureaucracy. In this manner, they

message reminders to re-file the FAFSA. Among are analogous to remedial courses, but are lower-­

community college students, receipt of text stakes as they are usually pass/fail and grant only

reminders improved retention into the fall and a credit at most.

spring semesters of sophomore year by 12 and 14 College skills courses offered to or required of

percentage points, respectively. Impacts were first-semester freshmen are common, but there is

larger among students with lower high school little rigorous research on their effectiveness and

GPAs. Barr et  al. (2016) carried out an experi- most studies are descriptive in nature (Derby and

ment with new student loan applicants at the Smith 2004; O’Gara et al. 2009; Zeidenberg et al.

Community College of Baltimore in which 2007). MDRC evaluated the impacts of two col-

treated students were sent, over the course of a lege skills course programs for students on aca-

month, a series of texts with student loan facts. demic probation at a community college in the

The texts told students that they could borrow Los Angeles area. In both programs, the treatment

less (and sometimes more) than the amount consisted of a two-semester college skills course

offered by their institution, that monthly repay- taught by a college counselor, a “Success Center”

ments depend on the amount borrowed and the that provided tutoring services, and a modest

repayment plan, and that there are lifetime limits voucher to cover the cost of textbooks. In the first

on borrowing. Students receiving the texts bor- program, the skills course was presented to those

rowed 9% less in Stafford loans and 12% less in randomized into treatment as optional; they were

unsubsidized Stafford loans, and larger declines merely encouraged to enroll, and participation in

in borrowing were witnessed among new enroll- tutoring was not enforced. As a result, only half of

ees, Blacks, low-income students, and students the treatment group took the first-semester course,

with lower GPAs. very few took the second-semester course, tutor-

Turner (2015) is presently conducting an ing services were rarely utilized, and treatment

experiment with community college students in effects were nonexistent. In the second iteration,

three states which randomizes the default option students were told (falsely) that they were required

presented to loan applicants. For some, the to take the first-semester skills course and were

default option will be to take out a loan, and stu- strongly encouraged to take the second-semester

dents will have to take action to opt out, while for course, and attendance at tutoring sessions was

the others the opposite will be true. Additionally, enforced by instructors. Take-up was much better

the experiment will randomly assign some to be in this iteration, and the treatment group earned

presented with a particular loan amount as a 2.7 additional credits on average over the two

default while others will have to choose a loan semesters of the program and was 7 percentage

amount, and some applicants will be prompted to points more likely to pass all of their classes. At

complete a worksheet helping them take stock of the end of the program year, the experimental

their resources and expenses before making a group was 10 percentage points less likely to be

decision while others will not. on academic probation, though this impact did not

persist after one additional semester (Scrivener

24.5.4 C ollege Skills Classes et  al. 2009; Weiss et  al. 2011). While the pro-

grams’ evaluations suggest that skills course pro-

grams might be a promising strategy, the

College skills classes are one of many interven- interventions involve additional resources on the

tions that community colleges provide premised part of community colleges (e.g., services and

546 D. Monaghan et al.

vouchers). However, existing evaluations do not with the Wisconsin HOPE Lab to assess the

describe the resources required for implementa- impacts of similar interventions at six two-year

tion, nor the programs’ costs. colleges in Wisconsin.

24.5.5 Social-Psychological 24.5.6 Incentivizing Academic
Interventions Momentum

Social psychologists have recently explored the The academic momentum perspective suggests
impacts of teaching individuals that intelligence that the speed at which a student makes progress
is not fixed but rather can be augmented through towards a degree—through accumulating credits
training and effort. Interventions designed to or clearing remedial requirements—has an inde-
instill a “growth mindset” informed by this incre- pendent causal impact on their likelihood of com-
mental theory of intelligence have been found to pletion (Adelman 1999; Attewell et  al. 2012;
effectively boost the academic performance of Attewell and Monaghan 2016). This may be
four-year college students and other groups because rapid completion minimizes cumulative
(Blackwell et  al. 2007). Building off this work, exposure to the risk of an event that could derail
Paunesku, Yeager, and colleagues developed a schooling, or because students who spend more
30-min intervention (a webinar and reinforce- time involved in schoolwork will be more aca-
ment activity) that teaches viewers that intellec- demically integrated into the institution. Attewell
tual skills are learned rather than fixed, and tested conducted a pair of randomized control trials at
it in a community college context. In one field community colleges in the City University of
experiment involving mostly Latino students at a New York to test two applications of this theory.
Los Angeles-area community college, treated In the first, students who were attending college
students earned overall GPAs which were 0.18 part-time (fewer than 12 credits) in the fall
grade points higher in the following semester. In semester were incentivized to “bump up” to 12 or
a second experiment the intervention was tested more credits in the spring. In the second, students
among students in remedial math courses. In this who had elected not to sign up for summer
case, the treated group dropped out of their math courses after their first year in college were
class at less than half the rate of the control group incentivized to do so. In both cases, the incentive
(9% vs 20%) (Yeager and Dweck 2012; Yeager was a generous $1000. In the experiment involv-
et al. 2013). ing increased credit load, the treatment group
was more likely to be retained into the second
Other researchers have investigated ways to year, and at the end of the second year had accu-
impact students’ motivation and therefore perfor- mulated six additional credits on average. In the
mance in academic contexts. Harackiewicz and summer coursework experiment, treated students
colleagues have investigated the impacts of both were 8 percentage points more likely to still be
“utility” interventions and “values” interven- enrolled two semesters after treatment, and had
tions. In the former, students are provided infor- accumulated an additional three credits by the
mation about the labor market value of science end of their second year of college (Attewell and
and math skills; in the latter, students complete a Douglas 2016).
brief in-class writing assignment in which they
select and explore values (such as spiritual or 24.6 S chool-Level Interventions
religious values, career, or belonging to a group)
that are important to them. Such interventions In contrast to the interventions outlined above,
have been found to improve outcomes among school-level interventions augment or alter the
both high school and university students institutional environment that individuals must
(Harackiewicz et  al. 2014; Harackiewicz et  al.
2015). The researchers are currently working

24  Experimental Evidence on Interventions to Improve Educational Attainment at Community Colleges 547

navigate in order to attain their goals. They may experience as “good” or “excellent,” report that
by extension augment students’ stock of knowl- they had a campus staff member on whom they
edge or capacities, and they do not necessarily relied for support, and receive financial aid in the
presume that individual deficits do not contribute spring semester. After the program year these
to generating unacceptable outcomes. But they gains did not persist, but many students contin-
do presuppose that the institutional environment ued to seek out the counselor formerly assigned
is changeable, and that the status quo may con- to them (Scrivener and Au 2007; Scrivener and Pi
tain unnecessary barriers to goal-attainment. We 2007; Scrivener and Weiss 2009).
identified 15 such interventions.
While the programs’ impacts were substantial,
24.6.1 “Enhanced” Student Services replicating this program elsewhere is hampered
somewhat by the absence of information on the
As noted earlier, counseling centers have been resources community colleges dedicated to its
singled out for critique by scholars of late. implementation. It is important to note that these
Because of the complexity of community col- programs required community colleges to poten-
leges as institutions and students’ lack of assis- tially dedicate additional personnel hours to carry
tance from knowledgeable family members, out the intervention, particularly counselors with
effective counseling emerges as utterly crucial to whom students met more frequently. However,
providing the information and guidance neces- the study does not describe in detail how com-
sary for student success (Allen et  al. 2013). munity colleges allocated the personnel hours
Effective counseling could also help students feel required—either by reallocating or expanding
more connected to the institution by establishing existing counselor time or by adding additional
a relationship with at least one trustworthy staff personnel.
member. But this is simply not present at most
community colleges, where counseling services 24.6.2 M entoring
are student-initiated and at which counselors are
responsible for a large number of students and MRDC carried out an evaluation of a “light-­
provided little training or time to serve them. touch” mentoring program for students taking
developmental and early college-level math
One RCT conducted by MDRC at two Ohio courses at a community college in McAllen,
community colleges investigated the effect of Texas. In the program, students’ math sections
“enhanced” counseling services. In this evalua- were randomly assigned to treatment and control
tion, treatment group students were assigned to a categories. Treated sections were assigned a non-­
specific counselor, with whom they were faculty college employee who acted as a mentor
expected to meet regularly, and this counselor for the students in the course and informed them
was assigned a reduced caseload (160:1 rather about additional support services, such as the
than the usual 1000:1). Treated students were tutoring center. The program succeeded in
also assigned a designated contact person in the increasing students’ utilization of on-campus ser-
financial aid office and were given a $150 stipend vices such as the tutoring center, and treated stu-
per semester conditional on meeting with coun- dents were more likely to report feeling that they
selors. During the two semesters the program had someone on campus to whom they could turn
was active, impacts were substantial. Treated stu- to for help. However, there were no statistically
dents’ fall-to-spring retention was 7 percentage significant differences in pass rates, GPA, or final
points higher than the control group, and treated exam score. Among part-time students, however,
students accumulated a half credit extra over the the treatment group was more likely to pass their
course of the year. In surveys, the program group math course and earned slightly higher scores on
also was more likely to describe their college the final (Visher et al. 2010).

548 D. Monaghan et al.

24.6.3 Testing and Remediation either directly into a college-level class or into a

remedial course (Sharon 1972). The students

As previously discussed, remediation is the near-­ assigned to remedial courses were retained at

universal institutional compromise strategy com- rates similar to control group students, and passed

munity colleges have adopted to resolve the college-level English at similar rates. However,

dilemma of being open-door institutions of they tended to earn higher grades in this course,

advanced education. Analogous to the situation suggesting some positive impact of remediation

with financial aid, the effect of taking a remedial on academic skills. Forty years later, Moss and

course must be separated empirically from the Yeaton (2013) conducted an experiment in which

effects of the academic weaknesses that landed students immediately below the remedial cutoff

students in the remedial course (Levin and on a math placement test were randomly placed

Calcagno 2008). But the matter is even more com- into either remedial courses or college-level

plicated because though taking a remedial course courses. The authors do not present results for

could improve one’s skills and odds of comple- retention or accumulation of college-level cred-

tion, being assigned to remediation has consider- its, but find a positive impact of taking remedial

able (likely negative) consequences in its own courses on grades in college-level math. Their

right. The net impact of a school’s testing and RCT sample was very small (N  =  63), but the

remediation policy is the balance of these two authors also conducted supplemental analyses

opposing effects—something that is typically using regression-discontinuity designs and found

overlooked in the research literature. In part similar effects. Finally, Logue and others at the

because of this methodological confusion, schol- City University of New York randomly assigned

ars have failed to reach consensus on remedia- students identified as requiring remediation into

tion’s impacts (Bailey 2009; Melguizo et al. 2011). either remedial algebra, remedial algebra with

Observational studies that compare those who take additional tutoring, or college-level statistics

remedial courses and those who do not tend to find with tutoring (Logue et  al. 2016). Early results

only small differences in completion, and their show no difference between the two groups tak-

authors have interpreted this as demonstrating that ing developmental algebra, but the group assigned

remedial courses are effective (Adelman 1998, to take statistics passed their assigned course at

1999; Attewell et  al. 2006; Bahr 2008; Fike and far higher rates, accumulated more credits in both

Fike 2008). But studies employing more sophisti- the program and post-program semesters, and

cated quasi-experimental designs have found were retained at similar rates. The researchers

impacts to be neutral-to-n­egative (Boatman and attribute the gain in credits among the “main-

Long 2010; Calcagno and Long 2008; Martorell streamed” group to three factors: They passed

and McFarlin 2011; Scott-C­ layton and Rodriguez their assigned course at higher rates, this course

2012; for a counter-­example, see Bettinger and counted for college credit, and it served as a pre-

Long 2009). A recent meta-analysis of this work requisite for other courses, enabling students to

finds that being placed into remediation has a pursue their majors more freely. The costs—to

small, but statistically significant, negative impact students and institutions—of these remedial

on credit accumulation, ever passing the course for course interventions are not well understood. The

which remediation was needed, and degree attain- evaluations did not incorporate direct measures

ment (Valentine et al. 2016). of costs in their analysis.

There are four randomized control trials that Another experiment investigated the impacts

deal with remediation at community colleges in of alternative methods of remedial placement

one form or another. Three RCTs investigated the (Evans and Henry 2015). This project contains

effects of taking remedial courses versus entering two separate experimental groups, both of which

directly into college-level work. An early RCT take an alternative test called the ALEKS, which

conducted in the late 1960s randomly placed stu- provides self-paced personalized learning mod-

dents identified as needing remediation in English ules for those who fail the test and allows them to

24  Experimental Evidence on Interventions to Improve Educational Attainment at Community Colleges 549

retake it. One of the treatment groups, charged to take these courses. Observational
“ALEKS-2­ ,” could only retake the test once, and research indicates that these programs effectively
only after completing all assigned modules. The boost persistence and even six-year attainment
other, “ALEKS-5,” could retake the test up to (Douglas and Attewell 2014).
four times but were not required to complete
learning modules. Control students took the stan- MDRC, in conjunction with the National
dard placement test (the COMPASS, in this case). Center for Postsecondary Research (NCPR),
Only first-semester results are at this point avail- carried out an experimental evaluation of sum-
able, but both treatment groups were less likely to mer bridge programs at eight colleges in Texas,
be placed into remediation. In addition, the including six community colleges (Barnett
ALEKS-2 group was more likely than the control et  al. 2012; Wathington et  al. 2011). Early
group to take college-level math in their first impacts were encouraging: Treated students
semester, and the ALEKS-5 group was more were more likely to take and to pass college-
likely to pass it. level math and English courses in their first
year than control-g­roup students, suggesting
Two more remediation interventions will be that the bridge program successfully enabled
evaluated in the near future. In the first study, some students to quickly clear remedial require-
being conducted by MDRC, the Community ments. But there was no impact on one-year
College Research Center (CCRC) and CUNY, retention, and the advantages in college-level
students placed into remediation will be ran- course completion and credit accumulation nar-
domly assigned to complete a one-semester rowed to statistical insignificance by the fourth
intensive developmental immersion program semester. Researchers at CUNY carried out
(entitled CUNYStart) prior to official matricula- another experimental evaluation of summer
tion. The second, MDRC’s “Developmental bridge programs. In this i­ntervention, students
Education Acceleration Project,” evaluates two who missed the enrollment deadline for bridge
innovative formats for administering develop- courses were recruited into an experimental
mental education. The first treatment group will evaluation, and those selected for treatment
be assigned to developmental courses that are were offered $1000 to enroll in sections of
personalized, module-based, and which permit these courses reserved for the experiment.
them to enter and exit at their own pace. The sec- Researchers estimated a non-significant nega-
ond treatment group will take an accelerated pro- tive 5 percentage point effect of taking bridge
gram which squeezes two remedial courses into courses on one-year retention, and a non-­
one single semester. significant negative effect on credit accumula-
tion (Attewell and Douglas 2016).

24.6.4 S ummer Bridge Courses 24.6.5 Learning Communities

“Summer bridge” programs—courses or pro- Learning communities are geared towards pro-
grams that take place during the summer prior to viding community college students the opportu-
freshmen year—are widespread in higher educa- nity to build social connections to other students
tion. These programs vary substantially in their and to faculty that they typically do not form
content and are nearly always voluntary. At com- because of their loose connection to the college.
munity colleges, bridge courses are oriented They proceed on the notion that “social and aca-
nearly exclusively to teaching basic skills to demic integration” into the social world of the
incoming students who scored low enough on college is a key mechanism for retaining students.
placement exams to require remediation, offering Social bonds engender a feeling of belonging and
such students an opportunity to complete at least an obligation to make good on implicit promises
some required remedial coursework prior to the to return and complete degrees. They addition-
first semester. Oftentimes students are not

550 D. Monaghan et al.

ally provide students with information networks Sommo 2005; Scrivener et al. 2008; Weiss et al.

and sources of emotional support. 2014, 2015).

Learning communities seek to cultivate stu- Subsequently, in conjunction with the NCPR,

dent success through three interconnected mech- MDRC carried out experimental evaluations of

anisms (Tinto 1997). First, a group of students learning communities at six separate community

take multiple courses together, providing oppor- colleges beginning in fall 2007. These evaluations

tunities for students to form social bonds and to involved, collectively, more than 6500 students,

support each other across courses (Karp 2011). and the programs evaluated varied from the earlier

Second, the courses are linked in terms of con- study in two important respects. First, they for the

tent, allowing for deeper engagement with mate- most part lacked any supplementary services, thus

rial. Third, faculty who teach the linked courses presenting purer tests of learning community

collaborate and share information about student impacts. Second, whereas the earlier project evalu-

progress and engagement. Additionally, many ated an established learning community at scale,

learning communities feature reduced class sizes, the later evaluations involved either newly-created

block-scheduling, and auxiliary services such as learning community programs or existing pro-

advising and tutoring. Frequently, one of the grams which were rapidly scaled up, incorporating

linked courses is a first-year college skills semi- faculty with little experience with learning com-

nar. Observational research on learning commu- munities and no history of collaborating on linked

nities almost uniformly finds positive impacts on courses. As was the case previously, all were one-

outcomes such as student engagement, interac- semester interventions. The resources required in

tion with faculty, relationships with peers, per- order to implement learning communities in com-

ceptions of institutions, academic performance, munity colleges, and their corresponding costs,

and retention (Minkler 2002; Raftery 2005; Tinto associated with i­mplementing these learning com-

et al. 1994). munities are essentially unknown.

