41 motivation towards e-learning) were strong predictors of student satisfaction and motivation. It is worth noting that, although Yilmaz’s (2017) study overlooked the importance of LMS self-efficacy as an aspect of e-learning self-efficacy, his research findings support the strong association between e-learning self-efficacy and student satisfaction that studies previously mentioned underscored. Academic honesty is another variable that has been correlated with e-learning contexts (Aresgado Estidola et al., 2021; Szabo et al., 2018). Research across several higher education systems not only revealed the need of preventive measures to ensure academic honesty in e-learning contexts but also indicated that false results from anticheat software and Internet restrictions placed by teachers can be detrimental for students’ e-learning journey and their relationship with instructors (Bylieva et al., 2020). Moreover, experts have recommended academic integrity e-learning training for university students as a way to stop the ever-growing wave of plagiarism present in elearning settings (Benson et al., 2019; Nikjo et al., 2021). However, although there is a vast body of research on academic dishonesty in e-learning contexts, there are few to none studies concerned with the relationship between e-learning self-efficacy and academic dishonesty. Section three looked at concepts and theories related to self-efficacy in general and e-learning self-efficacy in particular. Also, this section discussed factors that have been associated with e-learning self-efficacy, namely academic achievement and student satisfaction. Moreover, it was asserted that while there is ample research on the association between online education and academic dishonesty, studies about the impact of e-learning self-efficacy on academic honesty issues are scarce.
42 Conclusions Several ideas discussed in the first two chapters of this thesis contributed to producing a conceptual framework that associates all three variables explored in the present study. First, academic dishonesty is a variable that has been associated with low levels of English proficiency, e-learning, and religious beliefs among university students. In a more specific sense, poor academic writing skills and the use of online technologies for academically dishonest purposes are factors that predict academic dishonesty. Concerning religious beliefs, experts have different opinions when it comes to seeing them as a predictor of academic dishonesty. Second, many studies have indicated a strong relationship between English proficiency levels and language learning strategies among college students that have English as their second language. As such, L2 learning strategies have been considered effective tools for acquiring English and helping international university students improve their listening, speaking, reading, and writing skills. Ideally, the use of language learning strategies not only helps students learn the target language but also be more confident about their academic skills, which should translate into a decrease of academically dishonest behavior. Also, researchers have suggested that demographic variables like gender and year of study are predictors of what language learning strategies university students use. In this line of thought, it is important to mention that there are at least six types of L2 strategies ESL students use, namely memory, cognitive, compensation, metacognitive, affective, and social strategies. Online technologies in educational settings are also a common variable related to language learning strategies.
43 Third, e-learning self-efficacy has been linked with academic achievement and student satisfaction. However, there is scarce research on the relationship between elearning self-efficacy and academic dishonesty. Exploring such an association is crucial since academic dishonesty is a topic of growing concern in tertiary educational settings, especially nowadays when universities have been forced to transition into online instruction due to the spread of COVID-19. Fourth and last, while some studies have looked at the impact of language learning strategies on academic dishonesty and other surveys have explored the association between language learning strategies and e-learning, there is limited research on the relationship between all three variables. As such, a conceptual framework involving all three variables could help to assess students’ levels of academic dishonesty, preferences of language learning strategies, and perceptions of e-learning self-efficacy. Moreover, the inclusion of demographic variables could help ascertain whether there are differences in gender, year of study, major, and religion in terms academic honesty levels, language learning strategies preferences, and e-learning self-efficacy perceptions among international higher education students at APIU. Based on the aforementioned ideas, a conceptual framework was developed. Academic dishonesty was the dependent variable. On the other hand, language learning strategies, e-learning self-efficacy, and four demographic factors were the independent variables included in the model. The independent variables language learning strategies and e-learning self-efficacy had several sub-variables each, namely six and three, respectively. Figure 1 shows a graphic illustration of the conceptual framework developed for the study.
44 Figure 1. Conceptual Framework of the Study
45 CHAPTER 3 METHODOLOGY Introduction The purpose of this study is to explore the relationship between language learning strategies, e-learning self-efficacy, and academic honesty among Asia-Pacific International University (APIU) students. The present chapter aims to explain how such an exploration will take place. Accordingly, this section will not only restate the study’s research questions but also describe the research design, population and sample, procedures of data collection, instrumentation, methods of data analysis, and ethical considerations concerned with the study. There are five research questions developed for the present study: 1. What is the level of academic dishonesty among APIU students? 2. What learning strategies do APIU students use to learn English as a second language? 3. What are the APIU students’ perceptions of their e-learning self-efficacy? 4. Are there differences in the academic dishonesty levels, language learning strategies preferences, and e-learning self-efficacy perceptions of APIU students based on demographics? 5. Is academic dishonesty related to language learning strategies, e-learning selfefficacy, gender, year of study, major, and religion among APIU students?
46 Research Design Based on the conceptual framework developed for this study, the main objective of the research was to explore the relationship between academic dishonesty and language learning strategies, e-learning self-efficacy, and demographic variables among APIU students. The exploration of such relationship was possible via a quantitative research approach. In this sense, a survey was used to collect data among APIU international students. Survey research is not only a practical way for conducting a quantitative study but also a valuable method for including in the sample a sufficient number of participants that represent the target population in a relatively quick and timeefficient manner (Ponto, 2015). According to Jacobsen (2008), survey research falls under observational study designs and it is useful for getting a glimpse of a population’s “attitudes/beliefs” and “practices/behaviors” (p. 43). Lastly, survey research can be particularly useful when researchers face time and financial constraints (Kelley et al., 2003), which was the case in the present study. Population and Sample The target population for the present study are international undergraduate students at Asia-Pacific International University, a private higher education institution located in Saraburi, Thailand. Around 900 pupils make up the student body with 55% of them enrolled in international programs and 45% enrolled in Thai programs. Since the present research focused on international students, only pupils from the international program were asked to participate in the study. As such, the size of the sample was 225 respondents, which were involved in completing the survey used to collect data.
47 Convenience sampling, also known as accidental sampling, was used to make the process of data collection easier. Such sampling approach is considered an efficient manner to involve anyone who desires to participate in the research (Gay & Mills, 2019). Moreover, convenience sampling is acknowledged as a cost- and time-efficient manner to conduct research (Jager et al., 2017). Countries like Angola, Bangladesh, Brazil, Cambodia, Cameroon, China, Congo, Hong Kong, India, Indonesia, Japan, Kenya, Laos, Malaysia, Myanmar, Philippines, Russia, South Africa, South Sudan, Sri Lanka, Sweden, United States of America, Vietnam, Zambia, and Zimbabwe were represented among the international student body. Such a culturally diverse student population is in line with the target population for this study. Also, a representation of more than 20 countries provided a varied sample. Data Collection The process of data collection took place in three different phases. In the first phase, the research sought permission from the Institutional Review Board (IRB) committee to collect data from students. Also, some teachers and students were selected and asked to help in the process of distributing and collecting the survey. In the second phase, teachers and students chosen by the researcher were given envelopes with copies of the survey. This second part of the data collection required those distributing the survey to explain to participants what the research was about even though the instrument itself gives a clear idea of the study. Teachers and students delivering the survey were asked to invite as many participants as possible, always respecting students’ right to not participate in the research.