There have been seven experimental evalua- “No frills” learning communities at Merced

tions of learning communities, all by MDRC. In College in California, Hillsborough Community

2003 MDRC evaluated an existing learning com- College in Florida, and the Community College

munity for entering students at Kingsborough of Baltimore had negligible results. At Merced,

Community College in New York City. The treat- the treated group was about a third of a course

ment group was split into learning communities ahead of the control group in the completion of

of roughly 25 students who took three courses remedial sequences, and at Hillsborough the

together in their first semester: introductory treatment group was 5 percentage points more

English (mostly remedial), a course in their likely to be retained into the second semester. No

major, and a college skills course. There were further impacts were detected on academic per-

substantial support services: Treated students formance or credit accumulation, and no effects

were assigned an academic advisor (who was lasted beyond the first post-program semester

granted a smaller caseload), had reduced class (Weiss et  al. 2010; Weissman et  al. 2012).

sizes, were provided enhanced and often in-class Learning communities at Houston Community

tutoring, and were granted a $150 book voucher College and Queensborough Community College

for the semester. These supports, and the learning in New  York were slightly more elaborate. The

community itself, only lasted a single semester. Houston program linked remedial math to a stu-

The program had encouraging early impacts on dent success course, and tutoring and counseling

retention and completion of remedial courses, as was inconsistently provided. Treated students

well as on non-cognitive outcomes such as self-­ completed their first remedial math course at a

reported academic engagement and reported feel- rate 14 percentage points higher during the pro-

ings of belonging at the school. Positive impacts gram semester, and this advantage persisted for

faded out after four semesters (Bloom and two semesters after the program (Weissman et al.

24  Experimental Evidence on Interventions to Improve Educational Attainment at Community Colleges 551

2011). The Queensborough learning community ments. There are mandatory support services:
was supported with a full-time coordinator and a Students are assigned an advisor (who has a
college advisor assigned solely to treatment reduced case-load) and required to meet with
group students. The treatment group was sub- them at least twice per month, and they are also
stantially more likely to pass the first develop- required to meet once per semester with a career
mental math course in their sequence during the services and employment counselor (dedicated to
program semester and the second math course in ASAP). Students are required to attend tutoring if
the first post-program semester, and there were they are in remedial courses, on academic proba-
modest effects on credit accumulation (Weissman tion, or are re-taking a course they have previ-
et  al. 2011). Finally, researchers returned to ously failed. In each semester ASAP students are
Kingsborough Community College to evaluate required to take a non-credit seminar focused on
learning communities aimed at students pursuing building and developing college skills. Learning
particular occupational majors. The program was communities are also involved in students’ first
beset by implementation and recruitment prob- year, though precisely how these are conducted
lems, and the school was forced to alter the pro- varies across CUNY campuses. Students are
gram repeatedly throughout the evaluation. Not strongly encouraged to take required remedial
surprisingly, no effects were found on outcomes courses as early as possible, to attend tutoring for
of interest (Visher and Torres 2011). courses in which they are struggling, and to make
use of winter and summer intercessions to accu-
24.7 A Comprehensive Support mulate credits more rapidly. ASAP courses also
Intervention: CUNY ASAP tend to be somewhat smaller than average courses
at CUNY community colleges.
In 2007, with support from the City’s Center for
Economic Opportunity, the City University of But perhaps most important, in contrast to
New York launched what is likely the single most most interventions reviewed thus far, ASAP is
ambitious program to boost degree completion in not limited to a single semester or year. Instead,
a community college setting. The Accelerated conditional on meeting certain requirements—
Study in Associate Programs (ASAP) initiative such as remaining enrolled full-time—students
does not rely on a single intervention such as can participate in and access the benefits of ASAP
financial aid or smaller class sizes. Instead, it for three full years. Most interventions reviewed
builds on prior research, such as the 2003 learn- above were at least modestly successful during
ing community evaluation at Kingsborough, program semesters, but effects faded out rapidly
which suggested that multifaceted programs that thereafter. One reaction to this is to conclude that
address multiple student needs simultaneously the interventions “don’t work” because they did
tend to have more robust impacts. not produce “lasting gains.” ASAP planners drew
the opposite conclusion: In order to be success-
ASAP draws on many of the strategies ful, an intervention strategy needs to be not only
involved in the interventions we have already comprehensive but sustained.
described and adds a few more. First, there is
financial support: Tuition and fees not met In its first few years, ASAP was open only to
through other grants are waived, and students are “college-ready” students—that is, students with
provided with subway passes and can rent text- no remedial requirements. Internal evaluations,
books free of charge. Building on the academic utilizing propensity-score matching methods,
momentum perspective, participating students suggested that participation in ASAP was associ-
are required to enroll full-time (at least 12 cred- ated with a 28.4 percentage point gain in three-­
its), though they have the alternative of enrolling year degree completion and a half-semester’s
at slightly less than full-time and using winter difference in credits accumulated after 3  years
and summer intercessions to meet credit require- (Linderman and Kolenovic 2012). Encouraged
by these findings, CUNY contracted with MDRC
to carry out a randomized assignment evaluation.

552 D. Monaghan et al.

This evaluation began in the spring semester of College in Fort Worth, Texas, researchers are
2010, and involved just under 900 students at carrying out an experimental evaluation of a pro-
three CUNY community colleges. Instead of lim- gram called Stay the Course. Operated in part-
iting eligibility to college-ready students, partici- nership with a local non-profit, Stay the Course
pation was limited to low-income entering is designed to address non-academic obstacles
students who demonstrated some, though not faced by low-income community college stu-
deep, remedial need (1 or 2 required courses). dents through provision of comprehensive case
management and emergency financial assistance
The evaluation found that ASAP generated (Evans et al. 2014).
large early impacts. By the end of the first year,
the treatment group was 25 percentage points 24.8 Discussion and Conclusion
more likely to have completed all required reme-
dial courses, and had earned 3 more college-level As the vast majority of new jobs require postsec-
credits on average (Scrivener et al. 2012). These ondary training (Carnevale et al. 2013), low rates
impacts grew, rather than attenuating, over time. of degree completion increasingly disadvantage
After 3 years, treated students had accumulated lower-SES and minority youth. Community col-
7.7 more credits on average than the control leges are positioned to play a central role in
group. And whereas only 21.8% of the control expanding educational attainment and narrowing
group had completed a degree, 40.1% of the educational disparities. But in order to do so, they
treatment group had done so—an 83% gain. must pivot institutionally from guaranteeing
Treated students were also 9.4 percentage points access to facilitating degree completion—with-
more likely to have transferred to a four-year col- out compromising on the former. But making
lege within 3  years (Scrivener et  al. 2015). The community colleges deliver on promises of edu-
ASAP evaluation stands out as one of the few that cational opportunity will require not timid
systematically evaluated program costs, provid- reforms or tinkering, but bold innovation and
ing some guidance to community colleges seek- substantial resources.
ing to replicate the program. That said, the
accompanying cost study shows that ASAP’s Community colleges arose in the era of post-
gains did not come cheaply. The direct costs were war educational optimism with an explicit set of
estimated to be over $14,000 per student over goals—expanding access to college, providing a
3  years. ASAP students also took more classes broad and comprehensive set of programs, and
than control students, and incorporating these serving local communities—which they have
costs could raise the per-student total to between emphatically achieved. Today, politicians, schol-
$16,000 and $18,500. However, given the large ars, and foundations are demanding that commu-
increase in completion, researchers estimated nity colleges do better in terms of degree
that per degree, ASAP spent $13,000 less than completion. The simplest method for community
was spent on the control group (Scrivener et al. colleges to increase degree completion is to
2015). Despite information on program costs and restrict access to those who are “college-ready.”
effects, given the absence of other similar studies Or community colleges could reduce institutional
it is impossible to evaluate this evidence relative complexity by eliminating scores of occupational
to other interventions, leaving a lingering ques- programs that serve millions and are valued by
tion—is ASAP a cost-effective alternative rela- employers. However, few policymakers wish to
tive to other possible interventions? see community colleges abandoning either their
democratic mission or the provision of voca-
Efforts to evaluate comprehensive interven- tional certifications at low cost. Instead, in an era
tion models like ASAP are continuing. MDRC is of withering public support, community colleges
currently conducting a replication of ASAP at are being ordered to do more with less (Jenkins
three community colleges in Ohio; the evalua- and Belfield 2014).
tion cohort enrolled in fall 2015 and will be
tracked for 3  years. And at Tarrant County

24  Experimental Evidence on Interventions to Improve Educational Attainment at Community Colleges 553

The new focus on completion has brought scarcity or academic weaknesses or slight college

unprecedented scholarly attention—supported by knowledge do not vanish when a program closes

unprecedented research funding—to community up shop, but reassert themselves vigorously.

colleges, leading to a number of promising Policymakers should not expect short-term pro-

experimentally-e­valuated interventions. Need-­ grams to have anything other than short-term

based financial aid, particularly when accompa- impacts.

nied with supports, has increased retention and As community colleges operate with limited

credit accumulation. Learning communities do and unpredictable resources, policymakers and

not seem to generate large gains on their own, but educational leaders considering reforms must

have short-run impacts on retention and move- attend to the resources required for implementa-

ment through remedial sequences when coupled tion (Belfield et  al. 2014; Belfield and Jenkins

with counseling and other supports. “Enhanced” 2014). However, as we noted, existing evalua-

counseling appears to benefit students as long as tions largely ignore such matters (Belfield 2015).

it remains available. And there is evidence that Community college leaders need to know where

limiting exposure to remediation and can speed to invest scarce dollars and how programmatic

progress toward degrees. decisions influence resource requirements. State-­

Other interventions should be evaluated level policymakers are also are at a disadvantage.

experimentally. For example, scholars have pro- There are few benchmarks for determining at

posed developing “guided pathways,” clear what level community colleges should be funded

sequences of courses leading directly to creden- (Chancellor’s Office of the California Community

tials and/or transfer to a four-year college. Others Colleges 2003), and none are explicitly tied to

suggest providing housing or food support— performance (Kahlenberg 2015). Future evalua-

campus food pantries or a collegiate equivalent of tions should examine interventions’ relative cost

free and reduced lunch—will provide low- effectiveness and clearly delineate resources

income students greater security and improve entailed for implementation, or community col-

educational outcomes (Broton et  al. 2014; leges will risk squandering scarce resources or

Goldrick-Rab et  al. 2015). Another promising selecting interventions for which they have insuf-

intervention is emergency financial assistance for ficient capacity to implement. Research con-

students facing unexpected crises that endanger ducted by MDRC, the Wisconsin HOPE Lab, and

their persistence (Dachelet and Goldrick-Rab the Center for Benefit-Cost Studies in Education

2015; Geckler et  al. 2008). Single-stop centers, has begun to incorporate estimates of cost, but

which provide information about and access to a more is needed.

range of benefits and services in a single location, The evaluation literature also devotes inade-

are being established on campuses across the quate attention to the context in which interven-

country, and could be evaluated using random- tions occur. As we discussed, structural features

ized encouragement (Goldrick-Rab et al. 2014). of community colleges work at cross-purposes

The available evidence strongly suggests two with efforts to raise completion rates, and recent

tentative conclusions. First, simple interventions fiscal developments have further eroded capacity

do not appear to work as well as multifaceted for improvement. Additionally, community col-

programs. Complex interventions like ASAP can lege students confront a broader opportunity

lead researchers to wonder which interventions structure which presents immense obstacles to

are most impactful. But this assumes components improving their situation through educational

to have independent, additive effects, when they upgrading. If evaluators do not take these struc-

may interact with and reinforce each other. tural realities into consideration, unrealistic

Second, that many programs impacts are positive expectations will be set and improper conclu-

while in operation but fade away afterwards sug- sions reached. Too often, when an intervention

gests that effective interventions must be pro- has small, short-lived impacts, this is taken as

longed. Underlying problems such as resource evidence that the strategy in the abstract “doesn’t

554 D. Monaghan et al.

work.” A more realistic conclusion is likely that Attewell, P., Lavin, D., Domina, T., & Levey, T. (2006).
the intervention is, by itself, inadequate to over- New evidence on college remediation. Journal of
come the collective weight of countervailing Higher Education, 77(5), 886–924.
structural forces bearing upon individuals and
institutions at the bottom of the educational and Attewell, P., Heil, S., & Reisel, L. (2012). What is aca-
social hierarchy. demic momentum? And does it matter? Educational
Evaluation and Policy Analysis, 34(1), 27–44.
Failure to take resources and power into
account enables the tacit assumption that the only Bahr, P. R. (2008). Does mathematics remediation work?
actors that matter in determining community col- A comparative analysis of academic attainment among
lege students’ success are the colleges and the community college students. Research in Higher
students themselves. This conceals the real and Education, 49(5), 420–450.
pressing need for broader structural reforms to
ensure that community colleges are able to pro- Bahr, P. R. (2009). College hopping: Exploring the occur-
vide real educational opportunity to all rence, frequency, and consequences of lateral transfer.
Americans, regardless of background. Community College Review, 36(4), 271–298.

References Bailey, T. (2009). Challenge and opportunity: Rethinking
the role and function of developmental education in
Adelman, C. (1998). The kiss of death? An alternative community college. New Directions for Community
view of college remediation. National Crosstalk, 6(3), Colleges, 2009(145), 11–30.
11.
Bailey, T. R., Kienzl, G., & Marcotte, D. E. (2004). The
Adelman, C. (1999). Answers in the toolbox: Academic return to a sub-baccalaureate education: The effects
intensity, attendance patterns, and bachelor’s degree
attainment. Washington, DC: U.S.  Department of of schooling, credentials and program of study on
Education. economic outcomes (Report prepared for the National
Assessment of Vocational Education, U.S. Department
Allen, J.  M., Smith, C.  L., & Muehleck, J.  K. (2013). of Education). New  York: Community College
What kinds of advising are important to community Research Center.
college pre-and post-transfer students? Community Bailey, T., Jeong, D. W., & Cho, S. W. (2010). Referral,
College Review, 41(4), 330–345. enrollment, and completion in developmental educa-
tion sequences in community colleges. Economics of
Alon, S. (2011). Who benefits most from financial aid? Education Review, 29(2), 255–270.
The heterogeneous effect of need-based grants on stu- Bailey, T.  R., Jaggars, S.  S., & Jenkins, D. (2015).
dents’ college persistence. Social Science Quarterly, Redesigning America’s community colleges: A clearer
92(3), 807–829. path to student success. Cambridge, MA: Harvard
University Press.
Anderson, D. M., & Goldrick-Rab, S. (2016). Impact of a Barnett, E.  A., Bork, R.  H., Mayer, A.  K., Pretlow, J.,
private need-based grant on beginning two-year col- Wathington, H.  D., & Weiss, M.  J. (2012). Bridging
lege students (Working Paper). Madison: Wisconsin the gap: An impact study of eight developmental sum-
HOPE Lab. mer bridge programs in Texas. New York: MDRC.
Barr, A.  Bird, K., & Castleman, B.  J. (2016).
Angrist, J. D., Autor, D., Hudson, S., & Pallais, A. (2015, Prompting active choice among high-risk bor-
January). Leveling up: Early results from a random-
ized evaluation of a post-secondary aid. Presented rowers: Evidence from a student loan counseling
at the annual conference of the American Economic experiment (EdPolicyWorks Working Paper No. 41).
Association. Boston. Charlottesville: The Center on Education Policy and
Workforce Competitiveness.
Astin, A. W. (1984). Student involvement: A developmen- Barrow, L., & Rouse, C.  E. (2013). Financial incen-
tal theory for higher education. Journal of College tives and educational investment: The impact of
Student Personnel, 25(4), 297–308.
performance-b­ ased scholarships on student time use
Attewell, P., & Douglas, D. (2016). When randomized con- (NBER Working Paper w19351). Cambridge, MA:
trol trials and observational studies diverge: Evidence National Bureau of Economic Research.
from studies of academic momentum (Working Paper). Barrow, L., Richburg-Hayes, L., Rouse, C. E., & Brock,
CUNY Academic Momentum Project. T. (2014). Paying for performance: The education
impacts of a community college scholarship program
Attewell, P., & Monaghan, D. B. (2016). How many cred- for low-income adults. Journal of Labor Economics,
its should an undergraduate take? Research in Higher 32(3), 563–599.
Education, 57(6), 682–713. Baum, S., & Kurose, C. (2013). Community colleges in
context: Exploring financing of two- and four-year
institutions. In Bridging the higher educational divide:
Strengthening community colleges and restoring the