48 In the third phase of the data collection, the survey was compiled and given back to the researcher. This third phase entailed two things. First, it involved the keeping and securing of the data, which was carefully done by the researcher in order to comply with IRB regulations concerning participants’ anonymity and careful storing of the data. Second, it encompassed the collation, numeration, and input of all surveys into an Excel spreadsheet in order to analyze the data. Instrumentation The instrument was divided into three parts. The first section was an informed consent where purpose, procedures, risks, benefits, cost/reimbursement, participants’ rights, confidentiality, and questions/complaints were discussed. The second section was concerned with demographic information of the respondents, including gender, year of study, major, and religion. The third section of the instrument consisted of 67 Likert-type items. This section was the sum of three scales that measured students’ level of academic dishonesty, their use of language learning strategies, and their perceptions of e-learning self-efficacy. The academic dishonesty and language learning strategies scales employed a frequency/likelihood scaling technique based on a five-point scale while the e-learning self-efficacy scale used a performance approach based on a six-point scale. As such, a frequency/likelihood five-point scale (1 = Never/Almost Never, 2 = Seldom, 3 = Sometimes, 4 = Often, and 5 = Always/Almost Always) was used to measure all three variables previously mentioned. The reason for this was to help respondents answer the survey without worrying about different scaling techniques. Moreover, such an approach is consistent with the use, application, and conversion of the e-learning self-efficacy scale
49 into a five-point scale with a different scaling technique utilized by Yavuzalp and Bahcivan (2019) in the study they carried among Turkish university students. Definition of Variables Regarding language learning strategies, Oxford (2018) described them as “purposeful mental actions used by a learner to regulate his or her second or foreign (L2) learning” (p. 81). E-learning self-efficacy is, based on Bandura’s (1977) understanding of self-efficacy, the individual’s perception on how well they can perform tasks related to online learning, courses, and technologies (Yavuzalp & Bahcivan, 2019). Shoaib and Ali (2020) defined academic dishonesty as “a major problem,” present “in many educational systems in the world” and “increasing at a rapid rate with each passing day” (p. 62). Academic Dishonesty (AD) Scale The AD scale (17 items) was adopted from Eastman et al. (2008). This scale appraised students’ attitudes and experiences toward academic honesty issues. Also, the scale provided valuable insights concerning participants’ levels of academic dishonesty. Furthermore, Eastman et al. (2008) “treated academic dishonesty as a single construct” since this is the way the scale has been used in other studies they consulted (p. 216). Some items from this scale are “I fabricated or falsify a bibliography” and “I purchased or find a paper off the Internet to submit as my own work.” The original scaling technique used to measure academic dishonesty had a frequency/likelihood approach (1 = Never, 2 = Once, 3 = Few Times, 4 = Several Times, and 5 = Many Times). In terms of validity, Eastman et al. (2008) acknowledged that previous studies did not yield “any confirmatory factor analysis” (p. 216). The Cronbach’s alpha for this scale was .87.
50 Language Learning Strategies (LLS) Scale. The LLS scale (28 items) was adopted from Ardasheva and Tretter (2013). This scale evaluated participants’ attitudes and experiences when choosing L2 strategies, thus providing insights about students’ preference and use of language learning strategies. Also, the scale was divided into 6 categories or subscales, namely memory, cognitive, compensation, metacognitive, affective, and social strategies. Some items from this scale are “I act out new English words” and “I read for fun in English.” The original scaling technique used to identify preferred language learning strategies had a frequency/likelihood approach that ranged from “1 = Never of Almost Never True of Me, to 5 = Always or Almost Always True of Me” (Ardasheva & Tretter, 2013, p. 478). In terms of validity, Ardasheva and Tretter (2013) employed Confirmatory Factory Analysis (CFA), Exploratory Factor Analysis (EFA), and a panel of experts to validate the scale. Results showed that “GFI [goodness-of-fit index] and CFI [comparative fit index] met the cut-off criterion of close to .95” (Ardasheva & Tretter, 2013, p. 482), thus rendering the scale valid. The Cronbach’s alpha for this scale was .95. E-Learning Self-Efficacy (ELSE) Scale The ELSE scale (22 items) was adopted from Zimmerman and Kulikowich (2016). This scale assessed respondents’ attitudes and experiences toward e-learning selfefficacy, thus provided insights into students’ perceptions of their personal skills in online settings. Additionally, scale items were divided into three categories, namely learning in the online environment, technology use, and time management. Some items from this scale are “I navigate the online grade book” and “I communicate using asynchronous technologies (discussion boards, e-mail, etc.).”
51 The original scaling technique used to evaluate students’ perceptions of their elearning self-efficacy had “a rating of 1 [which] signified that [students] believed they would perform the task poorly,” while “a rating of 6 signified they believed they could perform the task at an expert level” (Zimmerman & Kulikowich, 2016, p. 184). In terms of validity, “evidence of convergent and divergent validity was examined through the use of correlational techniques” with values above .40 (Zimmerman & Kulikowich, 2016, p. 188), which is the minimum to confirm convergent validity (Taherdoost, 2016). The Cronbach’s alpha for this scale was .98. Data Analysis Three different software were used to collate and analyze the data, namely Microsoft Excel, open-source statistics software R Studio, and SPSS Statistics. Also, there were no items that required reversed-scoring. Missing values in the data were tackled by using a code for such eventualities. In this sense, all codes employed for data analysis are given in the appendices section. Lastly, seeing that the research instrument was multidimensional, a table was produced to explain how items, subscales, and scales are structured in the survey. Table 1 shows how scales and subscales encompass survey items, thus giving a comprehensive overview of how the third section of the research instrument is structured. In terms of data analysis, it is crucial to give a brief description of how the research questions were addressed via statistical analysis. The first three research questions were answered by using descriptive statistics. In other words, the number of respondents, means, standard deviations, and percentages for the observed variables, subvariables, and individual items of the survey were calculated. The fourth research
52 Table 1. Structure of the Survey Scales Subscales Items Language Learning Strategies (6 Subscales – 28 Items) 1–28 Memory 1–7 Cognitive 8–12 Compensation 13–17 Metacognitive 18–21 Affective 22–24 Social 25–28 E-Learning Self-Efficacy (3 Subscales – 22 Items) 29–50 Online Learning 32, 34, 38–40, 43, 45, 46, 49, 50 Technology Use 29–31, 33, 35, 41, 42 Time Management 36, 37, 44, 47, 48 Academic Dishonesty (Single Construct – 17 Items) 51–67 question was addressed by using comparative statistics. Consequently, t-test was conducted to determine whether there were gender differences in terms of academic honesty, language learning strategies, and e-learning self-efficacy among APIU students. Also, ANOVA was used to ascertain whether there were differences in academic dishonesty, language learning strategies, and e-learning self-efficacy variables based on year of study, major, and religion among APIU students. The fifth research question was answered by using relational statistics. Subsequently, Pearson correlation was applied in order to ascertain the strength of the association between academic dishonesty, language learning strategies, and e-learning self-efficacy. Also, a multiple regression analysis was utilized for the same purpose,
53 evaluating the relationship between academic dishonesty as the dependent variable and language learning strategies, e-learning self-efficacy, gender, year of study, major, and religion as the independent variables. Ethical Considerations The approval of the Institutional Review Board Committee was sought for the present study. As stated before, the survey used to collect data had a consent form section to comply with the Institutional Review Board regulations regarding human subject research. Furthermore, every respondent that answered the survey did so voluntarily. Lastly, the process of data collection did not put the participants’ physical or mental health at risk. Conclusions The present chapter aimed to explain the methodology of the study. Moreover, details concerning the study’s research design and procedures of data collection, population and sample, instrumentation, variables, and ethical considerations were given. The following chapter will examine the results of the study in relation to the questions developed for the research.
54 CHAPTER 4 RESULTS Introduction The purpose of chapter four is to answer the research questions developed for the study. Such an objective involves the analysis of the data collected among APIU international students. In this sense, the chapter is divided into four sections. The ollowing chapter overview outlines how the study’s results are presented. Overview of the Chapter The introductory section gives a summary of the chapter and shows important demographic characteristics of the sample. The second section, descriptive statistics, provides valuable information (i.e., means, standard deviations, skewness, and Cronbach’s alpha) about the scales and subscales of the research instrument. Moreover, it seeks to answer the first research question, which focuses on the academic dishonesty levels of APIU students. Also, section two is concerned with the second research question, which pursues to ascertain the respondents’ preferences of language learning strategies. Lastly, section two tackles the third research question, which is related to APIU students’ perceptions of their e-learning self-efficacy. The third section, comparative statistics, answers the fourth research question of the study, seeking to determine whether there were differences in the levels of academic dishonesty, preference of language learning strategies, and perceptions of e-learning self-
55 efficacy among respondents based on subgroups determined by demographic variables. For gender differences, t-test was employed. For differences based on other categorical variables (i.e., year of study, major, and religion), ANOVA was utilized. The fourth and last section of the chapter, relational statistics, answers the fifth research question of the study by (1) looking at the correlation between academic dishonesty, language learning strategies, and e-learning self-efficacy and (2) exploring via multiple regression analysis the strength of the relationship between academic dishonesty as the dependent variable and language learning strategies, e-learning selfefficacy, gender, year of study, major, and religion as the independent variables. Demographics In terms of demographics, the study provided information concerning the gender, year of study, major, and religion of the respondents. Concerning gender, 56% of the participants were female. Sorted by college year, 27% were Seniors, followed by Freshmen at 26%, Juniors at 24%, and Sophomores at 23%. Regarding major, English (25%), Business (21%), and Education (20%) majors constituted majority of the sample while Religion, Nursing, Science, and Information Technology majors were thinly distributed (see Table 2). Regarding religious affiliation, Adventists (45%), Christians (32%), and Buddhists (16%) were majority while 6% of the respondents identified themselves with other faiths. Table 2 not only shows the statistics previously mentioned but also the number of participants pertaining to each demographic category.