American dream: The report of the Century Task Force

on preventing community colleges from becoming

24  Experimental Evidence on Interventions to Improve Educational Attainment at Community Colleges 555

separate and unequal (pp.  73–108). New  York: The ence of heterogeneity and complex counterfactuals.
Century Foundation Press. Sociological Science, 1(1), 448–465.
Belfield, C. (2015). Efficiency gains in community col- Brint, S., & Karabel, J.  (1989). The diverted dream:
leges: Two areas for further investigation (CCRC Community colleges and the promise of educational
Working Paper 80). New  York: Community College opportunity in America, 1900–1985. New  York:
Research Center. Oxford University Press.
Belfield, C., & Jenkins, D. (2014). Community college Brock, T., & Ritchburg-Hayes, L. (2006). Paying for
persistence: Early results of a Louisiana scholarship
economics for policymakers: The one big fact and the
one big myth (CCRC Working Paper 67). New York: program for low-income parents attending community
Community College Research Center. college. New York: MDRC.
Belfield, C., Crosta, P.  M., & Jenkins, D. (2014). Can Broton, K., Frank, V., & Goldrick-Rab, S. (2014). Safety,
community colleges afford to improve completion? security, and college attainment: An investigation
Measuring the costs and efficiency effects of college
reforms. Educational Evaluation and Policy Analysis, of undergraduates’ basic needs and institutional
36(3), 327–345. response (Wisconsin HOPE Lab Working Paper).
Bettinger, E. (2004). How financial aid affects persistence. Madison: Wisconsin HOPE Lab.
In C. M. Hoxby (Ed.), College choices: The econom- Broton, K., Goldrick-Rab, S., & Benson. (2016). Working
for college: The causal impacts of financial grants on
ics of where to go, when to go, and how to pay for it undergraduate employment. Educational Evaluation
(pp. 207–238). Chicago: University of Chicago Press. and Policy Analysis, 38(3), 477–494.
Bettinger, E. P. (2012). Financial aid: A blunt instrument Calcagno, J.  C., & Long, B.  T. (2008). The impact of
for increasing degree attainment. In A.P.  Kelly & postsecondary remediation using a regression dis-
M. Schneider. Getting to graduation: The completion
agenda in higher education (pp. 157–174). Baltimore: continuity approach: Addressing endogenous sorting
The Johns Hopkins University Press. and noncompliance (NBER Working Paper w14194).
Bettinger, E. (2015). Need-based aid and college persis- Cambridge, MA: National Bureau of Economic
tence: The effects of the Ohio College Opportunity Research.
Grant. Educational Evaluation and Policy Analysis, Carnevale, A. P., Smith, N., & Strohl, J. (2013). Recovery:
37(1 supplementary), 102S–119S. Job growth and education requirements through
Bettinger, E.  P., & Long, B.  T. (2009). Addressing the 2020. Washington, DC: Center for Education and the
needs of underprepared students in higher education: Workforce, Georgetown University.
Does college remediation work? Journal of Human Castleman, B. L., & Long, B. T. (2013). Looking beyond
Resources, 44(3), 736–771. enrollment: The causal effect of need-based grants on
Betts, J.  R., & McFarland, L.  L. (1995). Safe port in a college access, persistence, and graduation (NBER
storm: The impact of labor market conditions on Working Paper w19306). Cambridge, MA: National
community college enrollments. Journal of Human Bureau of Economic Research.
Resources, 30(4), 741–765. Castleman, B.  L., & Page, L.  C. (2015). Freshman year
Bird, K., & Castleman, B.  L. (2014). Here today, gone financial aid nudges: An experiment to increase
tomorrow? Investigating rates and patterns of finan- FAFSA renewal and college persistence. Journal of
cial aid renewal among college freshmen. Research in Human Resources, 51(2), 389–415.
Higher Education, 57(4), 395–422. Cha, P., & Patel, R. (2010). Rewarding progress, reducing
Blackwell, L.  S., Trzesniewski, K.  H., & Dweck, C.  S. debt: Early results from Ohio’s performance-based
(2007). Implicit theories of intelligence predict scholarship demonstration for low-income parents.
achievement across an adolescent transition: A longi- New York: MDRC.
tudinal study and an intervention. Child Development, Chancellor’s Office of the California Community
78(1), 246–263. Colleges. (2003). The real cost project: Preliminary
Bloom, D., & Sommo, C. (2005). Building learning report. Sacramento: California Community Colleges.
Retrieved from http://www.immagic.com/eLibrary/
communities: Early results from the Opening Doors ARCHIVES/GENERAL/CCC_CAUS/C030929R.pdf
Demonstration at Kingsborough community college. Clark, B. R. (1960). The “cooling-out” function in higher
New York: MDRC. education. American Journal of Sociology, 65(6),
Blumer, H. (1971). Social problems as collective behav- 569–576.
ior. Social Problems, 18(3), 298–306. Cohen, A. M., Brawer, F. B., & Kisker, C. B. (2014). The
Boatman, A., & Long, B.  T. (2010). Does remediation American community college (6th ed.). San Francisco:
John Wiley & Sons.
work for all students? How the effects of postsec- Crisp, G., & Delgado, C. (2014). The impact of devel-
opmental education on community college persistence
ondary remedial and developmental courses vary and vertical transfer. Community College Review,
by level of academic preparation (NCPR Working 42(2), 99–117.
Paper). New York: National Center for Postsecondary Crosta, P. M. (2014). Intensity and attachment: How the
Research. chaotic enrollment patterns of community college
Brand, J.  E., Pfeffer, F.  T., & Goldrick-Rab, S. (2012).
Interpreting community college effects in the pres-

556 D. Monaghan et al.

students relate to educational outcomes. Community Evans, B., & Henry, G. (2015, March). Self-paced remedi-
College Review, 42(2), 118–142. ation and math placement: A randomized field experi-
Dachelet, K., & Goldrick-Rab, S. (2015). Investing in ment in a community college. Presented at the spring
conference of the Society for Research on Educational
student completion: Overcoming financial barriers to Effectiveness. Washington, DC.

retention through small-donor grants and emergency Evans, W., Kearney, M., & Sullivan, J.  (2014). Stay the
aid programs. Madison: Wisconsin HOPE Lab. course: Evaluating an intervention to promote com-
Deil-Amen, R. (2011). Socio-academic integrative munity college persistence and graduation rates (AEA
moments: Rethinking academic and social integra- RCT Registry). https://www.socialscienceregistry.org/
tion among two-year college students in career-related trials/223/history/1148
programs. The Journal of Higher Education, 82(1),
54–91. Fairlie, R. W., & Grunberg, S. H. (2014). Access to tech-
Deil-Amen, R., & DeLuca, S. (2010). The underserved nology and the transfer function of community col-
third: How our educational structures populate an edu- leges: Evidence from a field experiment. Economic
cational underclass. Journal of Education for Students Inquiry, 52(3), 1040–1059.
Placed at Risk, 15(1–2), 27–50.
Deil-Amen, R., & Rosenbaum, J.  E. (2002). The unin- Fairlie, R.  W., & London, R.  A. (2012). The effects of
tended consequences of stigma-free remediation. home computers on educational outcomes: Evidence
Sociology of Education, 75(3), 249–268. from a field experiment with community college stu-
Deil-Amen, R., & Rosenbaum, J.  E. (2003). The social dents. The Economic Journal, 122(561), 727–753.
prerequisites of success: Can college structure reduce
the need for social know-how? The Annals of the Fike, D. S., & Fike, R. (2008). Predictors of first-year stu-
dent retention in the community college. Community
American Academy of Political and Social Science, College Review, 36(2), 68–88.
586(1), 120–143.
Deming, D., & Dynarski, S. (2010). College aid. In P. B. Geckler, C., Beach, C., Pih, M., & Yan, L. (2008). Helping
Levine & D. J. Zimmerman (Eds.), Targeting invest- community college students cope with financial emer-

ments in children: Fighting poverty when resources gencies: Lessons from the Dreamkeepers and Angel
are limited (pp.  283–302). Chicago: University of Fund financial aid programs. New York: MDRC.
Chicago Press. Goldrick-Rab, S. (2010). Challenges and opportunities for
Denning, J.  T. (2014). College on the cheap: Costs and improving community college student success. Review
benefits of community college (Working Paper). of Educational Research, 80(3), 437–469.
Retrieved from http://www.terry.uga.edu/media/ Goldrick-Rab, S. (2016). Paying the price: College costs,
events/documents/Denning_Paper.pdf financial aid, and the betrayal of the American dream.
Derby, D. C., & Smith, T. (2004). An orientation course Chicago: University of Chicago Press.
and community college retention. Community College Goldrick-Rab, S., & Kinsley, P. (2013). School integra-
Journal of Research and Practice, 28(9), 763–773. tion and the open-door philosophy: Rethinking the
Desrochers, D.  M., & Kirshstein, R.  J. (2012). College economic and racial composition of community col-
leges. In Bridging the higher educational divide:
spending in a turbulent decade: Findings from the Strengthening community colleges and restoring the
Delta Cost Project: A delta data update, 2000–2010.
Washington, DC: American Institutes for Research. American dream: The report of the Century Task
Dougherty, K.  J., Jones, S.  M., Lahr, H., Natow, R.  S.,
Pheatt, L., & Reddy, V. (2016). Looking inside the Force on preventing community colleges from becom-
black box of performance funding for higher educa- ing separate and unequal (pp. 109–136). New York:
tion: Policy instruments, organizational obstacles, The Century Foundation Press.
and intended and unintended impacts. The Russell Goldrick-Rab, S., Harris, D. N., & Trostel, P. A. (2009).
Sage Foundation Journal of the Social Sciences, 2(1), Why financial aid matters (or does not) for college
147–173. success: Toward a new interdisciplinary perspec-
Douglas, D., & Attewell, P. (2014). The bridge and the tive. In Higher education: Handbook of theory and
troll underneath: Summer bridge programs and degree research (pp. 1–45). Dordrecht: Springer.
completion. American Journal of Education, 121(1), Goldrick-Rab, S., Broton, K., & Frank, V.  M. (2014).
87–109. Single Stop USA’s Community College Initiative:
Dynarski, S. M. (2003). Does aid matter? Measuring the Implementation assessment. Madison: Wisconsin
effect of student aid on college attendance and com- HOPE Lab.
pletion. American Economic Review, 93(1), 279–288. Goldrick-Rab, S., Broton, K., & Eisenberg, D. (2015).
Dynarski, S.  M., & Scott-Clayton, J.  (2013). Financial Hungry to learn: Addressing food and housing inse-
aid policy: Lessons from research (NBER Working curity among undergraduates. Madison: Wisconsin
Paper w18710). Cambridge, MA: National Bureau of HOPE Lab.
Economic Research. Goldrick-Rab, S., Kelchen, R., Harris, D. N., & Bensen,
Eagan, M. K., Jr., & Jaeger, A. J. (2009). Effects of expo- J. (2016). Reducing income inequality in educational
sure to part-time faculty on community college trans- attainment: Experimental evidence on the impact
fer. Research in Higher Education, 50(2), 168–188. of financial aid on college completion. American
Sociological Review, 121(6), 1762–1817.
Grissmer, D.  W., Subotnik, R.  F., & Orland, M. (2009).
A guide to incorporating multiple methods in ran-

24  Experimental Evidence on Interventions to Improve Educational Attainment at Community Colleges 557

domized controlled trials to assess intervention Kezar, A., & Maxey, D. (2013, June). The chang-
effects. Washington, DC: American Psychological ing academic workforce. Trusteeship Magazine.
Association. Retrieved from http://agb.org/trusteeship/2013/5/
Grossman, J.  B., Quint, J., Gingrich, J., Cerna, O., changing-academic-workforce
Diamond, J., Levine, A., & Willard, J.  (2015).
Kirshstein, R. J., & Hubert, S. (2012). Revenues: Where
Changing community colleges: Early lessons from does the money come from? A Delta update (pp. 2000–
Completion by Design. New York: MDRC. 2010). Washington, DC: American Institutes for
Grubb, W. N. (2001). “Getting into the world”: Guidance Research.
and counseling in community colleges. New  York:
Community College Research Center. Lareau, A. (2008). Narrow questions, narrow answers:
Grubb, W.  N. (2006). “Like, what do I do now?” The The limited value of randomized control trials for edu-
dilemmas of guidance counseling. In T.  Bailey & cation research. In P. B. Walters, A. Lareau, & S. Rains
V. S. Morest (Eds.), Defending the community college (Eds.), Education research on trial: Policy reform and
equity agenda (pp.  195–222). Baltimore: The Johns the call for scientific rigor (pp. 145–162). New York:
Hopkins University Press. Routledge.
Harackiewicz, J. M., Tibbetts, Y., Canning, E., & Hyde,
J.  S. (2014). Harnessing values to promote motiva- Leigh, D. E., & Gill, A. M. (2003). Do community col-
tion in education. Motivational Interventions, 18, leges really divert students from earning bachelor’s
71–105. degrees? Economics of Education Review, 22(1),
Harackiewicz, J. M., Canning, E. A., Tibbetts, Y., Priniski, 23–30.
S. J., & Hyde, J. S. (2015). Closing achievement gaps
with a utility-value intervention: Disentangling race Linderman, D., & Kolenovic, Z. (2012). Results this far
and social class. Journal of Personality and Social
Psychology, 111(5), 745–765. and the road ahead: A follow-up report on CUNY
Harris, D.  N., & Goldrick-Rab, S. (2012). Improving Accelerated Study in Associate Programs (ASAP).
the productivity of education experiments: Lessons New York: The City University of New York.
from a randomized study of need-based financial aid. Logue, A.  W., Watanabe-Rose, M., & Douglas, D.
Education, 7(2), 143–169. (2016). Should students assessed as needing remedial
Heckman, J. J. (2005). The scientific model of causality. ­mathematics take college-level quantitative courses
Sociological Methodology, 35(1), 1–97. instead? Educational Evaluation and Policy Analysis,
Heckman, J.  J., & Smith, J.  A. (1995). Assessing the 38(3), 578–598.
case for social experiments. The Journal of Economic Martorell, P., & McFarlin, I., Jr. (2011). Help or hin-
Perspectives, 9(2), 85–110. drance? The effects of college remediation on aca-
Hillman, N. W., Tandberg, D. A., & Gross, J. P. (2014). demic and labor market outcomes. The Review of
Performance funding in higher education: Do financial Economics and Statistics, 93(2), 436–454.
incentives impact college completions? The Journal of Mayer, A. K., Patel, R., & Gutierrez, M. (2015). Four-year
Higher Education, 85(6), 826–857.
Hughes, K.  L., & Scott-Clayton, J.  (2011). Assessing effects on degree receipt and employment outcomes
developmental assessment in community colleges.
Community College Review, 39(4), 327–351. from a performance-based scholarship program in
Jenkins, D., & Belfield, C. (2014). Can community col- Ohio. New York: MDRC.
leges continue to do more with less? Change: The Melguizo, T., Hagedorn, L.  S., & Cypers, S. (2008).
Magazine of Higher Learning, 46(3), 6–13. Remedial/developmental education and the cost of
Kahlenberg, R. (2015). How higher education short- community college transfer: A Los Angeles County
changes community colleges. Washington, DC: The sample. The Review of Higher Education, 31(4),
Century Foundation. 401–431.
Kane, T.  J. (2003). A quasi-experimental estimate of Melguizo, T., Bos, J., & Prather, G. (2011). Is develop-
the impact of financial aid on college-going (NBER mental education helping community college students
Working Paper w9703). Cambridge, MA: National persist? A critical review of the literature. American
Bureau of Economic Research. Behavioral Scientist, 55(2), 173–184.
Karp, M.  M. (2011). Toward a new understanding of Merton, R.  K. (1936). The unanticipated consequences
of purposive social action. American Sociological
non-academic student support: Four mechanisms Review, 1(6), 894–904.
Miller-Adams, M. (2015). Promise nation: Transforming
encouraging positive student outcomes in the com- communities through place-based scholarships.
munity college (CCRC Working Paper 28). New York: Kalamazoo: W.E.  Upjohn Institute for Employment
Community College Research Center. Research.
Kena, G., Musu-Gilette, L., Robinson, J., Wang, X., Minkler, J. E. (2002). ERIC review: Learning communi-
Rathbun, A., Zhang, J., Wilkinson-Flicker, S., Barmer, ties at the community college. Community College
A., & Velez, E. D. (2015). The condition of education Review, 30(3), 46–63.
2015. Washington, DC: National Center for Education Monaghan, D. B., & Attewell, P. (2015). The community
Statistics, U.S. Department of Education. college route to the bachelor’s degree. Educational
Evaluation and Policy Analysis, 37(1), 70–91.
Morgan, S. L., & Winship, C. (2014). Counterfactuals and

causal inference: Methods and principles for social
research. New York: Cambridge University Press.