56 Table 2. Demographics of the Study Variable N % Gender Male 99 44.0 Female 126 56.0 Year of Study Freshman 59 26.2 Sophomore 51 22.7 Junior 54 24.0 Senior 61 27.1 Major Business 47 20.9 Education 46 20.4 English 56 24.9 Religion 25 11.1 Science 17 7.6 Information Technology 16 7.1 Nursing 18 8.0 Religious Affiliation Adventist 102 45.3 Christian 72 32.0 Buddhist 37 16.4 Other 14 6.2 Descriptive Statistics The means, standard deviations, skewness, and reliability of all scales and subscales were calculated. Such an analysis is not only considered good practice but also confirmed the reliability of the survey as a whole, as well as its different parts. In this sense, all three scales, namely academic dishonesty (AD), language learning strategies (LLS), and e-learning self-efficacy (ELSE), were found reliable with Cronbach’s alpha indicators of .94, .90, and .91, respectively. Table 3 not only contains the scales’ reliability estimates previously discussed but also gives more information about descriptive statistics of the scales and subscales of the research instrument.
57 Table 3. Descriptive Statistics and Reliability Estimates of Scales and Subscales Scale (Subscale) M SD Skewness No. of Items Cronbach’s alpha AD 2.13 0.84 0.77 17 .94 LLS 3.35 0.58 -0.27 28 .90 (Memory) 3.13 0.72 -0.14 7 .79 (Cognitive) 3.35 0.72 -0.39 5 .67 (Compensation) 3.36 0.72 -0.33 5 .62 (Metacognitive) 3.89 0.81 -0.50 4 .82 (Affective) 2.82 1.04 0.03 3 .73 (Social) 3.54 0.87 -0.35 4 .77 ELSE 3.40 0.61 0.03 22 .91 (Online Learning) 3.36 0.67 0.06 10 .83 (Technology Use) 3.47 0.72 -0.05 7 .79 (Time Management) 3.37 0.72 -0.18 5 .75 Level of Academic Dishonesty For academic dishonesty (AD), APIU international students indicated they seldom engage in academically dishonest behavior (M = 2.13, SD = 0.84). However, a closer look at academic dishonesty items via descriptive statistics highlights several results. For instance, the item with the highest mean was “I manually pass answers in an exam” (see Table 4 Item 55). Around 28% of the respondents indicated they “often” or “always/almost always” engage in this sort of dishonest behavior, which can suggest that students frequently share information with their classmates during a test. Moreover, about 20% of the participants suggested they regularly “work with others on an individual project” (see Table 4 Item 59) and “ask someone about the content of an exam from someone who has taken it” (see Table 4 Item 57). On the other hand, the statement “I purchase or find a paper off the internet to submit as my own work” had the lowest mean
58 Table 4. Statistics of Academic Dishonesty Items Item No./Statement N M SD %a I55 I manually pass answers in an exam. 224 2.60 1.28 27.7 I59 I work with others on an individual project. 225 2.50 1.15 20.0 I51 I use crib notes on a test. 224 2.42 1.17 16.6 I57 I ask someone about the content of an exam from someone who has taken it. 225 2.33 1.20 19.6 I62 I receive substantial, unprecedented help on an assignment. 224 2.25 1.06 11.2 I58 I give information about the content of an exam to someone who has not yet taken it. 225 2.22 1.17 16.0 I63 I copy a few sentences of material from a published source without footnoting it. 225 2.21 1.12 14.2 I56 I have someone check over a paper before turning it in. 223 2.10 1.25 16.6 I54 I cheat on a test in any other way. 225 2.07 1.13 12.4 I61 I take credit for full participation in a group project without doing a fair share of the work. 224 2.03 1.16 14.3 I60 I visit a professor to influence a grade. 224 2.03 1.20 14.7 I53 I help someone cheat on a test. 225 2.02 1.15 13.8 I52 I copy from another student on the test. 225 1.99 1.09 11.5 I64 I fabricate or falsify a bibliography. 224 1.92 1.16 12.5 I67 I use a cell phone or another device to photograph an exam. 225 1.88 1.12 10.7 I66 I use a cell phone to text messages asking for help during an exam. 224 1.87 1.15 13.0 I65 I purchase or find a paper off the Internet to submit as my own work. 225 1.74 1.11 9.8 Note: %a = participants’ percentage indicating ‘often’ and ‘always/almost always.’ and percentage (see Table 4 Item 65). Table 4 shows the total responses, means, standard deviations, and percentages related to items from the academic dishonesty scale. Use of Language Learning Strategies For language learning strategies (LLS), respondents indicated neutrality toward their practice of thoughts, actions, and techniques that would be beneficial for acquiring English as a second language (M = 3.35, SD = 0.58). In terms of students’ preferences or
59 choice of specific categories of L2 learning strategies, metacognitive strategies scored the highest, followed by social, compensation, cognitive, memory, and affective strategies (see Table 5). Another interesting result concerning individual metacognitive strategies is that majority of students participating in the study answered they “often” or “always/almost always” used these strategies to acquire the target language (see Table 6). Additionally, it is noteworthy that the compensation strategy “if I can’t think of an English word, I use a word that means the same thing” (M = 3.95, SD = 0.96) and the social strategy “I practice English with other students” (M = 3.71, SD = 1.09) scored higher than other items. Lastly, the memory strategy “I use flashcards to learn new words in English” (M = 2.48, SD = 1.11) scored the lowest among all scale LLS items. Table 5 shows the total responses, means, and standard deviations of all six categories in the language learning strategies scale. Table 6 shows the total responses, means, standard deviations, and percentages related to metacognitive items. Table 5. Descriptive Statistics of LLS Subscales LLS Subscale N M SD Metacognitive 225 3.89 0.81 Social 225 3.54 0.87 Compensation 225 3.36 0.72 Cognitive 225 3.35 0.72 Memory 225 3.13 0.72 Affective 225 2.82 1.04
60 Table 6. Statistics of Metacognitive Items Item No./Statement N M SD %a I18 I see my English mistakes and try to do better. 225 4.00 0.96 71.6 I20 I look for ways to be a better student of English. 224 3.94 0.96 68.3 I19 I listen well (carefully) when people speak English. 225 3.93 0.96 68.9 I21 I think about how well I am doing in English. 225 3.69 1.11 59.6 Note: %a = participants’ percentage indicating ‘often’ and ‘always/almost always.’ Perception of E-Learning Self-Efficacy For e-learning self-efficacy (ELSE), participants demonstrated a neutral stance in relation to their perceptions of e-learning self-efficacy (M = 3.40, SD = 0.61). The subscales of elearning self-efficacy, namely online learning, technology use, and time management, indicated relatively similar values (see Table 7). The three statements that scored the highest among all ELSE items were related to technology use skills (see Table 8). It is also worth mentioning that “I use the library’s online resources efficiently” (see Table 8 Item 49) scored the lowest among all ELSE items. Table 7 shows the descriptive statistics for all three ELSE categories. Table 8 shows the three ELSE statements with the highest score and the three ELSE statements with the lowest score among all participants, as wells as these selected items’ total responses, means, and standard deviations. Table 7. Descriptive Statistics of ELSE Subscales ELSE Subscale N M SD Technology Use 225 3.47 0.72 Online Learning 225 3.38 0.67 Time Management 225 3.37 0.72
61 Table 8. Descriptive Statistics of Selected ELSE Items Item No./Statement N M SD High I33 I submit assignments to an online drop box. 224 3.82 1.12 I41 I search the Internet to find the answer to a courserelated question. 225 3.78 1.04 I42 I search the online course materials. 224 3.66 1.06 Low I44 I meet deadlines with very few reminders 225 3.17 1.01 I38 I find the course syllabus online. 225 3.16 1.13 I49 I use the library’s online resources efficiently. 224 2.92 1.12 Comparative Statistics Using comparative statistics, it was evaluated whether there were differences in the levels of academic dishonesty, preferences of language learning strategies, and perceptions of e-learning self-efficacy among respondents based demographic variables. For gender differences, t-test was employed. For differences based on other categorical variables (i.e., year of study, major, and religion), ANOVA was utilized. Differences in Academic Dishonesty by Demographics Concerning differences in the levels of academic dishonesty (AD) by gender among APIU students, results suggested there were no differences, t(223) = 0.99, p = 0.33 (see Table 9). Also, there were no differences by year of study [F(3, 221) = 1.47, p = 0.22] (see Table 10) or by religious affiliation [F(3, 221) = 1.84, p = 0.14] (see Table 12) with academic dishonesty as the dependent variable. However, there was a statistically significant difference by major for academic dishonesty levels [F(6, 218) = 2.53, p = .022] (see Table 11). Post Hoc comparisons using the Tukey HSD test revealed that the academic dishonesty mean score of Nursing students (M = 2.67, SD = 0.96) was
62 significantly higher than that of Education students (M = 1.88, SD = 0.62). Table 9 shows the t-test results for gender differences in academic dishonesty. Tables 10, 11, and 12 show the ANOVA results when year of study, major, and religious affiliation, respectively, are the predictors for academic dishonesty. Table 9. Results for Gender Differences in AD Gender N M SD t df p ES(d) Male 99 2.19 0.84 0.99 223 .33 0.13 Female 126 2.08 0.84 Table 10. ANOVA Results for AD as Criterion and Year as Predictor Year N M SD F df1,df2 p η 2 Freshman 59 1.97 0.83 1.47 3, 221 .223 0.020 Sophomore 51 2.31 0.96 Junior 54 2.15 0.70 Senior 61 2.11 0.83 Table 11. ANOVA Results for AD as Criterion and Major as Predictor Major N M SD F df1,df2 p η 2 Business 47 2.14 0.89 2.53 6, 218 .022 .07 Education 46 1.88 0.62 English 56 2.19 0.78 Religion 25 2.27 0.95 Science 17 1.97 0.96 Information Technology 16 1.91 0.65 Nursing 18 2.67 0.96
63 Table 12. ANOVA Results for AD as Criterion and Religion as Predictor Religion N M SD F df1,df2 p η 2 Buddhist 37 2.23 1.00 1.84 3, 221 .14 .02 Christian 72 2.27 0.82 Adventist 102 1.98 0.77 Other 14 2.16 0.81 Differences in Language Learning Strategies by Demographics Regarding differences in the preferences of language learning strategies (LLS) by gender among APIU students, no differences were found except in compensation strategies, t(223) = -2.18, p = 0.03 (see Table 13). Furthermore, no differences were found by year of study (see Table 14) or major (see Table 15) in any language learning strategies. However, there was a statistically significant difference by religious affiliation for language learning strategies, namely for metacognitive strategies [F(3, 221) = 4.62, p = .004] (see Table 16). In this sense, Post Hoc comparisons using the Tukey HSD test revealed that Adventist students’ use of metacognitive strategies (M = 4.07, SD = 0.79) was significantly higher than that of Buddhist students (M = 3.54, SD = 0.90). Table 13 shows the t-test results for gender differences in language learning strategies. Tables 14, 15, and 16 show the ANOVA results when year of study, major, and religious affiliation, respectively, are the predictors for language learning strategies.
64 Table 13. Results for Gender Differences in LLS Subscale Gender N M SD t df p ES(d) Memory Male 99 3.18 0.72 0.76 223 .45 0.102 Female 126 3.10 0.73 Cognitive Male 99 3.40 0.67 0.88 223 .38 0.117 Female 126 3.31 0.76 Compensation Male 99 3.24 0.68 - 2.18 223 .03 - 0.292 Female 126 3.45 0.73 Metacognitive Male 99 3.92 0.76 0.46 223 .65 0.061 Female 126 3.87 0.84 Affective Male 99 2.82 1.02 0.03 223 .98 0.003 Female 126 2.82 1.05 Social Male 99 3.52 0.81 - 0.27 223 .79 - 0.036 Female 126 3.55 0.92 Table 14. ANOVA Results for LLS as Criterion and Year as Predictor Subscale Year N M SD F df1,df2 p η 2 Memory Freshman 59 3.12 0.52 0.40 3 , 221 .75 0.005 Sophomore 51 3.09 0.63 Junior 54 3.23 0.86 Senior 61 3.10 0.84 Cognitive Freshman 59 3.34 9.62 0.44 3, 221 .73 0.006 Sophomore 51 3.45 9.62 Junior 54 3.33 9.80 Senior 61 3.30 9.82 Compensation Freshman 59 3.23 0.74 1.96 3, 221 .12 0.026 Sophomore 51 3.56 0.69 Junior 54 3.34 0.76 Senior 61 3.32 0.66 Subscale Year N M SD F df1,df2 p η 2 Metacognitive Freshman 59 3.96 0.92 0.30 3, 221 .83 0.004 Sophomore 51 3.92 0.75 Junior 54 3.83 0.95 Senior 61 3.85 0.56
65 Table 14 (continued) Affective Freshman 59 2.67 0.96 0.71 3, 221 .55 0.009 Sophomore 51 2.80 1.07 Junior 54 2.89 1.04 Senior 61 2.93 1.08 Social Freshman 59 3.47 0.94 1.83 3, 221 .14 0.024 Sophomore 51 3.76 0.81 Junior 54 3.57 0.82 Senior 61 3.39 0.88 Table 15. ANOVA Results for LLS as Criterion and Major as Predictor Subscale Major N M SD F df1,df2 p η 2 Memory Business 47 3.11 0.68 0.88 6, 218 .51 0.024 Education 46 3.11 0.77 English 56 3.30 0.64 Religion 25 2.99 0.71 Science 17 2.96 0.95 Information Tech 16 3.20 0.78 Nursing 18 3.06 0.71 Cognitive Business 47 3.34 0.69 1.59 6, 218 .15 0.042 Education 46 3.33 0.86 English 56 3.38 0.54 Religion 25 3.10 0.94 Science 17 3.55 0.70 Information Tech 16 3.71 0.46 Nursing 18 3.20 0.72 Cognitive Business 47 3.46 0.67 2.01 6, 218 .07 0.052 Education 46 3.38 0.86 English 56 3.52 0.63 Religion 25 3.06 0.79 Science 17 3.05 0.74 Information Tech 16 3.38 0.62 Nursing 18 3.23 0.50 Subscale Major N M SD F df1,df2 p η 2 Metacognitive Business 47 3.77 0.88 1.65 6, 218 .14 0.043 Education 46 3.90 0.86 English 56 3.89 0.67 Religion 25 3.82 0.70 Science 17 4.31 0.78 Information Tech 16 4.17 0.61 Nursing 18 3.63 1.04