558 D. Monaghan et al.

Moss, B. G., & Yeaton, W. H. (2013). Evaluating effects Rubin, D.  B. (1974). Estimating causal effects of treat-
of developmental education for college students using ments in randomized and nonrandomized studies.
a regression discontinuity design. Evaluation Review, Journal of Educational Psychology, 66(5), 688–701.
37(5), 370–404.
Schneider, B., & McDonald, S.  K. (2007a). Scale-up
National Student Clearinghouse Research Center. (2015). in education: Vol. I: Ideas in principle. Lanham:
Snapshot report: Persistence—Retention. Herndon: Rowman and Littlefield Publishers.
National Student Clearinghouse.
Schneider, B., & McDonald, S.  K. (2007b). Scale-up
O’Gara, L., Karp, M.  M., & Hughes, K.  L. (2009). in education: Vol. II: Ideas in practice. Lanham:
Student success courses in the community college: An Rowman and Littlefield Publishers.
exploratory study of student perspectives. Community
College Review, 36(3), 195–218. Schudde, L. T. (2011). The causal effect of campus resi-
dency on college student retention. The Review of
Park, V. C., Nations, J., & Nielsen, K. (2013). What mat- Higher Education, 34(4), 581–610.

ters for community college success? Assumptions and Scott-Clayton, J.  (2015). The shapeless river: Does a
lack of structure inhibit students’ progress at com-
realities concerning student supports for low-income munity colleges? In B.  J. Castleman, S.  Schwartz,
women. Los Angeles: All-Campus Consortium on & S.  Baum (Eds.), Decision making for student
Research for Diversity. success: Behavioral insights to improve college
Pascarella, E. T., & Terenzini, P. T. (2005). How college access and persistence (pp.  102–123). New  York:
affects students: A third decade of research (Vol. 2). Routledge.
San Francisco: Jossey-Bass.
Patel, R., & Rudd, T. (2012). Can scholarships alone help Scott-Clayton, J., & Rodriguez, O. (2012). Development,
discouragement, or diversion? New evidence on
students succeed? Lessons from two New  York City the effects of college remediation (NBER Working
community colleges. New York: MDRC. Paper w18328). Cambridge, MA: National Bureau of
Patel, R., & Valenzuela, I. (2013). Moving forward: Early Economic Research.

findings from the Performance-Based Scholarship Scrivener, S., & Au, J. (2007). Enhancing student services
Demonstration in Arizona. New York: MDRC. at Lorain County Community College: Early results
Patel, R., Ritchburg-Hayes, L., de la Campa, E., & Rudd, from the Opening Doors Demonstration in Ohio.
T. (2013). Performance-based scholarships: What New York: MDRC.
have we learned? In Interim findings from the PBS
demonstration. New York: MDRC. Scrivener, S., & Pi, M. (2007). Enhancing student services
Perin, D. (2006). Can community colleges protect both at Owens Community College: Early results from the
access and standards? The problem of remediation. Opening Doors Demonstration in Ohio. New  York:
Teachers College Record, 108(3), 339. MDRC.
Radford, A. W., & Horn, L. (2012). An overview of classes
Scrivener, S., & Weiss, M.  J. (2009). More guidance,
taken and credits earned by beginning postsecond- better results? Three-year effects of an enhanced stu-
ary students (Web tables). Washington, DC: National dent services program at two community colleges.
Center for Education Statistics, U.S.  Department of New York: MDRC.
Education.
Raftery, S. (2005). Developmental learning communities Scrivener, S., Blom, D., LeBlanc, A., Paxton, C., Rouse,
at metropolitan community college. New Directions C., & Sommo, C. (2008). A good start: Two-year effects
for Community Colleges, 2005(129), 63–72. of a learning community program at Kingsborough
Reynolds, C. L. (2012). Where to attend? Estimating the Community College. New York: MDRC.
effects of beginning college at a two-year institution.
Economics of Education Review, 31(4), 345–362. Scrivener, S., Sommo, C., & Collado, H. (2009). Getting
Ritchburg-Hayes, L., Brock, T., LeBlanc, A., Paxson, C., back on track: Effects of a community college program
Rouse, C. E., & Barrow, L. (2009). Rewarding persis- for probationary students. New York: MDRC.

tence: Effects of a performance-based scholarship for Scrivener, S., Weiss, M. J., & Sommo, C. (2012). What
low-income parents. New York: MDRC. can a multifaceted program do for community col-
Ritchburg-Hayes, L., Sommo, C., & Welbeck, R. (2011). lege students? Early results from an evaluation of
Accelerated Study in Associate Programs (ASAP)
Promoting full-time attendance among adults in com- for developmental education students. New  York:
MDRC.
munity college: Early impacts from the Performance-
Based Scholarship Demonstration in New  York. Scrivener, S., Weiss, M.  J., Ratledge, A., Rudd, T.,
New York: MDRC. Sommo, C., & Fresques, H. (2015). Doubling gradu-
Rosenbaum, J. E., Deil-Amen, R., & Person, A. E. (2006). ation rates: Three-year effects of CUNY’s Accelerated
Study in Associate Programs (ASAP) for developmen-
After admission: From college access to college suc- tal education students. New York: MDRC.
cess. New York: Russell Sage Foundation.
Rouse, C. E. (1995). Democratization or diversion? The Shapiro, D., Dundar, A., Yuan, X., Harrell, A., &
effect of community colleges on educational attain- Wakhungu, P.  K. (2014). Completing college: A
ment. Journal of Business & Economic Statistics, national view of student attainment rates: Fall 2008
13(2), 217–224. cohort (Signature report No. 8). Herndon: National
Student Clearinghouse Research Center.

24  Experimental Evidence on Interventions to Improve Educational Attainment at Community Colleges 559

Sharon, A. T. (1972). Assessing the effectiveness of reme- sion–discontinuity approach. International Economic
dial college courses. The Journal of Experimental Review, 43(4), 1249–1287.
Education, 41(2), 60–62. Visher, M. G., & Torres, J. (2011). Breaking new ground:
An impact study of career-focused learning communi-
Simone, S. (2014). The transferability of postsecondary ties at Kingsborough Community College. New York:
credit following student transfer or coenrollment. National Center for Postsecondary Research.
Washington, DC: National Center for Education Visher, M.  G., Butcher, K.  F., & Cerna, O.  S. (2010).
Statistics, U.S. Department of Education. Guiding developmental math students to campus ser-
vices: An impact evaluation of the Beacon Program at
Singell, L. D. (2004). Come and stay a while: Does finan- South Texas College. New York: MDRC.
cial aid effect retention conditioned on enrollment at Ware, M., Weissman, E., & McDermott, D. (2013). Aid
a large public university? Economics of Education Like a Paycheck: An incremental aid program to pro-
Review, 23(5), 459–471. mote student success. New York: MDRC.
Wathington, H. D., Barnett, E. A., Weissman, E., Teres, J.,
Sommo, C., Boynton, M., Collado, H., Diamond, J., Pretlow, J., & Nakanishi, A. (2011). Getting ready for
Gardenhire, A., Ratledge, A., Rudd, T., & Weiss, M. college: An implementation and early impacts study of
(2014). Mapping success: Performance-based schol- eight Texas developmental summer bridge programs.
New York: MDRC.
arships, student services, and developmental math Weiss, M.  J., Visher, M.  G., & Wathington, H. (2010).
at Hillsborough Community College. New  York: Learning communities in developmental reading: An
MDRC. impact study at Hillsborough Community College.
Stuart, G. R., Rios-Aguilar, C., & Deil-Amen, R. (2014). New  York: National Center for Postsecondary
“How much economic value does my credential Research.
have?” Reformulating Tinto’s model to study stu- Weiss, M. J., Brock, T., Sommo, C., Rudd, T., & Turner,
dents’ persistence in community colleges. Community M. C. (2011). Serving community college students on
College Review, 42(4), 327–341. probation: Four-year findings from Chaffey College’s
Sturgis, I. (2014, January 8). Gates foundation invests Opening Doors program. New York: MDRC.
half-billion in success of community college students. Weiss, M.  J., Mayer, A., Cullinan, D., Ratledge, A.,
Diverse Issues in Higher Education. Retrieved from Sommo, C., & Diamond, J.  (2014). A random
http://diverseeducation.com/article/59973/ assignment evaluation of learning communities at
The Century Foundation. (2013). Bridging the higher Kingsborough Community College: Seven years later.
New York: MDRC.
educational divide: Strengthening community colleges Weiss, M.  J., Mayer, A.  K., Cullinan, D., Ratledge,
A., Sommo, C., & Diamond, J.  (2015). A ran-
and restoring the American dream: The report of the dom assignment evaluation of learning communi-
ties at Kingsborough Community College—Seven
Century Task Force on preventing community colleges years later. Journal of Research on Educational
from becoming separate and unequal. New York: The Effectiveness, 8(2), 189–217.
Century Foundation Press. Weissman, E., Butcher, K.  F., Schneider, E., Teres, J.,
The College Board. (2015). Trends in student aid 2015. Collado, H., & Greenberg, D. (2011). Learning
New York: The College Board. communities for students in developmental math:
Tinto, V. (1987). Leaving college: Rethinking the causes Impact studies at Queensborough and Houston
and cures of student attrition. Chicago: University of Community Colleges. New York: National Center for
Chicago Press. Postsecondary Research.
Tinto, V. (1997). Colleges as communities: Taking Weissman, E., Cullinan, D., Cerna, O., Safran, S., &
research on student persistence seriously. The Review Richman, P. (2012). Learning communities in devel-
of Higher Education, 21(2), 167–177. opmental English: Impact studies at Merced College
Tinto, V., Russo, P., & Kadel, S. (1994). Constructing and the Community College of Baltimore. New York:
educational communities: Increasing retention in chal- National Center for Postsecondary Research.
lenging circumstances. Community College Journal, Yeager, D.  S., & Dweck, C.  S. (2012). Mindsets that
64(4), 26–29. promote resilience: When students believe that per-
Trow, M. (2007). Reflections on the transition from elite to sonal characteristics can be developed. Educational
mass to universal access: Forms and phases of higher Psychologist, 47(4), 302–314.
education in modern societies since WWII.  In J.  J. Yeager, D.  S., Paunesku, D., Romero, C., Brown, R.,
F. Forest & P. G. Altbach (Eds.), International hand- Muhich, J., & Walton, G. (2013). Engineering psy-
book of higher education (pp.  243–280). Dordrecht: chological interventions for scale: The case of the
Springer Netherlands. “growth mindset” in community colleges (Working
Turner, L. (2015, November). Loan nudges: Experimental Paper). Austin: University of Texas at Austin.
Zeidenberg, M., Jenkins, P. D., & Calcagno, J. C. (2007).
evidence on the effects of default options on the fram- Do student success courses actually help community
ing of federal loans. Presented at the conference college students succeed? (CCRC Brief No. 36).
of the Association for Public Policy Analysis and New York: Community College Research Center.
Management. Miami.
Valentine, J.  C., Konstantopolous, S., & Goldrick-Rab,
S. (2016). A meta-analysis of regression discontinu-

ity studies investigating the effects of placement into
remedial education: A working paper. Madison:
Wisconsin HOPE Lab.
Van der Klaauw, W. (2002). Estimating the effect of
financial aid offers on college enrollment: A regres-

Research–Practice Partnerships 25
in Education

Paula Arce-Trigatti, Irina Chukhray,
and Ruth N. López Turley

Abstract relationships between researchers and practitio-
The field of education has seen a sharp ners that are formal and long-term in nature,
increase in the formation and participation of there has been a notable recent increase in their
research–practice partnerships (RPPs) over formation and persistence in education. In this
the last two decades. Bringing together two chapter, we seek to understand why partnerships
parties in education that share a concern for have been accepted as an important strategy for
improved student outcomes but differ dramati- potentially addressing the long-e­ stablished edu-
cally in their approaches to that end, RPPs in cation research-to-practice gap using theoretical
education have not only grown in number and foundations grounded in sociology. Guiding our
type, but complementary organizations and work is a key concept from organizational the-
efforts have begun to emerge as well. In this ory: the description of an organization’s envi-
contribution, we explore the reasons for these ronment as a field. In particular, knowing how
changes, grounding our work in the organiza- the field is structured or organized, understand-
tional and institutional theories literature from ing the individual organizations within the field,
sociology. and defining the challenges faced by organiza-
tions are especially useful for gaining an under-
25.1 I ntroduction standing of organizations. To that end, we
explore the following three questions: First,
The world of education research–practice part- have we actually seen an increased presence of
nerships (RPPs) has evolved dramatically over RPPs in education? How might we account for
the last two decades. Perhaps most simply the rapid growth in the number of RPPs in the
understood as collaborative, mutually beneficial last two decades given that institutional con-
straints typically slow adoption of new innova-
P. Arce-Trigatti (*) · I. Chukhray · tions? Second, how can organizational theory
R. N. López Turley help illustrate why we observe multiple models
Rice University, Houston, TX, USA or types of RPPs in education? Finally, we note
e-mail: [email protected]; [email protected]; that in addition to RPPs, complementary organi-
[email protected] zations and efforts (e.g., those related to RPPs
but are not in and of themselves partnerships)
have emerged as well. What role do these com-
plementary organizations play within the larger
RPP ecosystem?

© Springer International Publishing AG, part of Springer Nature 2018 561
B. Schneider (ed.), Handbook of the Sociology of Education in the 21st Century, Handbooks
of Sociology and Social Research, https://doi.org/10.1007/978-3-319-76694-2_25

562 P. Arce-Trigatti et al.

Fig. 25.1  Illustration of the development of research– launch of the UChicago Consortium on School Research
practice partnerships over time separately, to demonstrate that until their arrival, there
Notes: This figure helps illustrate the growth in research– was a virtual drought of RPPs for an extended time. After
practice partnerships in education over time. The numbers their launch, several RPPs modeled after the UChicago
to the right of each category indicate how many RPPs of Consortium emerged (“research alliances”), as well as a
that type emerged starting in a particular year. We note the large collection of IES-sponsored RPPs

25.2 Is There an Increase in RPPs? Historically, the research-to-practice model
How Can Organizational that has been the modus operandi of many con-
Theory Help Us Understand sisted of a one-way conversation between educa-
the Growth Patterns of RPPs? tion practitioners and researchers (see Huberman

1994 for a clear illustration of this model). This

We begin our discussion by inquiring if there has simple linear model, where basic research leads

indeed been a spike in the number of research– to applied research, which then leads to the

practice partnerships (RPPs) more recently rela- development of products and/or professional

tive to their historical development. Quite simply, practices, and finally, dissemination to educa-

data suggest this is the case. Figure 25.1 displays tional practitioners and systems demonstrates

the pattern of growth of several types of RPPs the difficulty in changing frameworks once they

over time. As shown in the figure, the IES-­ have been accepted as the norm. How we con-

sponsored Regional Education Laboratories were ceptualize the problem with the research–prac-

the main type of RPP in the U.S. for multiple tice gap in education matters because it influences

decades, before the launch of the UChicago the policy solutions that are pursued. Because of

Consortium on School Research in 1990. the linear model assumption, for example, there

Following a much shorter dry spell, the next wave has been special attention devoted to the role of

of RPPs began in the mid-2000s, with several “linking agents,” which are organizations or

partnerships modeled after the UChicago individuals who transform research findings into

Consortium as well as a large collection (over 20 understandable material for the public (Hood

new grantees) of IES-funded RPPs emerging 1982). Efforts to improve this translational gap

onto the landscape. Related research investigat- led to federal funding for the Regional Education

ing school district decision-making processes Laboratories, for example, which have been

also support the notion of a recent increased pres- around since the 1960s (Coburn and Stein

ence of RPPs (e.g., Honig and Coburn 2008).

25  Research–Practice Partnerships in Education 563

2010).1 Other federal initiatives that have lier), the slow research process itself compared to
emerged based on this framing include the the rapidity with which practitioners need infor-
Education Resources Information Center (ERIC) mation, and the perception of researchers as out-
funded by the Institute of Education Sciences side agents that can provide little usable
(IES), which is a “nationwide information net- knowledge for the classroom. Huberman (1989)
work that acquires, catalogs, summarizes, and additionally finds that research use within dis-
provides access to education information from tricts is heavily dependent on the social transac-
all sources,”2 as well as three clearinghouses tions within that setting, including how leaders
(e.g., National Clearinghouse for Comprehensive relate to the research or which parties are impli-
School Reform, National Clearinghouse for cated in the policy recommendations, for exam-
Educational Facilities, and the What Works ple. In sum, many of these examples reiterate the
Clearinghouse) that serve solely to disseminate dichotomous and distinct environments in which
different types of research and information in practitioners and researchers operate. Changing
education. On the research front, there has been accepted ways of doing business, however, is
a large effort to study how to improve district oftentimes a slow process. We next turn to insti-
access to research (e.g., Coburn and Stein 2010). tutional theories to help us reconcile the initial
slow growth of RPPs in light of the inadequacies
The linear model in practice, however, is prob- of the linear model.
lematic because reality quite often deviates from
this clearly structured pathway. Weiss (1980) 25.2.1 Initial Slow RPP Growth
describes a decidedly non-linear way in which
policymakers interact with research, suggesting Institutional theory can at least partially explain
that information gradually filters through multi- the consistent behavioral patterns of organiza-
ple channels and “creeps” into thought processes. tions (in our case, districts or schools from the
Simply releasing findings to practitioners is practitioner side and academics from the
therefore insufficient (Spillane et  al. 2002). researcher side) (Zucker 1987; Powell and
Fleming (1988) documents the myriad chal- DiMaggio 1991). In this context, “institutions”
lenges explaining why teachers are so unlikely to may have developed around both sets of parties
use research in their activities. Some examples that have likely contributed to the delayed emer-
include the overwhelming amount of time it takes gence of RPPs. For example, institutional theo-
to find research and interpret it (hence the focus rists argue that organizations adopt procedures,
on translational linking agents mentioned ear- formal structures, and vocabularies consistent
with expectations of what is “acceptable” and
1 Title IV of ESEA authorized the formation of what even- “legitimate” given their operating environment.
tually came to be known as the “Regional Education The importance of institutional legitimacy is
Laboratories” (RELs). The motivating ideas behind the underscored by its role in ensuring organizational
introduction of the RELs were first, to facilitate the gen- survival: Adopting innovative practices are often
eration of more useful research in education and second, viewed as threatening and incompatible with the
to somehow encourage practitioners to actually use it internal structures of the organization. We next
(Guthrie 1989). In the years since their inception, the explore how these pressures may have shaped the
RELs have been reauthorized several times and are cur- slow participation of first, researchers and sec-
rently operating 79 research alliances within ten different ond, practitioners in collaborating within an RPP.
RELs across the U.S. They are funded by the Institute of
Education Sciences (IES) to conduct research, dissemi- For academics, typical research culture has
nate findings, and to provide training and technical assis- been notoriously isolated (e.g., “ivory tower”):
tance to link research-proven practices with educational Building from existing theories, developing new
practitioners. In their most recent iteration, there will be
greater opportunity to engage with research–practice part-
nerships (Sparks 2016).