66 Table 15 (continued) Affective Business 47 2.98 1.10 1.00 6, 218 .43 0.027 Education 46 2.90 1.06 English 56 2.93 0.89 Religion 25 2.59 1.04 Science 17 2.53 1.17 Information Tech 16 2.54 0.91 Nursing 18 2.72 1.17 Social Business 47 3.59 0.83 0.52 6, 218 .79 0.014 Education 46 3.48 0.92 English 56 3.64 0.77 Religion 25 3.50 0.80 Science 17 3.59 1.07 Information Tech 16 3.52 1.09 Nursing 18 3.25 0.87 Table 16. ANOVA Results for LLS as Criterion and Religion as Predictor Subscale Religion N M SD F df1,df2 p η 2 Memory Buddhist 37 3.19 0.59 0.51 3, 221 .676 0.007 Christian 72 3.19 0.69 Adventist 102 3.09 .081 Other 14 2.99 0.91 Cognitive Buddhist 37 3.49 0.74 1.98 3, 221 .119 0.026 Christian 72 3.33 0.61 Adventist 102 3.27 0.79 Other 14 3.69 0.54 Compensation Buddhist 37 3.29 0.72 2.43 3, 221 .067 0.032 Christian 72 3.50 0.67 Adventist 102 3.25 0.73 Other 14 3.59 0.71 Subscale Religion N M SD F df1,df2 p η 2 Metacognitive Buddhist 37 3.54 0.90 4.62 3, 221 .004 0.059 Christian 72 3.86 0.71 Adventist 102 4.07 0.79 Other 14 3.93 0.81 Affective Buddhist 37 2.96 1.00 1.30 3, 221 .277 0.017 Christian 72 2.85 1.00 Adventist 102 2.70 1.06 Other 14 3.19 1.08
67 Table 16 (continued) Social Buddhist 37 3.59 0.87 0.22 3, 221 .885 0.003 Christian 72 3.48 0.73 Adventist 102 3.55 0.96 Other 14 3.63 0.93 Differences in E-Learning Self-Efficacy by Demographics Relating to perceptions of e-learning self-efficacy (ELSE) by gender among APIU students, there were no differences in online learning, technology use, or time management (see Table 17). Moreover, no differences were found by year of study (see Table 18), major (see Table 19), or religious affiliation (see Table 20) in any of the three e-learning self-efficacy subscales. Table 17 shows the t-test results for gender differences in online learning, technology use, and time management. Tables 18, 19, and 20 show the ANOVA results when year of study, major, and religious affiliation, respectively, are the predictors for e-learning self-efficacy subscales. Table 17. Results for Gender Differences in ELSE Subscale Gender N M SD t df p ES(d) Online Learning Male 99 3.37 0.65 -.19 223 .847 - 0.026 Female 126 3.38 0.69 Technology Use Male 99 3.43 0.70 -.74 223 .456 - 0.099 Female 126 3.50 0.74 Time Management Male 99 3.33 0.67 -76 223 .449 - 0.102 Female 126 3.41 0.75
68 Table 18. ANOVA Results for ELSE as Criterion and Year as Predictor Subscale Year N M SD F df1,df2 p η 2 Online Learning Freshman 59 3.33 0.70 1.12 3, 221 .342 0.015 Sophomore 51 3.35 0.64 Junior 54 3.30 0.71 Senior 61 3.51 0.63 Technology Use Freshman 59 3.43 0.81 0.23 3, 221 .873 0.003 Sophomore 51 3.45 0.73 Junior 54 3.46 0.65 Senior 61 3.53 0.69 Time Management Freshman 59 3.46 0.68 1.08 3, 221 .358 0.014 Sophomore 51 3.33 0.74 Junior 54 3.25 0.76 Senior 61 3.44 0.68 Table 19. ANOVA Results for ELSE as Criterion and Major as Predictor Subscale Major N M SD F df1,df2 p η 2 Online Learning Business 47 3.34 0.61 1.69 6, 218 .124 0.044 Education 46 3.38 0.65 English 56 3.26 0.60 Religion 25 3.23 0.64 Science 17 3.61 0.67 Information Tech 16 3.73 0.52 Nursing 18 3.47 0.71 Technology Use Business 47 3.51 0.64 1.75 6, 218 .110 0.046 Education 46 3.41 0.69 English 56 3.34 0.57 Religion 25 3.38 0.70 Science 17 3.84 0.87 Information Tech 16 3.78 0.62 Nursing 18 3.44 0.80 Time Management Business 47 3.51 0.78 1.18 6, 218 .331 0.031 Education 46 3.45 0.85 English 56 3.19 0.71 Religion 25 3.30 0.81 Science 17 3.52 0.91 Information Tech 16 3.34 0.57 Nursing 18 3.42 0.84
69 Table 20. ANOVA Results for ELSE as Criterion and Religion as Predictor Subscale Religion N M SD F df1,df2 p η 2 Online Learning Buddhist 37 3.41 0.64 1.01 3, 221 .391 0.013 Christian 72 3.28 0.58 Adventist 102 3.40 0.73 Other 14 3.58 0.75 Technology Use Buddhist 37 3.60 0.73 1.60 3, 221 .191 0.021 Christian 72 3.33 0.63 Adventist 102 3.53 0.75 Other 14 3.39 0.82 Time Management Buddhist 37 3.57 0.71 1.85 3, 221 .138 0.025 Christian 72 3.31 0.62 Adventist 102 3.31 0.77 Other 14 3.61 0.69 Relational Statistics In order to ascertain the association between academic dishonesty, language learning strategies, and e-learning self-efficacy, relational statistics were used in two different ways. First, the Pearson Product Correlation was calculated to evaluate the strength of the relationship between academic dishonesty, the six language learning strategies subscales, and the three e-learning self-efficacy dimensions. Second, multiple regression was used to assess the link between academic dishonesty as the dependent variable and language learning strategies categories (i.e., memory, cognitive, compensation, metacognitive, affective, and social), e-learning self-efficacy subscales (i.e., online learning, technology use, and time management), and demographics (i.e., gender, year of study, major, and religious affiliation) as independent variables. Results concerned with correlation and regression are given below.
70 Correlation Results Results indicated a positive weak correlation between academic dishonesty and three language learning strategies, namely memory (r = .16, n = 225, p > .05, 95% CI [.03, .28]), cognitive (r = .15, n = 225, p > .05, 95% CI [.02, .28]), and compensation (r = .17, n = 225, p > .05, 95% CI [.04, .29]) strategies. On the other hand, a negative weak correlation was found between academic dishonesty and metacognitive strategies (r = - .14, n = 225, p > .05, 95% CI [-.27, -.01]). Additionally, there was a significant slight correlation between academic dishonesty and affective strategies (r = .29, n = 225, p > .01, 95% CI [.16, .40]). Still, no significant correlation between academic dishonesty and any of the e-learning self-efficacy subscales was found (see Table 21). Correlation results also indicated significant relationships between e-learning selfefficacy subscales and language learning strategies. For instance, a mild correlation was found between the subscale online learning and all language learning strategies, metacognitive approaches (r = .57, n = 225, p > .01, 95% CI [.47, .65]) being the ones with the highest correlation. Also, there was a moderate correlation between the subscale technology use and all language learning strategies but one (see Table 21). In this sense, a strong correlation between technology use and social strategies was found (r = .84, n = 225, p > .01, 95% CI [.79, .87]). Table 21 shows the correlations mentioned above. It also shows all the correlations between academic dishonesty (AD), language learning strategies categories (i.e., memory (ME), cognitive (CO), compensation (CM), metacognitive (MC), affective (AF), and social (SC)), and e-learning self-efficacy dimensions (i.e., online learning (OL), technology use (TC), and time management (TM)). Lastly, Table 21 shows the 95% confidence interval for each correlation.