2 See: http://www2.ed.gov/about/contacts/gen/othersites/
eric.html.

564 P. Arce-Trigatti et al.

ones, testing them out, and so forth are activities resource dependence theory suggests that
that have been carried out in concert with other organizations respond to external actors who
academics or solo. Additionally, academic control the resources upon which the organization
researchers are subject to other primary depends; this seems to apply to universities
objectives, such as meeting tenure, which directly (Kraatz and Zajac 2001). We would predict that
influences their choices over what types of based on this theory, if grant makers at the federal
research activities to engage in. Institutional and private foundational levels increasingly
forces such as peer recognition and the potential include line items for RPP start-up and persistence
to improve one’s position in his/her respective costs, this might mitigate the reluctance of
field may have led to adherence towards these universities to invest in such organizations.
more traditional research inquiries. Indeed, these Indeed, the Spencer Foundation just recently
influences shape how scientists choose to pursue launched a new competitive grant program for
a particular research problem; innovation, in partnerships.3 Greater funding opportunities for
terms of novel methodologies or questions that academic researchers from outside their
divert towards a greater focus on practice for universities may thus have helped fuel the growth
example, is perceived as a gamble (Foster et al. of RPPs.
2015). When academics pursue involvement in
developing or contributing to an RPP, they are Finally, other forces, such as policy pressures,
essentially taking a gamble, given the large time may pull universities in directions opposite to the
commitment that is required, as well as the institutional norms or financial pressures.
distinct shift in the types of research questions Institutional theorists argue that changes in the
that could potentially be examined. Thus, we external environment can facilitate new logics to
might predict an initial slow growth of RPPs spread within a given organization (Berman and
while such entities remain squarely in a “novel” Stivers 2016). For example, Berman (2012)
phase. Consider that after the UChicago documents how changes in U.S. policies to
Consortium on School Research launched in promote science led universities to increasingly
1990, more than a decade passed before new focus on the economic value of science.
RPPs with similar arrangements began to emerge. Furthermore, new resources became available to
Further complicating matters, researchers help support this new focus. Within this
generally operate with much longer timelines framework, we might expect a similar pattern for
relative to practitioners, who commonly need RPPs, as the U.S. government continues to
information to make policy decisions very promote the use of evidence to inform decision
quickly. With a set of organizational norms that making for education policymakers through laws
differ dramatically from those practitioners face, such as No Child Left Behind (NCLB) and the
it is not at all surprising that partnerships in Every Student Succeeds Act (ESSA). NCLB
education have taken such a long time to “take (2002) requires evidence from “scientifically
off” and instead, simpler solutions of “linking based research,” while ESSA (2015) has kept the
agents” have been adopted. spirit of the law in tact, but has broadened
available research to include “evidence based
Financial pressures on the university may research.” Policy changes in the external
have also contributed to slow institutional support environment such as these may present new
for RPPs. In the last few decades, institutions of opportunities for researchers to engage with
higher education have experienced massive cuts partnerships that were absent before.
in federal and state funding, impacting research
universities the most (Scott and Biag 2016). In Turning our attention to U.S. schools and dis-
addition to the risk associated with pursuing tricts, we begin by focusing on the influence of
more innovative research paths, funding institutional norms in shaping these organiza-
challenges may have also presented an obstacle
to greater investments in RPPs. In particular, 3 See: http://www.spencer.org/research-practice-partnership-
program.

25  Research–Practice Partnerships in Education 565

tions, and the long-lasting effects that can accom- district central offices. This means that new
pany these forces. With roots in rural areas, models of professional practice that require cen-
one-room schoolhouses in the U.S. functioned as tral office administrators to break some previ-
efficient transmitters of basic skills essential for ously held routines may be necessary to enable a
societal success. With the arrival of the industrial new culture of evidence-based decisions (Honig
revolution, however, educators felt pressure to 2006). Absent those, we would expect a slow
make education more systematic, mimicking the adoption of RPPs among school districts to gen-
factory-type model common in the business sec- erally be the case.
tor (Tyack 1974). The resulting bureaucratization
of the educational system in general was, some Beyond the cultural factors or norms that may
would argue, necessary to adequately address cur- inhibit districts from embracing research, there
rent needs. A combination of businessmen, uni- are other constraints that may contribute to this
versity professors and presidents, school deterrence as well. Burch and Thiem (2004) and
superintendents, and middle-class reformers Reichardt (2000) suggest that central office
facilitated the shift to a centralized school system administrators may lack the human capital and
with a top-down structure of school management. technological infrastructure to engage in
Post centralization, school boards were comprised evidence-based decision making. Additionally,
mostly of business people and professionals, as it working knowledge also appears to strongly
largely remains today. To be clear, schools and mediate evidence use. For example, some
districts in the U.S. were not created with the research finds that district central offices are
explicit goal of acting as research and develop- more likely to search and pay attention to
ment (R&D) centers, which certainly contributes evidence that fits in with their conceptions or
to the lack of research capacity that still pervades conforms to their expectations (Birkeland et  al.
the practitioner side today. With the absence of 2005; Spillane 2000). Individual preferences for
R&D as a primary function within schools, it certain types of evidence can also play a role
should not be surprising that successful practitio- (Coburn and Talbert 2006). Collectively, these
ner interaction with research has been marred by additional constraints may also lessen the
numerous challenges. Work by Rowan (1982) likelihood that a district is willing or able to work
suggests that an institutional environment charac- collaboratively with external researchers in an
terized as contentious and unfocused (which may RPP.
commonly occur in schools and districts given the
multiplicity of objectives that comes with numer- However, despite factors hindering evidence
ous stakeholders) dramatically slows the adoption use, district central offices have long used some
of innovative structures. form of evidence in their decision making. For
example, practitioner or “local” knowledge, such
The culture surrounding the use of research as input from principals, teachers, parents, and
evidence, in general, can also mediate how dis- students, is very common (Gonzalez et al. 2005;
tricts and schools interact with research. For Datnow et  al. 2002). Districts have also been
example, Honig and Coburn (2008, p. 594) shown to consult with social science research and
define evidence use as a process that involves to incorporate the use of student-level data to
“searching for and incorporating evidence” into inform decisions (Honig and Coburn 2008;
decision making. Furthermore, evidence use Massell 2001; Massell and Goertz 2002). We
within districts is more likely when district would expect the culture of evidence use within
norms, expectations, and routines encourage districts to continue to evolve, even without the
ongoing engagement with empirical research presence of RPPs, especially given the recent
(Honig 2003; Honig and Coburn 2008; Massell passage of ESSA. ESSA partially shifts authority
2001; Corcoran et al. 2001). However, the prac- back into the hand of the localities, enhancing the
tice of evidence-based decision making is argu- role of state and local policymakers that was
ably not yet common or a “norm” for many previously more restrictive under NCLB (Strauss
2015). In particular, states will have a greater say

566 P. Arce-Trigatti et al.

in which standards are adopted, greater control had existed for over a decade. Additionally, the

over their accountability systems, and greater reauthorization of ESSA and the NCLB of 2001,

flexibility over their teacher certification led to a greater need for localized capacity to

requirements and evaluation systems (Klein conduct research. The NCLB Act’s heavy use of

2016). This flexibility creates a larger role for the phrase “evidence-based” for describing what

local policymakers. Second, matching NCLB’s types of education policies should be imple-

previous emphasis of utilizing “evidence-based” mented created further incentives for school dis-

research in decision making, ESSA also explicitly tricts to invest in research-related skill sets (Feuer

defines this term and describes four levels of et al. 2002). How, then, did the collective efforts

rigor for research. Taken together, the need for on these many fronts influence the growth of

evidenced-based interventions and leeway in partnerships?

standards adoption creates a unique demand First, there was recognition that the current

appropriate for RPPs to meet. As a result, ESSA state of affairs was inadequate. The critique of

contains features that may possibly lead either to long-held institutional norms led the way for new

a greater number of RPPs or create a larger role ideas on the role of practitioners and researchers

for existing RPPs. in the research process. Second, it is likely that the

perceived success of the UChicago Consortium

25.2.2 Recent Burgeoning RPP led to new definitions for “legitimacy” among

Growth these types of institutions. Kramer (1981)

describes four key roles that non-­profit organiza-

tions tend to perform that sets them apart from

Growing concern over the large gap between other sectors: vanguard role, value guardian role,

research and practice and the failures to address it advocacy role, and the service provider role. Of

began to gain traction in the late 1990s and early these, the vanguard role, where experimentation

2000. Huberman (1994, p. 14) reports on the with innovative approaches to processes or pro-

“state of the art” of knowledge utilization in edu- grams leads to non-profits serving mainly as

cation and recognizes “the proliferation of cen- agents of change best describes early RPPs. If

ters, laboratories, intermediate units, and these organizations are proven successful (i.e., the

collaborative enterprises…is a sign that the pro- longevity of the UChicago Consortium), other

cess of ‘knowledge transfer’ is active in several agencies are more likely to adopt them.

forms.” He further describes the importance of Universities, which have typically been inflexible

“sustained interactivity” between researchers and with their institutional rules as described earlier,

practitioners in producing research itself, going may have shifted their stance somewhat given the

so far as to describe this interaction as “mutual” reputation of the University of Chicago, for exam-

in its benefits to both sides. While Huberman cri- ple. Moreover, increased opportunities for fund-

tiques and offers changes to the aforementioned ing (i.e., through the Spencer grantee program

linear research-to-practice framework, he stops mentioned earlier, as well as numerous IES-

short of naming this new enterprise a “partner- sponsored RPP initiatives separate from the

ship.” In 2003, a major task force from the RELs) have accompanied these trends.

National Research Council investigating the cur- Third, changes to the external environment via

rent approaches to addressing the research-to-­ NCLB likely contributed to the change in

practice gap produced “Strategic Education institutional norms as well. Tolbert and Zucker

Research Partnership,” a report offering an (1983), for example, find that when coercive

actionable change to business as usual (National pressures (either direct or indirect pressures to

Research Council 2003). At this point, the first conform to institutional expectations) are large,

education research–practice partnership that con- such as the changes brought in by NCLB,

sisted of a university and school district pairing, organizations are quick to adopt new structures.

the UChicago Consortium on School Research, More specifically, these accountability policies

25  Research–Practice Partnerships in Education 567

required districts to invest more heavily in their sharing feature is part and parcel of the
longitudinal data systems in order to regularly alliance commitment. Coburn and colleagues
use student performance data as required (Kerr consider both the Regional Education
et al. 2005). Other federal grants such as the one Laboratories (RELs) and partnerships such as
provided by IES to help support statewide the UChicago Consortium on School Research
longitudinal data systems have also contributed to fit within this category.
to their increased presence. This particular Design-based partnerships: are structurally very
change opens new doors for researchers to similar to research alliances, in that they are
interact with practitioners, given the supreme typically comprised of district and university
importance of administrative data in conducting pairings, such as the Middle-School
research. Taken together, then, these may have Mathematics in the Institutional Setting of
influenced the acceptance of and the growing Teaching (MIST) project at Vanderbilt
interest in RPPs as a promising mechanism to University. The authors chose to distinguish
address the research-to-practice gap. this type of model based on their scope of work,
which departs from that of the research alli-
25.3 What Are the Different RPP ances in that it tends to be more narrow (e.g.,
Models? Why Are There problems of practice as they relate to curricu-
Multiple Models/Types lum and instruction only). Design-b­ ased part-
of Research–Practice nerships feature an iterative research process
Partnerships? that focuses on developing as well as testing
conjectures; this additional work towards for-
Moving from describing the growth in the sheer mulating and developing theory is also not
number of partnerships that exist today, we next commonly part of the research process in alli-
turn to a brief presentation of some of the models ances (see Barab and Squire 2004 for a more
that have currently been identified in the litera- detailed introduction of design-based research).
ture, and then offer a theoretical exploration into Networked improvement communities: are net-
why we might observe multiple types of RPPs. works of districts and researchers that collabo-
rate on one problem of practice with the goals
Currently, to the best of our knowledge, there of understanding what works best, where, and
is only one study that has attempted a typology of in what context (see Bryk et al. 2011 for a more
the different types of RPPs. A white paper detailed introduction). One of the defining fea-
authored by Coburn et  al. (2013) and tures of this type of RPP is that it involves the
commissioned by the W.  T. Grant Foundation collaboration of many districts to exploit differ-
identifies three different types of RPPs: research ences in contexts in order to improve knowl-
alliances, design-based partnerships, and edge surrounding implementation of programs
networked improvement communities. We briefly and policies. The concept of “improvement sci-
define each of these: ence” is at its core, which is a model adapted
from the healthcare industry. The key example
Research alliances: are partnerships between a of this RPP type is the Carnegie Foundation’s
school district(s) and a research institution(s) Networked Improvement Communities.
such as a university or non-profit. By their
definition, research alliances are long-term This white paper is an excellent first attempt at
commitments where the researchers pursue describing the types of RPPs currently operating
questions of policy and practice that are in education. Note that there are likely several
relevant to both practitioners and researchers other models in operation today that have not
(Coburn et al. 2013). The researchers share the been captured here since the publishing of the
research findings with the district, the paper, with more likely to develop in the future.
community, and other stakeholders and this To help illustrate the myriad ways a partnership

568 P. Arce-Trigatti et al.

Table 25.1  This table presents a simplified illustration of the numerous ways research–practice partnerships can differ
across a multitude of partnership dimensions

Length of partnership commitment
• short, medium, long-term

Researcher side participation
• single or multiple universities, research institutions, or non-profit

organizations

Practitioner side participation
• number of school districts, level of school leader participation

(superintendent/principal/teacher)

Policy side participation
• state, local, school, classroom

Research agendas
• topic specific (narrow focus), collection of topics (broadly defined)

Intensity of collaboration
• Periodic check-ins vs systematic, regular meetings

may be arranged we offer an organized list of s­ pecialize in researching education (i.e., academ-
individual characteristics that can vary among ics, scholars, and generally, those working either
partnerships in Table 25.1. With a greater number within a university or a research institution) and
of RPPs emerging and growth in this field likely those who specialize in administering education
continuing, a more rigorous typology allowing (i.e., practitioners involved at all levels of educa-
for additional nuance across models may be pos- tion, such as teachers, principals, and district or
sible in the future. state leaders). Moving beyond these similarities,
we next explore how different social, political,
25.3.1 A ccording to Theory, What and institutional conditions may give rise to orga-
Might Account for Different nizational heterogeneity across RPPs.
Partnership Models?
One can draw from different disciplines to
It might first be instructive to define how research– explain organizational heterogeneity; this
practice partnerships are similar. They are best approach can shed light on the impact various
considered non-profit organizations, given that aspects of the organizational form has on strategy
their objectives rarely (if ever) focus on maximiz- or production. For example, in economics, the
ing profit: They do not operate in a typical market objectives of the firm can give rise to differences
featuring customers and suppliers, where supply in structural forms, as can the differential costs
and demand determine price, and efficiency can associated with varying production processes
be measured through clear measurement of pro- adopted. Similarly, sociologists also recognize
duction. On the other hand, RPPs also fail to be the importance of organizational goals in shaping
classified as a pure government agency, where structures and strategies, but they additionally
survival is directly linked to satisfying constituent consider how leaders’ backgrounds and cultures
preferences and revenues are generated from a influence the identity of the organization
mandated tax base. Structurally speaking, they (Fligstein and Dauter 2007). These basic con-
commonly include representatives from at least cepts can help us initially understand the visible
two sides of the education realm: those who differences in how RPPs are arranged and the
scope, areas of, or approaches to research they
specialize in. For example, the Houston Education

25  Research–Practice Partnerships in Education 569

Research Consortium (HERC), a research alli- and Freeman (1977) posited that differential
ance that is housed at Rice University and fea- opportunities in the market, in terms of resources
tures the Houston Independent School District as available, directly shaped the birth and survival
its practitioner partner, was founded and is cur- of organizations. Survival is ensured by
rently directed by one of the authors of this work, maintaining good relationships with already-­
Ruth López Turley, who is trained in sociology. known contacts, predictability in meeting funding
The research output produced by HERC will thus targets, and relying on an accepted approach
be framed within the context of sociology and the producing output.
methodologies utilized throughout the projects
will be those commonly found in the field of The availability of resources deserves special
sociology. On the other hand, MIST (mentioned attention in this case. RPPs are strongly dependent
earlier under the design-based partnership model) on the availability of funding; therefore, the
is housed in the College of Education at Vanderbilt number and types of potential funding sources
University. The project’s co-PIs are Erin Henrick will have large ramifications on the birth and
and Paul Cobb, who are both housed within the subsequent survival of RPPs and tracks taken.
Department of Teaching and Learning and are Because there is external control over resources,
trained specifically in education. The scope of the RPPs become interdependent on this environment.
MIST project is thus much more narrow, focus- More generally, what this means is that there is
ing on improving the instructional practices of an element of competition among RPPs that may
math teachers.4 Finally, the networked improve- not exist otherwise. The degree of competitive
ment communities of Carnegie arrive at their pressure for resources will likely vary widely
structural arrangement along a different path across localities; how rural or urban a city is, the
altogether. A specific problem of practice is first number of academic institutions that exist, as
identified and a network then forms consisting of well as the availability of private foundations
a variety of parties interested in working on the serving an area are all examples of how
problem. Organizational goals overall and more competition may be affected. Greater competitive
specifically, the background and training of the pressure for funding may lead to larger differences
leader, can, at least initially, explain some of the in partnerships (i.e., to stand out from the crowd),
differences in RPP models. while less competitive pressure may allow for
imitation of models perceived as successful.
Within organizational theory, we can further Because RPPs are not self-sustaining
identify at least two ways to frame the question of organizations (and will likely never be, given the
organizational heterogeneity, using either absence of a product from which to generate
organizational ecology or institutional theory. We revenues), they are implicitly wedded to the
explore each in turn. foundations that support their work. The
objectives and preferences of the foundations
Organizational or population ecology, an area themselves, then, are likely to have a strong
of research first introduced by Hannan and influence on the probability of birth and survival.
Freeman (1977, 1984), takes the view that the
rational adaptation model popular in economics In addition to the role of resources, we can also
overemphasizes the role of firm adaptation. highlight the general pressures that arise from the
Instead, this approach suggests that the environment as a whole. The number of organiza-
environment in which organizations operate tions that can co-exist in an environment is depen-
presents a fixed constraint; essentially, the dent on the environmental carrying capacity,
environment selects which types of organizations which is itself a function of the social, economic,
survive and which die. To explain the emergence and political conditions and available resources
of heterogeneous organizational forms, Hannan (Anheier 2005). Because many things can affect
the environment’s carrying capacity, this in turn
4 See http://peabody.vanderbilt.edu/departments/tl/teach- will affect the dynamics of organizations over
ing_and_learning_research/mist/ for more information. time. More precisely, the environment will affect

570 P. Arce-Trigatti et al.

how organizations choose to allocate resources; copy others that are perceived to be successful.
this in turn will produce variation across partner- Perhaps one example in the RPP context we can
ship strategies. For example, certain departments highlight is the recent surge in the number of
within universities may be more amenable to the research alliances that resemble the UChicago
notion of a partnership than others. If the sociol- Consortium on School Research. Founded in
ogy department is willing to provide support for 1990, the UChicago Consortium was the only RPP
the creation of a partnership, then the tools of that of its type for approximately 15 years; during that
particular discipline will shape and influence how time, it built a strong reputation among many
the research work is approached within the part- involved in education for producing rigorous, rel-
nership. Furthermore, tenure rules differ across evant, and timely research that has made important
institutions. Qualifying activities, then, could impacts on local decision making. More recently,
either be limited or numerous, and these environ- several RPPs modeled after the UChicago
mental constraints will alter features of the part- Consortium have emerged: the Baltimore
nership. Alternatively, the current research Education Research Consortium (2006), the
capacity and preferences of a school district will Research Alliance for New  York City Schools
also create pathways to some approaches and not (2008), the Los Angeles Education Research
others. Along the practitioner side, there is a Institute (2011), and the Houston Education
greater propensity for leadership turnover, which Research Consortium (2011), just to name a few.
gives rise to environmental instability. Previous
relationships and practices that may have held Normative isomorphism describes the process
promise for a partnership may have to change by which firms change due to external pressures
with immediacy. The infinite combinations of initiated by professions or legitimation directed
these two environmental features could conceiv- by professional practices. These types of forces
ably give rise to multiple types of partnerships. In lead organizations to conform to accepted ways
particular, different environmental characteristics of practice, given a particular profession or even
can help describe the shifts in partnership network of professionals. While new to the RPP
approaches that have occurred more recently (i.e., ecosystem, the National Network of Education
the introduction of design-based research and the Research–Practice Partnerships, a network con-
networked improvement communities). necting several types of RPPs in education, may
eventually influence how individual RPPs emerge
A second way to approach the question of or change over time.
organizational heterogeneity is to use institutional
theory, a research area advanced by the work of Finally, coercive isomorphism relates to the
DiMaggio and Powell (1983), Meyer and Scott changes organizations must undergo due to orga-
(1983), and Meyer and Rowan (1977). In contrast nizational, political, or social pressures of stake-
to the assumption in population ecology of a holders they are dependent upon. RPPs are
fixed environment, this line of thinking particularly susceptible to this type of isomor-
hypothesizes that the environment may be “at phism, given the previously documented reliance
least partially a social construction” (Fligstein upon foundational dollars. Thus, certain models
and Dauter 2007, p. 111). That is, the environment of RPPs may be more or less common simply due
is comprised of other related organizations that to the financial resources they are dependent
could influence the strategic behavior of a upon. For example, several of the design-based
particular firm. DiMaggio and Powell (1983) research partnerships share a STEM-related
argued that in this context, organizations tend to focus; unsurprisingly, the National Science
become more similar over time through three Foundation also funds many of these. Coercive
different isomorphic processes: imitative or isomorphism may also arise from the mere fact
mimetic, normative, and coercive. that RPPs operate in a new space, where research-
ers and practitioners must come together in ser-
Under the first type, imitative or mimetic iso- vice to solving problems of practice. In this case,
morphism describes how organizations imitate or relevant stakeholders include not only the practi-

25  Research–Practice Partnerships in Education 571

tioners themselves, but local decision makers, stu- To close this section, we summarize the dis-
cussion by reinforcing the notion that many vari-
dents, and communities-at large. Indeed, Roderick ables can contribute to organizational
heterogeneity. Furthermore, these differences can
et  al. (2009, p. 2) describe the founding of the arise at any age of the organization, from birth
and over time. More research investigating how
UChicago Consortium as follows: RPPs differentiate is needed, especially to further
our knowledge of what an “effective” RPP may
Given the magnitude of this experiment, the advo- look like.
cates of reform—largely the foundation commu-
nity and local reform organizations—believed it 25.4 W here Does
was important to establish an independent organi- the Development of the Field
zation that would be charged with conducting Currently Stand?
independent, objective evaluations of the progress
of reform and engaging in research that would
assist local schools in developing their own strate-
gies. Because universities seemed like natural part-
ners in this effort, the Chicago Public Schools
(CPS) invited local universities to become
involved.

Furthermore, the authors also illustrate the We have discussed the growth in the number and
importance of satisfying stakeholder needs types of RPPs across the U.S. in the last two
through their work: decades and provided possible reasons for these
trends grounded in organizational theory from
This new role—to provide a research-based frame- sociology. In this next section we widen our focus
work (but not a blueprint) for improvement, to pro- to examine the field of RPPs as a whole and ask:
vide critical measures of performance and feedback Where do we currently stand? If we think of
mechanisms to individual schools, and for RPPs as an “industry,” at what stage in the life-
researchers to engage in the core questions of what cycle do we find ourselves? What can we say
it will take to improve performance—has had a about the development of the field given the rise
significant impact in shaping the work of CCSR of complementary organizations and, most
[the UChicago Consortium] and the role of recently, a formal professional network of educa-
research in the city. CCSR researchers do not just tion RPPs?
comprise an independent group that does studies
on schools and occasionally announces findings. From the larger perspective of the field of
Rather, our studies and products (e.g., individual RPPs, it is likely that this “industry” is still in its
school reports) are resources that practitioners use infancy. The number of RPPs (total) across the
to manage their own improvement efforts. U.S. suggests they are still relatively uncommon
(Roderick et al. 2009, p. 2) among approaches that connect research and
practice.5 Hannan and Carroll (1992) suggest that
Over time, the UChicago Consortium has had the pattern of organizational density over time for
to evolve, and as we might expect, they tie these several industry types follows a regular path:
changes explicitly to stakeholder objectives. long, slow growth in the initial phases, followed
Moreover, they attribute their success specifically by an explosive period of growth, and later,
to this type of change: stabilization or perhaps even decline. Within this
context, RPPs seem to be on the cusp of explosive
Over time, CCSR has evolved into a more complex growth (e.g., Sect. 25.2 of this chapter, which
organization…But key to the success of CCSR has
been a consistent focus on these initial themes: (1) 5 Note: Currently, no resource, such as a directory, exists
research must be closely connected over time to on the number of RPPs currently in operation. The
the core problems facing practitioners and decision NNERPP website (nnerpp.rice.edu) contains a list of part-
makers; (2) making an impact means researchers nerships that are members of its network (which includes
must pay careful attention to the process by which most of the research alliances in operation today), while
people learn, assimilate new information and the R + P Collaboratory website (researchandpractice.org)
ideas, internalize that information, and connect it includes a list of DBIR-type partnerships.
to their own problems of practice; and (3) building
capacity requires that the role of the researcher
must shift from outside expert to interactive
participant in building knowledge of what matters
for students’ success. (Roderick et al. 2009, p. 3)

572 P. Arce-Trigatti et al.

provides an overview of the recent growth in zational form’s processes, structure, and services,
RPPs). Indeed, recent research by Coburn and and sociopolitical legitimation, referring to “the
Penuel (2016, p. 1) on the state of the field process by which key stakeholders, the general
describes RPPs as a “promising approach” that is public, key opinion leaders, or government offi-
currently witnessing an uptick in interest and cials accept a venture as appropriate and right,
funding. Despite a noticeable increase in RPPs as given existing norms and laws.”
an organizational form, they are arguably not yet
a “business as usual” approach. The majority of The inception of new ventures may naturally
states and school districts across the U.S. do not be accompanied by low cognitive legitimacy:
participate in RPPs and they are particularly “Without widespread knowledge and
scarce or nonexistent in more rural areas. understanding of their activity, entrepreneurs
may have difficulty maintaining the support of
25.4.1 W hat Can We Say About the key constituencies” (Aldrich and Fiol 1994,
Development of the Field p. 649). The authors furthermore suggest that in
Given the Rise the absence of developing cognitive legitimacy,
of Complementary especially as it relates to reaching a collective
Organizations? consensus regarding best practices, standards, or
procedures, new entrants into the field risk
Across the organizational theory literature, there possible failure. This could reflect poorly on the
are a few key concepts that can help us better organizational form as a whole, since potential
understand the current state of the RPP industry funders or future RPP leaders will be watching
and where it might be headed next. In an closely to see how individual organizations
industry’s infancy, new organizations must perform. This conceptual framework can help
develop several innovations—not just the organi- provide some grounding to explain the recent
zational structure itself or the process of work— emergence of several peripheral efforts related to
but also new workplace roles, without having RPPs, which we will call “complementary
much prior knowledge to build off of and within organizations.” We define these organizations to
a larger context that is not quite yet accepting of be those that support the work of RPPs in some
these ventures (Hannan and Carroll 1992; way, but are not in and of themselves partnerships.
Stinchcombe 1965). These early challenges may
partially explain why the beginning stage of a With respect to cognitive legitimation, we
new industry is characterized by a long, slow argue that complementary organizations work to
build: Moving from innovative, developmental advance the collective knowledge of RPPs that
production phases to systematic, efficient may indeed contribute towards creating
processes takes time, while the external conditions where partnerships are more likely to
environment in which the organization operates become permanent fixtures in the educational
may provide additional barriers to acceptance of arena. First, the number of individual research
new norms. To ensure survival, institutional studies on RPPs has exploded in the last two
theorists have long suggested the legitimacy of decades.6 This may partly be due to the simple
the new organizational form must be expanded fact that there are more RPPs today relative to
and directly addressed (e.g., Meyer and Rowan 20  years ago, but it could also be argued that
1977; Meyer and Scott 1983; DiMaggio and those working within RPPs are eager to produce
Powell 1991). More recently, Aldrich and Fiol knowledge that helps support their new venture.
(1994, p. 648) distinguish among two types of Second, two new IES-funded research centers
legitimacy, especially salient to entrepreneurs:
cognitive legitimation, describing the knowledge 6 Conducting a simple Google Scholar search on “research
building that must occur around the new organi- practice partnerships” + education and restricting the
results to the years 1960 through 1989 returns zero results.
When changing the yearly range from 1990 to 2000, ten
results are listed. Finally, adjusting the yearly range once
more, from 2001 to 2016, nearly 300 articles are returned.

25  Research–Practice Partnerships in Education 573

focusing on understanding the connections Members of these different organizational forms
between researchers, practitioners, and may often not be fully aware of the extensive dis-
policymakers have recently emerged. The similarities in terms of timelines, communication
National Center for Research in Policy and processes, and internal working structures, to
Practice (NCRPP), housed at the University of name a few examples. Given these potential bar-
Colorado, Boulder and the Center for Research riers to success, NNERPP has made one of its
Use in Education (CRUE) at the University of objectives to systematically collect, develop, and
Delaware, are likely to increase knowledge share best practices from a variety of RPP mod-
around RPP work into the next decade.7 Third, els. This is directly in line with raising the cogni-
three additional resources exist to help develop tive legitimacy of the approach, which may be
and support those interested in partnership work. especially salient at this stage in the industry’s
The R  +  P Collaboratory8 at the University of development. We hypothesize that collectively,
Colorado, Boulder, is an organization that helps these complementary organizations are likely to
support STEM-related work within RPPs as well directly impact the cognitive legitimacy of the
as DBIR-type partnerships, while the William field overall and will more than likely make it
T. Grant Foundation has organized a micro-site9 easier for new entrants to emerge and develop,
of RPP-related information and materials. given the relatively larger pool of knowledge they
will be able to draw from.
The third resource and most recent entrant
into this group of complementary organizations Commenting on the state of the field with
and the one most intimately known to the authors respect to sociopolitical legitimation is somewhat
is the National Network of Education Research– more challenging. In terms of measurement,
Practice Partnerships (NNERPP), which aims to Aldrich and Fiol (1994, p. 648) suggest evaluat-
construct a connected web of education RPPs ing the degree of this type of legitimacy by
across the country to support and develop RPPs.10 “assessing public acceptance of an industry, gov-
As we will focus on the “network” aspect of this ernment subsidies to the industry, or the public
organization shortly, in this section we highlight prestige of its leader.” It is likely that sociopoliti-
its role in expanding cognitive legitimacy. cal legitimation is still growing among stakehold-
Because RPPs require many skills for which ers. Of the scant evidence we can point to that
education researchers, education agency leaders, suggests this may indeed be occurring, we note
and decisionmakers are typically not trained, the increase in opportunities for funding. For
these collaborations tend to be challenging to set example, as mentioned previously, the Spencer
up and maintain. Although researchers often Foundation launched their first ever competitive
collaborate with other researchers, it is less com- RPP grants award in 2015 while IES has created
mon for them to collaborate with education agen- new initiatives to fund RPPs, in addition to a
cies in long-term partnerships, as noted earlier. reorganization of the RELs towards a greater
Substantial organizational differences between RPP orientation. The founding of NNERPP itself
research institutions and education agencies can also lends support to the idea that RPPs are gain-
lead to a prohibitive working environment. ing sociopolitical legitimacy, especially if we
consider that it is financially resourced by five
7 Given IES’ role in supporting the Regional Education different private foundations. From the govern-
Laboratories, these two centers should come as no sur- mental perspective, we note the increased demand
prise, lending support for the notion that advancing cogni- from policies mandating greater use of evidence-
tive legitimation matters. based research. While not explicitly directed at
RPPs, the shift towards connecting research and
8 See http://researchandpractice.org/ for more practice could arguably be a form of sociopoliti-
information. cal legitimation. Generally speaking, however,
we might expect sociopolitical legitimacy to be
9 See http://rpp.wtgrantfoundation.org/ for more
information.