71 Table 21. Correlations with Confidence Intervals Variable 1 2 3 4 5 6 7 8 9 1. AD 2. ME .16* [.03, .28] 3. CO .15* .48** [.02, .28] [.37, .57] 4. CM .17* .48** .46** [.04, .29] [.38, .58] [.36, .56] 5. MC -.14* .41** .45** .39** [-.27, -.01] [.30, .51] [.33, .54] [.28, .50] 6. AF .29** .53** .40** .32** .19** [.16, .40] [.43, .62] [.28, .50] [.20, .43] [.06, .31] 7. SC .08 .58** .39** .43** .47** .45** [-.05, .21] [.49, .66] [.27, .50] [.31, .53] [.37, .57] [.33, .54] 8. OL -.02 .42** .43** .35** .57** .32** .44** [-.15, .11] [.31, .52] [.32, .53] [.22, .46] [.47, .65] [.19, .43] [.33, .54] 9. TC .10 .64** .44** .44** .50** .59** .84** .56** [-.03, .23] [.55, .71] [.32, .54] [.33, .54] [.39, .59] [.50, .67] [.79, .87] [.47, .65] 10. TM -.01 .38** .38** .28** .54** .28** .29** .80** .42** [-.14, .12] [.26, .48] [.27, .49] [.16, .40] [.44, .63] [.15, .39] [.17, .41] [.75, .85] [.30, .52] Note: Values in square brackets indicate the 95% confidence interval for each correlation. The confidence interval is a plausible range of population correlations that could have caused the sample correlation (Cumming, 2014). * indicates p < .05. ** indicates p < .01. Also, two scatter plots were developed. The first scatter plot illustrates the relationship between academic dishonesty and affective strategies (see Figure 2). The
72 second one demonstrates the association between technology use and social strategies (see Figure 3). Figure 2. Scatter Plot of Academic Dishonesty and Affective Strategies Figure 3. Scatter Plot of Technology Use and Social Strategies Regression Results Hierarchical regression analysis was utilized to determine the strength of the relationship between academic dishonesty as the dependent variable and language learning strategies subscales, e-learning self-efficacy dimensions, and demographics (i.e.,
73 gender, year of study, major, and religion) as the independent variables. Hierarchical regression is a way to evaluate whether the variance in independent variables is statistically significant. In this sense, the idea was to develop several regression models that accounted for the impact of language learning strategies subscales, e-learning selfefficacy dimensions, and demographics on APIU students’ level of academic dishonesty. After keying in all thirteen independent variables, SPSS Statistics developed six regression models that pinpointed which independent variables were statistically significant predictors for academic dishonesty from strongest to weakest. Regression results indicated that the thirteen independent variables included in the analysis explained 22% of the variance (R 2 =.22, F (22,202) = 3.34, p < .001, R 2 adjusted = .20). Also, several variables constituted statistically significant predictors of academic dishonest. In this sense, it was found that affective strategies (β = .29, p <. 01) significantly explained the variance of academic dishonesty. Moreover, Nursing students (β = .18, p <. 01) and compensation strategies (β = .16, p <. 01) constituted important predictors of academic dishonesty. Among class levels, results indicated that Sophomore students (β = .13, p <. 05) as a group had a significant association with academic dishonesty. On the other hand, metacognitive strategies (β = -.25, p <. 01) and Education students (β = -.14, p <. 05) were important negative predictors of academic dishonesty. Moreover, in terms of hierarchy, metacognitive strategies were the second most important factor. Lastly, regression results suggested that e-learning self-efficacy subscales, gender, and religious affiliation were not important factors related to academic
74 dishonesty among APIU students. Table 22 shows the different regression models in their hierarchical order. Table 23 gives the results of the multiple regression analysis. Table 22. Regression Models in Hierarchical Order Model R R2 Adj R2 SE 1 .285a .081 .077 .803 2 .349b .122 .114 .787 3 .394c .155 .144 .778 4 .427d .182 .167 .763 5 .448e .201 .183 .755 6 .465f .216 .195 .750 a. Predictors: (Constant), Affective b. Predictors: (Constant), Affective, Metacognitive c. Predictors: (Constant), Affective, Metacognitive, Nursing d. Predictors: (Constant), Affective, Metacognitive, Nursing, Compensation e. Predictors: (Constant), Affective, Metacognitive, Nursing, Compensation, Education f. Predictors: (Constant), Affective, Metacognitive, Major Nursing, Compensation, Education, Sophomore
75 Table 23. Multiple Regression Analysis Results for AD as Criterion Model Variable b SE β t p r Partial Part 1 (Constant) 1.479 .156 9.503 .000 Affective .230 .052 .285 4.445 .000 .285 .285 .285 2 (Constant) 2.217 .277 8.014 .000 Affective .261 .052 .324 5.061 .000 .285 .322 .318 Metacognitive -.212 .066 -.205 -3.197 .002 -.143 -.210 -.201 3 (Constant) 2.096 .275 7.627 .000 Affective .263 .051 .326 5.180 .000 .285 .329 .320 Metacognitive -.194 .066 -.187 -2.963 .003 -.143 -.195 -.183 Nursing .565 .191 .184 2.961 .003 .193 .195 .183 4 (Constant) 1.734 .303 5.729 .000 Affective .225 .052 .279 4.322 .000 .285 .280 .264 Metacognitive -.260 .069 -.251 -3.762 .000 -.143 -.246 -.229 Nursing .571 .188 .186 3.034 .003 .193 .200 .185 Compensation .216 .080 .185 2.686 .008 .166 .178 .164 Model Variable b SE β t p r Partial Part 5 (Constant) 1.796 .301 5.965 .000 Affective .229 .052 .284 4.443 .000 .285 .288 .268 Metacognitive -.263 .069 -.253 -3.832 .000 -.143 -.251 -.231 Nursing .507 .189 .165 2.687 .008 .193 .179 .162 Compensation .216 .080 .185 2.710 .007 .166 .180 .164 Education -.288 .126 -.139 -2.282 .023 -.152 -.152 -.138 6 (Constant) 1.784 .299 5.965 .000 Affective .236 .051 .293 4.597 .000 .285 .297 .276 Metacognitive -.256 .068 -.247 -3.753 .000 -.143 -.246 -.225 Nursing .551 .189 .179 2.924 .004 .193 .194 .175 Compensation .188 .080 .161 2.348 .020 .166 .157 .141 Education -.295 .126 -.143 -2.350 .020 -.152 -.157 -.141 Sophomore .251 .122 .126 2.057 .041 .115 .138 .123
76 CHAPTER 5 CONCLUSIONS AND RECOMMENDATIONS Conclusions The findings of the study are as follows. First, international students at AsiaPacific International University indicated they seldom engage in academically dishonest behavior. This finding is not in agreement with several studies who have highlighted that ESL students at international universities tend to engage in academic dishonesty due to their low levels of English proficiency (Bista, 2011; Conzett et al., 2010). However, it is important to stress that occasionally committing cheating, plagiarism, or any other form of academic dishonesty still constitutes unacceptable behavior in the academic setting, regardless of the prevalence of academic dishonesty among students. Moreover, around 28% students admitted to “manually pass[ing] answers in an exam” to their friends and classmates. Explaining this significant finding entailed revisiting neutralization theory (Sykes & Matza, 1957). According to Sykes and Matza (1957), students excuse their act of cheating using different practices or techniques, appeal to higher loyalties being one of them. Thus, it is possible that more than a fourth of the sample believes passing answers is acceptable as long as it is done to help a friend or classmate. Such an attitude toward cheating could be understood in the context of collectivist societies, like the ones in Southeast Asia, where pupils do not consider helping others to do an individual assignment or get a good grade on a final exam as something dishonest (Thompson et al., 2017). In this sense, Broeckelman-Post (2008)
77 suggested that teachers should talk about academic dishonesty in the classroom, thus helping students get the conversation going among them, which could dissuade undergraduates from committing plagiarism and cheating. Another finding was that students in general indicated to use or prefer metacognitive strategies over social, compensation, cognitive, memory, and affective strategies. Such finding suggests that ESL students are inclined to use observation and self-assessment since items like “I listen well (carefully) when people speak English,” “I look for ways to be a better student of English,” and “I see my English mistakes and try to do better” scored the highest among APIU international students. The concept of mediation developed by Vygotsky (1978) could explain students’ use of observation and self-assessment strategies. Vygotsky’s (1978) theoretical model of mediation posits selfregulation as an important cornerstone of the learning process. In simple words, students tend to use learning strategies to mediate/regulate their acquisition of new knowledge more effectively. Consequently, APIU students’ preference of metacognitive strategies not only highlights them as the most effective strategies but also reaffirms Vygotsky’s (1978) ideas of self-regulation and mediation as a key component in the learning process. In a more particular sense, ANOVA results suggested that Adventist students’ use of metacognitive strategies was significantly higher than that of Buddhist students. Most of the Adventist students are international while most Buddhist students are Thai. Such finding is in agreement with studies that indicated that metacognitive strategies are favored by international students at universities where English is the lingua franca (BinHady et al., 2020; Raoofi et al., 2017; Solak & Cakir, 2015).
78 In relation to e-learning self-efficacy, a third finding emerged. Although international students demonstrated a neutral stance in relation to their perceptions of elearning self-efficacy, results using descriptive statistics indicated that APIU students infrequently “use the library’s online resources efficiently.” Considering the importance of knowing how to use the library, especially in the context of doing research, the fact that the aforementioned item scored the lowest is noteworthy. Perhaps, pupils feel it is not important to be familiar with online library resources to get through college. Still, not knowing effective ways in which to use the library might lead to struggle academically. A further conclusion was that the variable major constituted a predictor for academic dishonesty among APIU international students. In a more specific sense, Nursing students were found to be the most likely to engage in academic dishonesty. ANOVA results indicated that the mean score of Nursing students was significantly different from Education students. Also, regression results indicated a significant link between Nursing students and academic dishonesty. A similar result was found in previous research that looked at the association between Nursing as a major and academic dishonesty (Abusafia et al., 2018). Studies conducted among Nursing students pointed out that factors like high expectations from teachers, lack of strict academic honesty policies, and the need for getting good grades prompted university students to cheat and plagiarize (Birks et al., 2018; Kiekkas et al., 2020). As such, it can be asserted that academic is not only a recurrent issue among Nursing students but also has to be addressed. Perhaps, deterrence theory (Zimring & Hawkins, 1973) is one of the theoretical frameworks that could be used to tackle the issue at hand since previous research supported its application (Birks et al., 2018).