10 See http://nnerpp.rice.edu for more information.

574 P. Arce-Trigatti et al.

positively impacted as cognitive legitimacy sur- and consider how a professional network of RPPs,
rounding RPPs increases. such as NNERPP, might matter for individual
RPP behavior or performance.
We next turn our discussion to the “network”
aspect of NNERPP: Why does the development 25.4.2 W hat Do Networks Provide?
of a professional network of RPPs merit attention
here? What does it suggest about the state of the First and foremost, networks establish a clear
field overall or where it might be headed? While mechanism through which member organizations
the research in the previous part of Sect. 25.3 is can access a wide range of resources (Burt 1992;
more connected to organizational sociology (i.e., Gulati et al. 2011; Smith-Doerr and Powell 2005)
institutions), this next subsection relates more as well as provide order to an otherwise discon-
closely to economic sociology (i.e., networks). nected collection of related organizations (Burt
They are often two distinct research areas but 2000). Smith-Doerr and Powell (2005, p. 16)
share connections, as we will see. By shifting the suggest “organizations forge connections to other
lens slightly, we hope to further our understanding parties to access relevant expertise. Access to
of the important roles different actors play within centers of knowledge production is essential
the RPP ecosystem overall. when knowledge is developing at an unprece-
dented pace.” While knowledge about the inter-
We begin our discussion by exploring the defi- nal workings of an RPP is developing, it is not
nition of a “network.” Podolny and Page (1998, p. necessarily developing rapidly or systematically
59) broadly define a network as “any collection of but haphazardly. The UChicago Consortium is
actors (N ≥ 2) that pursue repeated, enduring one example of an RPP that has written about
exchange relations with one another and at the their founding (cf. Roderick et al. 2009) while the
same time, lack a legitimate organizational R + P Collaboratory, the W. T. Grant Foundation
authority to arbitrate and resolve disputes that RPP microsite, and NNERPP have created or
may arise during the exchange.” These authors made available various toolkits to help those
distinguish between a market, where exchanges interested in pursuing this work. Several addi-
are not necessarily enduring but instead, “epi- tional resources not mentioned here exist across
sodic,” and hierarchies, where there is a clear various other websites, but are less known. To
order to authority, especially regarding the resolu- access these knowledge centers, however, those
tion of disputes. Other authors have been more interested in launching an RPP would first need
explicit, defining a network as a collection of to know where to find them and second, may find
actors or nodes (in our case, RPPs) that are con- that the resources, while helpful in their own
nected by specific ties (Borgatti and Halgin 2011; right, are not quite sufficient. Indeed, the authors
Smith-Doerr and Powell 2005). In these cases, have often fielded phone calls, in-person meet-
ties among nodes are typically descriptive of the ings, and online video chats from interested par-
relationship between two nodes; for example, in ties seeking “relevant expertise,” as Smith-Doerr
NNERPP’s case, the ties may represent a collegial and Powell describe. Thus, while other, more
relationship among RPPs. Research in this area static resources are available, the dynamic nature
has focused on characterizing the structural of interacting with others may be quite difficult to
aspects of a network (e.g., Burt 1992), while oth- replace. The network itself becomes a centralized
ers have prioritized an analysis of interorganiza- hub, then, that facilitates an arguably more effi-
tional connections and their potential effect on cient distribution of information and knowledge
organizational behavior (e.g., Granovetter 1985). than individual organizations working alone.
A third perspective moves away from previous This is one reason why we might expect NNERPP
assumptions that organizations within a network to move the field forward more quickly than an
are essentially uninvolved and instead, examines RPP ecosystem without it.
how organizations actively rely on networks as a
wellspring of resources (e.g., Gulati et al. 2011).
In this subsection, we adopt the third framework,

25  Research–Practice Partnerships in Education 575

Second, in addition to its power of dissemina- research and policy for district/state improvement

tion, other research points to the role networks play shows evidence that research produced by RPPs is

in supporting innovation (Bryk et  al. 2011; likely to be more beneficial than research pro-

Goldsmith and Eggers 2004; Podolny and Page duced outside of RPPs, not only because research-

1998; Powell 1990). Smith-Doerr and Powell ers are more likely to produce work that is aligned

(2005, p. 17) go so far as to characterize the poten- with district needs but also because district lead-

tial for networks to become a “locus of innovation” ers are more likely to view the research as credible

due to the fostering of meaningful relationships and directly applicable to their context (e.g.,

across member organizations that goes beyond a Coburn et  al. 2009; Honig and Venkateswaran

simple knowledge exchange. Furthermore, Smith- 2012). However, there is mixed evidence that dis-

Doerr and Powell (2005, p. 25) posit that “[m]uch tricts engaging in RPPs use research in decision

research has suggested that close interaction among making more consistently than districts not

divergent organizations can produce novel recom- engaging in RPPs, and one possible explanation is

binations of information leading to greater innova- that it is difficult for researchers and district lead-

tion and learning (Cohen and Levinthal 1990; ers to learn from one another (Turley and Stevens

Powell 1990; March 1991; McEvily and Zaheer 2015). Coburn et  al. (2013, p. 25) conclude:

1999; Stuart and Podolny 1999; Ahuja 2000).” In “What is needed is a more robust dialogue in

the present case, there are “divergent” organiza- which district leaders, researchers, policymakers,

tions along two lines: First, within each individual and funders speak candidly about the strategic

RPP, there are at least two different institutions trade-offs partnerships face and the resources that

involved (i.e., university and school district), and are required for success.” By organizing these

often times, more.11 Thus, each individual RPP is syntheses through a network, greater diffusion of

essentially a mini-network of its own. The close knowledge and ideas that may then spur innova-

proximity within which each institution works tive solutions to current problems of practice may

together because of the partnership commitment is be possible.

very promising for the potential to produce innova- To close this section, we circle back to the

tions. Second, NNERPP itself consists of a collec- question of why network analysis may be relevant

tion of RPPs that differ in terms of arrangements, to the study of RPPs specifically, and to

geographical location, age, size, research organizational forms, generally. Owen-Smith and

approaches, and breadth of topics analyzed. Powell (2008, p. 600) suggest that “[n]etworks

NNERPP has further indicated that two of its are essential to fields in at least two senses: they

priorities include the facilitation of cross-­ are both a circulatory system and a mechanism

partnership collaboration and second, the synthe- for sensemaking. Fields are shaped by networks,

sis of research findings produced by RPPs and the which condition the formation of relationships

building of new knowledge based on RPP and help establish their consequences.”

research. Education leaders and researchers alike Furthermore, the authors also write that “[w]hile

can benefit from other partnerships’ research institutions shape structures and condition their

practices and findings. Research produced by effects, networks generate the categories and

RPPs can and should be synthesized in a manner hierarchies that help define institutions and con-

that enables researchers and policymakers from tribute to their efficacy. Thus, any effort to under-

all over the country to strategically build on that stand institutional processes must take networks

knowledge and use it to develop novel solutions to into account, and vice versa” (Owen-S­ mith and

persistent problems of practice. An emerging field Powell 2008, p. 594). Additionally, complemen-

of research that studies the relationship between tary partnerships may forge a path for further net-

work establishment. It is possible that networks

11 For example, some RPPs also partner with community may feed back into the lifecycle process of RPPs
non-profit organizations or non-university research and may help further establish normative culture
institutions. around RPPs.

576 P. Arce-Trigatti et al.

25.5 C onclusion an RPP “effective.”12 Because several different
models of RPPs exist (with greater variety in
We have seen growth both in the number and type structural arrangements likely occurring over
of research–practice partnerships (RPPs) in edu- time), this also adds complexity to the issue.
cation over the last two decades, as well as the Should all be judged equally? The literature on
emergence of complementary organizations and RPP failure is equally sparse. What conditions
even the launch of a professional network of lead an RPP to fail or close, for example? The
RPPs, all suggesting that the RPP model is gain- next stage of the field will require a more explicit
ing traction as a potentially useful way to connect definition of organizational performance.
research, policy, and practice in education. We
have explored the reasons for these changes using Beyond constructing an accepted definition of
many organizational and institutional theories success/failure, the interim process of how RPPs
found in sociology and what they might mean for evolve over time is not well known, either. For
the future of RPPs. We framed our analysis across example, how does organizational change occur
multiple levels: At the firm-level, we provided a or what leads to organizational change? There is
historical foundation to explain the rise of RPPs also, of course, the possibility that RPPs change
and additionally gave a current description of the very little over time. Because there are typically at
variety of RPP models in existence. At the indus- least two distinct types of institutions that come
try-level, we have explored how organizations together to form a research–practice partnership,
that are not themselves RPPs are situated within there are internal and external pressures affecting
the industry and how they may complement the multiple units within the partnership, which could
work of partnerships and more broadly, the field. individually and collectively lead to very different
Given limitations in space and scope of work, we types of changes over time. Analysis of this kind
aimed to provide the reader foundational knowl- is not straightforward. For those interested in
edge from which one can begin to think more implementing continuous improvement processes
deeply about the evolution of research–practice as they relate to RPP performance, what types of
partnerships and the promises they hold for the organizational policies would be most appropri-
future in education. In this final section, we leave ate? Outside of the institutional forms that make
the reader with several unanswered questions that up the RPP, there is also the larger external envi-
will require further analysis and consideration in ronment to consider. What political contexts or
the coming years, and will likely affect the contin- conditions are important for fostering future
ued growth and existence of this organizational growth of individual RPPs and the field as a
form. whole? Addressing these questions with rigorous
research will likely be important for the overall
First, defining the conditions that constitute survival of this organizational form.
“best practices” for an RPP is still very much in
development. Feedback loops are an essential Finally, it is important to note the dual roles
component to learning more precisely about that collaboration and competition between RPPs
“what works,” but these have been sparse for a can play with respect to individual organizational
couple of reasons. From the perspective of
innovation, multiple cycles of success and failure 12 For example, some have argued that the UChicago
(e.g., closure of the organization) have not yet Consortium has been a model for RPP success. It is not
occurred in this industry, mostly due to the clear if this accolade refers to its longevity within the
relative newness of the organizational form. industry or due to the strong reputation it has developed
Second, and perhaps more importantly, there is over time in being an exemplar for how RPPs can work, or
currently no consensus about how to define RPP other aspects of the partnership. Although these features
“success” or the features or outcomes that make may be indicators of success, it should be noted that our
general knowledge of RPP effectiveness is still in its
infancy.

25  Research–Practice Partnerships in Education 577

health and to the larger field. Is it possible for Barab, S., & Squire, K. (2004). Design-based research:
RPPs to continue to learn from one another and Putting a stake in the ground. Journal of the Learning
remain in a relatively collaborative space? Or Sciences, 13(1), 1–14.
will increased competition for funds inhibit the
type of knowledge sharing that could provide Berman, E.  P. (2012). Explaining the move toward the
beneficial growth to the field? Hannan and Carroll market in U.S. academic science: How institutional
(1992, p. 13) suggest that generally speaking, logics can change without institutional entrepreneurs.
there are “limits to the longevity of firms.” One Theory and Society, 41(3), 261–299.
possible explanation for this constraint has to do
with the tension between legitimation and com- Berman, E.  P., & Stivers, A. (2016). Student loans as a
petition: To survive initially and to ensure field pressure on U.S. higher education. In Berman &
growth, organizational forms must address legiti- Paradeise, (Eds.), The university under pressure
mation. This often leads to collaboration since (Research in the Sociology of Organizations) (Vol. 46,
systematic knowledge collection around the pp. 129–160).
organization and diffusion of this information is
particularly useful for raising legitimation. As a Birkeland, S., Murphy-Graham, E., & Weiss, C. (2005).
larger number of organizations emerge, however, Good reasons for ignoring good evaluation: The
competition for a variety of limited resources case of the drug abuse resistance education (DARE)
(e.g., financial and human capital related) places program. Evaluation and Program Planning, 28(3),
greater pressure on the survival of any given 247–256.
organization. This may preclude some organiza-
tions that may otherwise have supported others Borgatti, S. P., & Halgin, D. S. (2011). On network theory.
from sharing practices or knowledge. Organization Science, Articles in Advance, 1–14.

What, then, can we conclude about the RPP Bryk, A.  Gomez, L., & Grunow, A. (2011). Getting
landscape given these important issues? Given the ideas into action: Building networked improvement
relatively young age of the industry overall, some communities in education. In Maureen T.  Hallinan
of these questions will require more time in order (Ed.), Frontiers in sociology of education. Dordrecht:
to be adequately addressed. Rigorous research Springer.
examining RPPs of all types is just now com-
mencing (for example, the two IES-funded Burch, P. E., & Thiem, C. (2004). Private organizations,
research centers mentioned in Sect. 25.3). In gen-
eral, because the interest and momentum in RPPs school districts, and the enterprise of high stakes
as a mechanism for connecting research, policy, accountability. Unpublished manuscript.
and practice is currently in an upward trend, we Burt, R. (1992). Structural holes. Cambridge, MA:
are optimistic that the creation of knowledge Harvard University Press.
around these approaches is likely to grow. Burt, R. (2000). The network structure of social capital.
Research in Organizational Behavior, 22, 345–423.
References Coburn, C.  E., & Penuel, W.  R. (2016). Research–prac-
tice partnerships in education: Outcomes, dynamics,
Ahuja, G. (2000). Collaboration networks, struc- and open questions. Educational Researcher, 45(1),
tural holes, and innovation: A longitudinal study. 48–54.
Administrative Science Quarterly, 45, 425–455. Coburn, C., & Stein, M.  K. (2010). Research and prac-

Aldrich, H. E., & Fiol, C. M. (1994). Fools rush in? The tice in education: Building alliances, bridging the
institutional context of industry creation. The Academy divide. Blue Ridge Summit: Rowman & Littlefield
of Management Review, 19(4), 645–670. Publishers, Inc.
Coburn, C., & Talbert, J. E. (2006). Conceptions of evi-
Anheier, H. (2005). Nonprofit organizations: Theory, dence use in school districts: Mapping the terrain.
management, policy. London/New York: Routledge. American Journal of Education, 112(4), 469–495.
Coburn, C.  E., Honig, M.  I., & Stein, M.  K. (2009).
What’s the evidence on districts’ use of evidence? In
J. D. Bransford, D. J. Stipek, N. J. Vye, L. M. Gomez,
& D. Lam (Eds.), The role of research in educational
improvement (pp.  67–87). Cambridge, MA: Harvard
Education Press.
Coburn, C., Penuel, W.  R., & Geil, K.  E. (2013).

Research–­practice partnerships: A strategy for

leveraging research for educational improvements
in school districts. Commissioned by the William
T. Grant Foundation.
Cohen, W., & Levinthal, D. (1990). Absorptive capac-
ity: A new perspective on learning and innovation.
Administrative Science Quarterly, 35, 128–152.
Corcoran, T., Fuhrman, S.  H., & Belcher, C.  L. (2001).
The district role in instructional improvement. The Phi
Delta Kappan, 83(1), 78–84.

578 P. Arce-Trigatti et al.

Datnow, A., Hubbard, L., & Mehan, H. (2002). Extending policy and research agenda. Educational Policy, 22(4),
educational reform: From one school to many. 578–608.
London/New York: Routledge/Falmer. Honig, M. I., & Venkateswaran, N. (2012). School–central
office relationships in evidence use: Understanding
DiMaggio, P.  J., & Powell, W.  W. (1983). The iron evidence use as a systems problem. American Journal
cage revisited: Institutional isomorphism and col- of Education, 118, 199–222.
lective rationality in organizational fields. American Hood, P. D. (1982). The role of linking agents in school
Sociological Review, 48, 147–160.
improvement: A review, analysis, and synthesis of
Every Student Succeeds Act, Public Law 95, 114th Cong., recent major studies. Far West Lab for Educational
1st Sess. (2015). Research and Development.
Huberman, M. (1989). Predicting conceptual effects
Feuer, M.  J., Towne, L., & Shavelson, R.  J. (2002). in research utilization: Looking with both eyes.
Scientific culture and educational research. Knowledge in Society, 2(3), 6–24.
Educational Researcher, 31(8), 4–14. Huberman, M. (1994). Research utilization: The state
of the art. Knowledge and Policy: The International
Fleming, D. S. (1988). The literature on teacher utilization Journal of Knowledge Transfer and Utilization, 7(4),
of research: Implications for the school reform move- 13–33.
ment. In S.  D. Castle (Ed.), Teacher empowerment Kerr, K. A., Marsh, J. A., Ikemoto, G. S., Darilek, H., &
Barney, H. (2005). Districtwide strategies to promote
through knowledge linking research and practice for data use for instructional improvement. Paper presented
school reform. Papers presented at the annual meeting at the annual meeting of the American Educational
of the American Educational Research Association, Research Association, Montreal, QC, Canada.
New Orleans, April 5–9. Klein, A. (2016). Under ESSA, states, districts to share
Fligstein, N., & Dauter, L. (2007). The sociology of mar- more power. Education Week, January 5, 2016.
kets. Annual Review of Sociology, 33, 105–128. Kraatz, M. S., & Zajac, E. J. (2001). How organizational
Foster, J.  G., Rzhetsky, A., & Evans, J.  A. (2015). resources affect strategic change and performance
Tradition and innovation in scientists’ research strate- in turbulent environments: Theory and evidence.
gies. American Sociological Review, 80(5), 875–908. Organization Science, 12(5), 632–657.
Goldsmith, S., & Eggers, W. D. (2004). Governing by net- Kramer, R. (1981). Voluntary agencies in the welfare
work: The new shape of the public sector. Washington, state. Berkley: University of California Press.
DC: Brookings Institution Press. March, J. G. (1991). Exploration and exploitation in orga-
Gonzalez, N., Moll, L. C., & Amanti, C. (Eds.). (2005). nizational learning. Organization Science, 2, 71–87.
Massell, D. (2001). The theory and practice of using data
Funds of knowledge: Theorizing practices in house- to build capacity: State and local strategies and their
holds, communities and classrooms. Princeton: effects. In S.  H. Fuhrman (Ed.), From the capitol to
Lawrence Erlbaum Associates.
Granovetter, M. (1985). Economic action and social the classroom: Standards-based reform in the states.
structure: The problem of embeddedness. American
Journal of Sociology, 91, 481–510. One hundredth yearbook of the national society for the
Gulati, R., Lavie, D., & Madhavan, R. (2011). How do study of education (pp. 148–169). Chicago: National
networks matter? The performance effects of inter- Society for the Study of Education.
organizational networks. Research in Organizational Massell, D., & Goertz, M. E. (2002). District strategies for
Behavior, 31, 207–224. building instructional capacity. In A.  M. Hightower,
Guthrie, J. W. (1989). Educational laboratories: History M.  S. Knapp, J.  A. Marsh, & M.  W. McLaughlin
and prospect. Commissioned by the United States (Eds.), School districts and instructional renewal
Department of Education. (pp. 43–60). New York: Teachers College Press.
Hannan, M.  T., & Carroll, G.  R. (1992). Dynamics of McEvily, W.  J., & Zaheer, A. (1999). Bridging ties:
A source of firm heterogeneity in competitive
organizational populations: Density, legitimation, and capabilities. Strategic Management Journal, 20,
competition. New York: Oxford University Press. 1133–1156.
Hannan, M., & Freeman, J. (1977). The population ecol- Meyer, J. W., & Rowan, B. (1977). Institutionalized orga-
ogy of organizations. American Journal of Sociology, nizations: Formal structure as myth and ceremony.
82(5), 929–964. American Journal of Sociology, 83, 340–363.
Hannan, M., & Freeman, J. (1984). Structural inertia and Meyer, J. W., & Scott, R. (1983). Organizational environ-
organizational change. American Sociological Review, ments: Ritual and rationality. Beverly Hills: Sage.
49, 149–164. National Research Council. (2003). Strategic education
Honig, M.  I. (2003). Building policy from practice: research partnership. Washington, DC: The National
Central office administrators’ roles and capacity in Academies Press.
collaborative policy implementation. Educational No Child Left Behind Act of 2001, Public Law 107-110,
Administration Quarterly, 39(3), 292–338. 20 U.S.C. (2002).
Honig, M.  I. (2006). Street-level bureaucracy revisited: Owen-Smith, J., & Powell, W. (2008). Networks and
District central office administrators as boundary span- institutions. In R. Greenwood, C. Oliver, & R.
ners in complex policy implementation. Educational
Evaluation and Policy Analysis, 28, 357–383.
Honig, M. I., & Coburn, C. (2008). Evidence-based deci-
sion making in school district central offices: Toward a

25  Research–Practice Partnerships in Education 579

Suddaby (Eds.), The SAGE handbook of organiza- http://blogs.edweek.org/edweek/inside-schoolre-
tional institutionalism (pp. 596–623). London: SAGE search/2016/12/new_regional_research_labs_exp.html
Publications. Spillane, J. P. (2000). Cognition and policy implementa-
Podolny, J. M., & Page, K. L. (1998). Network forms of tion: District policymakers and the reform of math-
organization. Annual Review of Sociology, 24, 54–76. ematics education. Cognition and Instruction, 18(2),
Powell, W.  W. (1990). Neither market nor hierar- 141–179.
chy: Network forms of organization. Research in Spillane, J. P., Reiser, B. J., & Reimer, T. (2002). Policy
Organizational Behavior, 12, 295–336. implementation and cognition: Reframing and refocus-
Powell, W. W., & DiMaggio, P. J. (1991). The new insti- ing implementation research. Review of Educational
tutionalism in organizational analysis. Chicago: Research, 72, 387–431.
University of Chicago Press. Stinchcombe, A. (1965). Social structure and organiza-
Reichardt, R.  E. (2000). The state’s role in supporting tions. In J. G. March (Ed.), Handbook of organizations
data-driven decision-making: A view of Wyoming. (pp. 142–193). Chicago: Rand McNally.
Aurora: Mid-Continent Research for Education and Strauss, V. (2015). The successor to No Child Left Behind
Learning. has, it turns out, big problems of its own. Washington
Roderick, M., Easton, J.  Q., & Sebring, P.  B. (2009). Post, December 7, 2015.
Stuart, T. E., & Podolny, J. M. (1999). Positional conse-
CCSR: A new model for the role of research in sup- quences of strategic alliances in the semiconductor
porting urban school reform. UChicago Consortium industry. In S. Andrews & D. Knoke (Eds.), Networks
on School Research Report. Accessed from https:// in and around organizations (Vol. 16 of Research
consortium.uchicago.edu/publications/ccsr-new- in the Sociology of Organizations) (pp.  161–182).
model-role-research-supporting-urban-school-reform Greenwich: JAI Press.
Rowan, B. (1982). Organizational structure and the insti- Tolbert, P., & Zucker, L. (1983). Institutional source of
tutional environment: The case of public schools. change in organizational structure: The diffusion
Administrative Science Quarterly, 27(2), 259–279. of civil service reform, 1880–1935. Administrative
Scott, W. R., & Biag, M. (2016). The changing ecology Science Quarterly, 28, 22–39.
of U.S. higher education: An organizational field Turley, R., & Stevens, C. (2015). Lessons from a
perspective. In E. P. Berman & C. Paradeise (Eds.), school district–university research partnership:
University under pressure (Vol. 46 of Research in the The Houston Education Research Consortium.
Sociology of Organizations) (pp.  25–51). Bingley:
Emerald Group Publishing Limited. Educational Evaluation and Policy Analysis,
Smith-Doerr, L., & Powell, W. (2005). Networks and eco- 37(1S), 6S–15S.
nomic life. In N. J. Smelser & R. Swedberg (Eds.), The Tyack, D.  B. (1974). The one best system: A history of
handbook of economic sociology (2nd ed., pp.  379– American urban education. Cambridge, MA: Harvard
402). New  York/Princeton: Russell Sage Foundation University Press.
& Princeton University Press. Weiss, C. (1980). Knowledge creep and decision accre-
Sparks, S. D. (2016, December 23). New Regional tion. Knowledge, 1(3), 381–404.
Education Research Labs expand partnerships Zucker, L. (1987). Institutional theories of organization.
for ESSA support. Blog post. Retrieved from: Annual Review of Sociology, 13, 443–464.

Author Index

A Barnett, B.G., 517
Abdulkdiroglu, A., 223 Barnett, E.A., 375, 549
Adelman, C., 133, 197, 198, 209, 267, 366, 370, 416, Barr, A.B., 236
Barrera, D.B., 177
418, 432, 536, 548 Bastedo, M., 409
AERA, 415 Baum, S., 119, 185, 458, 537, 539
Ahearn, C., 431–452 Bean, F.D., 110, 119, 154, 156
Ainsworth-Darnell, J.W., 23, 30, 92, 122, 139, 330 Beattie, I.R., 171–188, 461
Alba, R.D., 154, 411 Becker, K.I., 351, 447
Alexander, K.L, 42, 46, 60–63, 65, 110, 111, 121, 254, Bednarek, A., 177, 186
Belfield, C., 419, 434, 539, 542, 552, 553
256, 259, 363, 414, 432, 515, 517 Benson, J., 173, 175, 408
Allen, D., 422, 487, 547 Berends, M., 221–243, 259, 260
Allison, K.W., 178 Bersudskaya, V., 118, 407
Alon, S., 117, 118, 141, 342, 463, 542 Besen-Cassino, Y., 351
Altenhofen, S., 48, 237, 239 Bettinger, E.P., 223, 224, 229, 375–377, 393, 415, 416,
Altonji, J.G., 225, 226, 242, 267, 370
Alvarez, G.A.S., 182 542, 548
An, B.P., 422 Betts, J.R., 222, 227–229, 267, 284, 370, 372, 539
Andreas, R.E., 178 Bianchi, S.M., 352, 353
Andrews, R., 418, 438 Bifulco, R., 229, 235, 236
Angrist, J.D., 33, 222, 225, 227, 241, 242, 542, 543 Billingham, C.M., 237, 238
Apel, R., 341, 345, 348, 350 Birani, A., 183, 188
Aries, E., 180, 186, 392, 393 Bishop, J.H., 267
Armstrong, E.A., 174, 175, 180, 181, 185, 187, Black, S.E., 345, 346, 364
Blau, P.M., 73, 173, 175, 304, 310, 434, 437, 441,
188, 386, 389, 392, 395–397, 399, 435,
437, 447 458, 494–496
Aronson, J., 101, 102, 136, 143, 146, 395 Blumer, H., 540
Aronson, P.J., 350 Boatman, A., 415, 548
Arum, R., 123, 143, 174–176, 185, 186, 228, 229, 266, Bodovski, K., 3, 13, 23–27
267, 318, 385–399, 444, 451, 465 Booker, K., 225, 229
Astin, A., 182, 391, 393, 537 Borman, G., 56
Attewell, P., 332, 336, 370, 410, 415, 416, 433, 539, 546, Bound, J., 375, 407, 464
548, 549 Bourdieu, P., 3–5, 8–10, 12, 14, 15, 19, 33, 45, 56, 91,
Austin, M., 221–243
Autor, D.H., 254, 386 98, 138, 174, 179, 182, 287, 362, 374, 375, 391
Bowles, S., 5, 7, 18, 19, 56, 89, 90, 99, 103, 138, 174,
B
Bachman, J.G., 78, 340–343, 346, 348, 351 252, 362
Badway, N., 387 Bozick, R., 133, 141, 352, 371, 376, 392
Bahr, P.R., 393, 416, 424, 443, 537, 548 Bradley, C.L., 140
Bailey, M.J., 364, 365, 376, 387, 423 Bradley, K., 31
Bailey, T.R, 137, 376, 408, 432, 434, 437, 438, 444, Bragg, D.D., 419, 422
Brame, R., 341, 345, 348, 350
536–539, 548 Brawer, F.B., 406
Barling, J., 342 Bremner, R.H., 339, 341
Brint, S., 376, 388, 393, 409, 414, 425, 536, 540

© Springer International Publishing AG, part of Springer Nature 2018 581
B. Schneider (ed.), Handbook of the Sociology of Education in the 21st Century, Handbooks
of Sociology and Social Research, https://doi.org/10.1007/978-3-319-76694-2

582 Author Index

Broda, M., 343, 350, 354 Cohen, J., 329
Brody, C., 341 Cohen, W., 575
Broh, B.A., 58, 60, 61, 63–65 Colbeck, C.L., 422
Brown, I., 140, 142, 434 Coleman, J., 4, 8, 19, 33
Brown, J., 266 Coleman, J.S., 40, 44, 45, 56, 57, 99, 182, 224, 225, 241,
Brown, S.K., 161
Bruch, S., 143, 144 242, 256, 259, 279, 284, 287, 289, 298, 306,
Bryk, A.S., 45, 224, 225, 242, 259, 286, 287, 291, 301, 320–322, 339, 341, 366, 463, 515, 516, 522
Collier, P.J., 180, 392
306, 307, 309, 319, 487, 495, 497, 499, 504, 505, Collins, P.H., 184, 398
517, 567, 575 Condron, D.J., 56, 60, 63, 67, 120
Buchmann, C., 67, 117, 132–134, 136, 142, 143, 174, Conger, D., 136, 154, 258
319, 364, 387, 395, 398, 406, 432, 434 Conger, R.D., 346
Buckley, J., 47, 226, 227, 238, 239 Conwell, J., 144
Buddin, R., 229 Costello, C., 175
Bugarin, R., 172, 173, 178 Covarrubias, R., 186
Burchinal, M., 22 Covay, E., 224, 242
Burdick-Will, J., 232, 235, 236, 240, 241 Cowen, J., 223, 226
Burris, C.C., 260 Crosnoe, R., 43, 49, 50, 119, 120, 197, 317–333, 517
Bush, G., 477 Crosta, P.M., 376, 537
Bush, V., 254 Crowley, M., 257
Bushway, S.D., 345, 348, 350 Cuccaro-Alamin, S., 172, 177
Butler, J.S., 227 Cullinane, J., 419
Butler, S.M., 420, 458, 464, 465, 469
D
C Dadgar, M., 422
Cahalan, M., 407 D’Amico, R., 339–341
Calcagno, J.C., 415, 439, 548 Dauber, S.L., 121, 256
Call, K.T., 341 Davis, J., 172
Callahan, R.E., 307 Davis, J.B., 339, 341
Callahan, R.M., 121, 261, 262 Davis, L., 341, 343
Campbell, M., 146 Deckman, S.L., 186
Carbonaro, W., 224, 225, 242, 319 Deil-Amen, R.J., 117, 173, 187, 366, 376, 410, 414–416,
Carr, R., 341
Carroll, C.D., 435, 439 423, 435, 436, 467, 537, 539
Carroll, G.R., 571, 572, 577 DeLuca, S., 123, 237, 241, 366, 374, 376, 435, 439, 440,
Carroll, J.M., 205, 251–268
Carroll, L., 205 444, 447
Carter, D.F., 178, 394 Deming, D., 368, 407, 408, 436, 438, 439, 446, 459,
Carter, P.L., 30, 93, 122, 147, 178, 181, 329, 330
Castleman, B., 377 483, 542
Castleman, B.L., 377, 542, 544, 545 Denice, P., 234, 237, 240, 241
Catsambis, S., 133, 260 Denning, J.T., 412, 413, 421, 542
Chakrabarti, R., 229, 230, 481 Destin, M., 186
Chambliss, D.F., 183 Dewey, J., 252
Charles, C.Z., 117, 136, 388, 393, 395, 398 Dillow, S.A., 134, 368, 406, 407
Charles, M., 136 DiPrete, T.A., 67, 133–136, 145, 174, 283, 364, 369,
Cheadle, J.E., 25, 27, 139, 327
Chingos, M.M., 226, 521 387, 395, 398, 406, 432, 434, 463, 518
Cho, S., 310 Dobbie, W., 57, 222, 225, 241, 242
Cho, S.-W., 424 Domina, T., 263–266, 332, 361, 363, 366, 370,
Choy, S., 172, 173, 178, 184, 439, 447
Clark, B.R., 409, 410, 414, 425, 432 375, 483
Clotfelter, C.T., 138, 265, 278, 280, 281, 370, 485, 494, Dougherty, K.J., 154, 238, 239, 241, 376, 406, 409–411,

518, 520 432, 446, 480, 541
Clouston, S.A.P., 268 Downey, D.B., 30, 55–68, 92, 120, 122, 124, 138, 139,
Coca, V., 232, 422
Cohen, A.M., 537 284, 330
Cohen, D.K., 289 Doyle, W.R., 411
Cohen, G.L., 364, 406, 467 Dumais, S.A., 3, 13, 17, 23, 180, 374
Duncan, G.J., 3, 4, 20, 28, 30, 62, 269, 366, 367, 379,

386, 434, 437
Duncan, O.D., 437, 458, 494
Dundar, A., 458
Dunnigan, T., 352, 353
Dynarski, S.M., 233, 364, 365, 368, 376, 387, 406, 542

Author Index 583

E Goldrick-Rab, S., 173, 175, 374, 408, 418–420, 424,
Eccles, J.S., 135, 136, 197, 256 431, 432, 435, 438, 441, 460, 535–554
Egalite, A.J., 230, 235, 518
Ehrenberg, R., 352, 518 Gonzalez, K.P., 172, 185
Eichorn, D.H., 339, 341 Gonzalez, N., 565
Elder, G.H. Jr., 225, 442 Goodman, J.D., 88, 412, 421, 425
Elder, T., 225, 226 Greenberger, E., 338, 339, 341
Elliott, D., 340 Greene, J.P., 223, 226, 230, 235, 236
Elliott, S.W., 340 Gregg, P.L., 178
Ellis, A., 133, 135 Grodsky, E., 139, 140, 143, 366, 375, 376, 405–426, 439
England, P., 3, 4, 33, 34, 92, 135–137, 396 Gross, B., 234, 237, 240, 241
Entwisle, D.R., 60, 61, 63, 65, 110, 111, 132, 256, 259, Gu, X., 419
Gumport, P., 423
284, 339, 346, 494 Gwosc, C., 177
Evans, L., 236
Evans, M., 47, 552 H
Evans, N., 199, 209 Hachen, D., 175
Evans, W.N., 226 Hacker, M., 350
Ewert, S., 134 Haeffele, L., 419
Haller, A.O., 174
F Hallinan, M.T., 120–122, 260, 318, 329, 332, 368, 369
Farkas, G., 3–35, 133, 259 Hamedani, M.G., 186
Ferrare, J.J., 267 Hamilton, L.T., 174, 175, 180, 181, 185, 187, 188, 276,
Figlio, D.N., 223, 224, 230
Finch, M.D., 341, 342 352, 366, 386, 389, 392, 395–397, 399, 435, 437,
Fink, R., 303, 305 447, 477, 479–481
Fischer, C.S., 68, 85, 91 Han, P.P., 64, 65
Fischer, M., 175, 184 Handel, S.J., 410
Flaherty, B., 340, 350 Hanselman, P., 143, 144
Fletcher, J.M., 303 Hansen, D.M., 342
Flores, S.M., 154, 155, 185 Hao, L., 137
Francis, D., 420 Harackiewicz, J.M., 546
Frank, K.A., 145, 257, 287, 289, 297–312, 328 Harley, C., 447, 449
Franklin, K., 407 Harrell, A., 465
Freedle, R., 143 Harris, A.L., 87, 92, 122, 324, 329
Freedman-Doan, P., 340 Harris, D.N., 42, 48, 142, 237, 280, 541
Freeman, K.J., 226 Harris, K.M., 87
Friedel, J.N, 419 Hart, B., 22, 28
Friedman, M., 228 Hart, C.M.D., 230
Fryberg, S.A., 176, 183 Hauser, R.M., 73, 87, 88, 339, 363, 434, 446
Fryer, R.G. Jr., 30, 57, 63, 64, 222, 225, 241, 242 Haveman, R.H., 406
Haviland, A.M., 345
G Heller, S.B., 345
Gaertner, M.N., 267 Herd, P., 268
Gamoran, A., 56, 66, 102, 121, 133, 140, 143, 256, 259, Herman, M., 146, 147
Heyns, B., 61, 63, 258
263, 266, 267, 279, 281, 303, 306, 366, 368, 369, Hilmer, M.J., 412
435, 515, 517, 518 Hirsch, B.J., 354, 521
Garcia, D.R., 73, 236 Hirschman, C., 341
Gardner, E., 137 Hoffer, T., 8, 44, 224, 225, 242, 259, 475–477, 479
Gibbs, B.G., 65, 111, 133 Hoffman, N., 422, 443
Gill, A.M., 413, 414 Holdaway, J., 123
Ginder, S.A., 351 Holland, M., 147, 374, 435, 439, 440, 444, 447
Gintis, H., 5, 7, 18, 19, 56, 89, 90, 99, 103, 138, 174, 362 Holme, J.J., 234, 237, 239, 240
Gleason, P., 222, 226, 227, 241 Hood, L., 419
Gofen, A., 182 Horn, L., 118, 154, 435, 439, 540
Goffman, E., 10, 197 Horn, S.S., 197
Golann, J.W., 241 Hout, M., 15, 100, 124, 262, 394, 432, 434, 439, 467
Goldin, C., 252, 253, 268, 378, 434, 447 Hovaguimian, L.D., 173, 179, 183
Howell, W.G., 223, 226, 235, 236