79 The fifth finding was the significant slight correlation between academic dishonesty and memory, cognitive, compensation, and affective strategies. Likewise, regression results indicated that affective strategies were the strongest indicator of academic dishonesty. These results imply that students using memory, cognitive, compensation, and affective strategies are likely to be involved in academic dishonesty. On the other hand, a negative weak correlation was found between academic dishonesty and metacognitive strategies. Furthermore, regression results pointed to metacognitive strategies as a negative predictor of academic dishonesty. Such findings could imply that students who are aware of their own learning patterns are less likely to engage in academic dishonesty. Significant correlations between e-learning self-efficacy subscales and language learning strategies infer that international students see technology and language acquisition as important and complementary aspects of their academic journey. Also, this finding suggests that students might enjoy using online technologies as a way for improving their acquisition of the English language. Other researchers that looked at the association between e-learning and L2 acquisition pointed out similar results among international students (Ahmadi, 2018; Zhou & Wei, 2018). The last finding is concerned with the relationship between academic dishonesty and the demographic variable year of study, namely the subgroup of Sophomore students. According to regression results, Sophomore students have a significant relationship with academic dishonesty. This is a troubling fact because Freshmen scored the lowest in terms of academic dishonesty levels. Two things could be inferred from this finding. First, international students tend to lower their standards of academic honesty after their
80 first year at the tertiary level. One reason for this could be the pressure that comes with adjusting to a new setting with higher academic expectations than high school. Another reason might be that, after a year in college, students have received the influence of friends and classmates, which could be positive or negative in terms of academic dishonesty. The second inference regarding Sophomore student’s association with academic dishonest is concerned with the university. In this sense, the university needs to uphold its high standards of academic honesty while college students go through their academic journey. Otherwise, students might believe it is not a big deal to commit plagiarism, cheat during examinations, and incur in other forms of academic dishonesty. In any case, the gap identified between Freshmen and Sophomore in terms of academic dishonesty is something to ponder upon. Recommendations Students, teachers, and administrators are the key players involved in the task of preventing academic dishonesty at the tertiary level. As such, this section gives recommendations to each one of them. Furthermore, suggestions related to further study are addressed after discussing what students, teachers and administrators can do to deal with academic dishonesty in e-learning settings where English is the lingua franca. For Students It would be beneficial for students to learn to use the library’s online resources effectively. In this sense, learning to find books and articles online is an important part of the research process, especially when it comes to preparing an essay or a term paper. Much of the stress students go through when working on a secondary-source research paper could be associated with not being able to efficiently utilize online resources. Asia-
81 Pacific International University offers free access to its online library resources, which include several full-text databases. Learning where, how, and what to search can be advantageous for international students’ academic journey. Also, students should know APIU’s academic honesty policy. The rules and regulations related to academic honesty can be found in the APIU Academic Bulletin for Undergraduate Students (pp. 41-42). Reading, understanding, and putting in practice the rules and procedures explained in the Academic Bulletin can help pupils succeed in their university journey. For Teachers The role that teachers play in preventing academic dishonesty cannot be stressed enough. In the context of the present study, four recommendations for teachers are given to help them play their part in addressing academic honesty issues. First, teachers could give assignments, projects, and even exams focused on the two-fold purpose of using technology and learning English as a second language. Involving technology use with second language acquisition will not only help students be more proficient in the target language but also teach them how technology can be a powerful tool used in benefit of their academic journey. In this sense, teachers are responsible for explaining in which ways technology should not be used, thus helping students avoid academic dishonesty. Second, based on the results of the study, teachers should explain to students what is acceptable and what is not when it comes to using language learning strategies for acquiring the English language. In other words, L2 strategies and academic honesty policies need to be clearly explained in the classroom so students have no excuse for using a friend’s help on a paper or exam as learning strategies. Teachers are responsible
82 for clearly delineating how students can rely on each other for acquiring the target language without engaging in academic dishonesty. Also, teachers should encourage students to use metacognitive strategies and evaluate which ways are most effective for their learning experience. After all, it seems that students who use metacognitive strategies and know what works for them as learners are less inclined to academic dishonesty Third, teachers should identify pupils who are struggling academically and might incur in academically dishonest behavior. According to the results of this study, Sophomores are the subgroup with a stronger association to academic dishonesty. APIU teachers should be aware of this and strive to support, guide, and provide students with sufficient information so they can make responsible decisions in their academic journey. Fourth and last, teachers could use moments dedicated to devotional thoughts to address concerns of academic honesty. To some extent, this could be a way of integrating faith and learning. As such, students would come to understand that Christianity, as a religion of morality and high ethical values, condemns plagiarism, cheating, and any other form of immoral and unethical behavior. For Administrators There are at least three recommendations for the institutional leadership at APIU. First, administrators could promote and conduct seminars among students to outline and explain the academic honesty policies given the Academic Bulletin for undergraduates. Such undertaking would be particularly helpful among Freshmen and Sophomores because of the gap found among them in terms of academic dishonesty.
83 Second, the leadership of the school would benefit from researching the possible factors that might be involved in making Sophomores the most likely subgroup to engage in academic dishonesty. Perhaps, a careful revision of the curriculum and the study load international students carry in their second year could shed some light on this issue. Third, in the long and short terms, it would be beneficial for students and teachers if the school leadership administered strong penalties to those who get involved in academically dishonest behavior. Moreover, the school could be stricter in its academic honesty policies by coming up with more severe consequences for dishonest students. For Further Study Regarding further study, several suggestions can be given. First, it would be valuable to include academic stress in the conceptual framework developed for this study. The reason for this is that previous research has associated academic dishonesty with academic stress (Maykel et al., 2018; Mi̇ ldaeni̇ et al., 2021). Moreover, although the present study involved a culturally diverse sample of international students, culture was not incorporated in the study as a variable that could predict academic dishonesty, which is something other studies have done (Payan et al., 2010; Thompson et al., 2017). Consequently, a second suggestion would to be to consider the cultural element in future research. Perhaps, including culture in the conceptual framework as a variable to consider when explaining academic dishonesty among international students could yield insightful results. Lastly, this study focused on collecting and analyzing data from only one international university. This could be considered a limitation of the study in terms of generalizability. Therefore, it would be prudent to include other international universities
84 in the sample, which could lead to more encompassing conclusions directed to the population of international students in Thailand and Southeast Asia.
85 REFERENCE LIST
86 REFERENCE LIST Abdul Rahim, A. F. (2020). Guidelines for online assessment in emergency remote teaching during the COVID-19 pandemic. Education in Medicine Journal, 12(2), 59–68. https://doi.org/10.21315/eimj2020.12.2.6 Aboagye, E., Yawson, J. A., & Appiah, K. N. (2021). COVID-19 and e-learning: The challenges of students in tertiary institutions. Social Education Research, 2(1), 1– 8. https://doi.org/10.37256/ser.212021422 Abulibdeh, E. S., & Hassan, S. S. S. (2011). E-learning interactions, information technology self-efficacy and student achievement at the University of Sharjah, UAE. Australasian Journal of Educational Technology, 27(6), Article 6. https://doi.org/10.14742/ajet.926 Abusafia, A. H., Roslan, N. S., Mohd Yusoff, D., & Mat Nor, M. Z. (2018). Snapshot of academic dishonesty among Malaysian nursing students: A single university experience. Journal of Taibah University Medical Sciences, 13(4), 370–376. https://doi.org/10.1016/j.jtumed.2018.04.003 Ahmadi, M. R. (2018). The use of technology in English language learning: A literature review. International Journal of Research in English Education, 3(2), 115–125. https://doi.org/10.29252/ijree.3.2.115 Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179–211. https://doi.org/10.1016/0749- 5978(91)90020-T Akbulut, Y., Şendağ, S., Birinci, G., Kiliçer, K., Şahin, M. C., & Odabaşi, H. F. (2008). Exploring the types and reasons of Internet-triggered academic dishonesty among Turkish undergraduate students: Development of Internet-Triggered Academic Dishonesty Scale (ITADS). Computers & Education, 51(1), 463–473. https://doi.org/10.1016/j.compedu.2007.06.003 Al-Balas, M., Al-Balas, H. I., Jaber, H. M., Obeidat, K., Al-Balas, H., Aborajooh, E. A., Al-Taher, R., & Al-Balas, B. (2020). Distance learning in clinical medical education amid COVID-19 pandemic in Jordan: Current situation, challenges, and perspectives. BMC Medical Education, 20(1), 341. https://doi.org/10.1186/s12909-020-02257-4 Alhaysony, M. (2017). Language learning strategies use by Saudi EFL students: The effect of duration of English language study and gender. Theory and Practice in Language Studies, 7(1), 18–28. https://doi.org/10.17507/tpls.0701.01 Alkhateeb, M. A., & Abdalla, R. A. (2021). Factors influencing student satisfaction towards using learning management system Moodle. International Journal of Information and Communication Technology Education (IJICTE), 17(1), 138– 153. https://doi.org/10.4018/IJICTE.2021010109
87 Almaiah, M. A., Al-Khasawneh, A., & Althunibat, A. (2020). Exploring the critical challenges and factors influencing the e-learning system usage during COVID-19 pandemic. Education and Information Technologies, 25(6), 5261–5280. https://doi.org/10.1007/s10639-020-10219-y Alqurashi, E. (2016). Self-efficacy in online learning environments: A literature review. Contemporary Issues in Education Research, 9(1), 45–52. https://doi.org/10.19030/cier.v9i1.9549 Alqurashi, E. (2019). Predicting student satisfaction and perceived learning within online learning environments. Distance Education, 40(1), 133–148. https://doi.org/10.1080/01587919.2018.1553562 Altaibekova, M. (2020). Teacher perceptions and promotion of academic honesty: A qualitative case study of humanities teachers within a diploma program (DP) at one international baccalaureate (IB) school in Kazakhstan [Nazarbayev University]. https://nur.nu.edu.kz/handle/123456789/4858 Ardasheva, Y., & Tretter, T. R. (2013). Strategy inventory for language learning-ELL student form: Testing for factorial validity. The Modern Language Journal, 97(2), 474–489. Aresgado Estidola, C., Sison Prudente, M., & Echevarria Aguja, S. (2021). Multiple attempts in online assessments address concerns in academic honesty: Perceptions of students and teachers. 2021 12th International Conference on E-Education, EBusiness, E-Management, and E-Learning, 220–224. https://doi.org/10.1145/3450148.3450178 Austin, Z., Simpson, S., & Reynen, E. (2005). ‘The fault lies not in our students, but in ourselves’: Academic honesty and moral development in health professions education—results of a pilot study in Canadian pharmacy. Teaching in Higher Education, 10(2), 143–156. https://doi.org/10.1080/1356251042000337918 Baird Jr., J. S. (1980). Current trends in college cheating. Psychology in the Schools, 17(4), 515–522. https://doi.org/10.1002/1520-6807(198010)17:4<515::AIDPITS2310170417>3.0.CO;2-3 Bandura, A. (1977). Social learning theory. Prentice-Hall. Bandura, A. (1986). Social foundation of thought and action. Prentice-Hall. Bandura, A. (1997). Self-efficacy: The exercise of control. Freeman. Bangthamai, E. (2009). The study of guideline to enhance ethics of e-learning among university students. วารสาร ICT เพื่อพัฒนาการเรียนรู้, 1(3), 1–5.
88 Basilaia, G., & Kvavadze, D. (2020). Transition to online education in schools during a SARS-CoV-2 coronavirus (COVID-19) pandemic in Georgia. Pedagogical Research, 5(4). https://eric.ed.gov/?id=EJ1263561 Basri, H., Hashim, H., & Yunus, M. M. (2019). Using Google apps as learning strategy to enhance ESL writing. Creative Education, 10(12), 2649–2657. https://doi.org/10.4236/ce.2019.1012192 Benson, L., Rodier, K., Enström, R., & Bocatto, E. (2019). Developing a university-wide academic integrity e-learning tutorial: A Canadian case. International Journal for Educational Integrity, 15(1), 5. https://doi.org/10.1007/s40979-019-0045-1 Bin-Hady, W. R. A., Al-kadi, A., Alzubi, A. A. F., & Mahdi, H. S. (2020). Assessment of language learning strategies in technology-enhanced learning environment. In ICT-based assessment, methods, and programs in Tertiary Education (pp. 83–99). IGI Global. https://doi.org/10.4018/978-1-7998-3062-7.ch005 Birks, M., Smithson, J., Antney, J., Zhao, L., & Burkot, C. (2018). Exploring the paradox: A cross-sectional study of academic dishonesty among Australian nursing students. Nurse Education Today, 65, 96–101. https://doi.org/10.1016/j.nedt.2018.02.040 Bista, K. K. (2011). Academic dishonesty among international students in higher education. To Improve the Academy, 30(1), 159–172. https://doi.org/10.1002/j.2334-4822.2011.tb00655.x Black, E. W., Greaser, J., & Dawson, K. (2008). Academic dishonesty in traditional and online classrooms: Does the “media equation” hold true? Journal of Asynchronous Learning Networks, 12(3), 23–30. Bong, M., & Skaalvik, E. M. (2003). Academic self-concept and self-efficacy: How different are they really? Educational Psychology Review, 15(1), 1–40. https://doi.org/10.1023/A:1021302408382 Bowen, N. E. J. A., & Nanni, A. (2021). Piracy, playing the system, or poor policies? Perspectives on plagiarism in Thailand. Journal of English for Academic Purposes, 51, 100992. https://doi.org/10.1016/j.jeap.2021.100992 Bravo, J. C., Intriago, E. A., Holguín, J. V., Garzon, G. M., & Arcia, L. O. (2017). Motivation and autonomy in learning English as foreign language: A case study of Ecuadorian college students. English Language Teaching, 10(2), 100–113. Broeckelman-Post, M. A. (2008). Faculty and student classroom influences on academic dishonesty. IEEE Transactions on Education, 51(2), 206–211. https://doi.org/10.1109/TE.2007.910428
89 Brownlee, B. J. (1986). Coping with plagiarism requires several strategies. The Journalism Educator, 41(4), 25–29. https://doi.org/10.1177/107769588604100406 Bylieva, D., Lobatyuk, V., Tolpygin, S., & Rubtsova, A. (2020). Academic dishonesty prevention in e-learning university system. In Á. Rocha, H. Adeli, L. P. Reis, S. Costanzo, I. Orovic, & F. Moreira (Eds.), Trends and Innovations in Information Systems and Technologies (pp. 225–234). Springer International Publishing. https://doi.org/10.1007/978-3-030-45697-9_22 Carnevale, D. (1999). How to proctor from a distance. The Chronicle of Higher Education, 46(12), 47–48. Cavico, F. J., & Mujtaba, B. G. (2009). Making the case for the creation of an academic honesty and integrity culture in higher education: Reflections and suggestions for reducing the rise in student cheating. American Journal of Business Education, 2(5), 75–88. https://doi.org/10.19030/ajbe.v2i5.4072 Chapman, K. J., & Lupton, R. A. (2004). Academic dishonesty in a global educational market: A comparison of Hong Kong and American university business students. International Journal of Educational Management, 18(7), 425–435. https://doi.org/10.1108/09513540410563130 Chen, C.-H., & Su, C.-Y. (2019). Using the BookRoll e-book system to promote selfregulated learning, self-efficacy and academic Achievement for university students. Journal of Educational Technology & Society, 22(4), 33–46. Chirikov, I., Shmeleva, E., & Loyalka, P. (2020). The role of faculty in reducing academic dishonesty among engineering students. Studies in Higher Education, 45(12), 2464–2480. https://doi.org/10.1080/03075079.2019.1616169 Chudzicka-Czupała, A., Grabowski, D., Mello, A. L., Kuntz, J., Zaharia, D. V., Hapon, N., Lupina-Wegener, A., & Börü, D. (2016). Application of the theory of planned behavior in academic cheating research: Cross-cultural comparison. Ethics & Behavior, 26(8), 638–659. https://doi.org/10.1080/10508422.2015.1112745 Chyung, S. Y., Moll, A. J., & Berg, S. A. (2010). The role of intrinsic goal orientation, self-efficacy, and e-learning practice in engineering education. Journal of Effective Teaching, 10(1), 22–37. Click, A. (2012). Issues of plagiarism and academic integrity for second-language students. MELA Notes, 85, 44–53. Coffey, S., & Anyinam, C. (2012). Trialing a contextual approach to academic honesty. Nurse Educator, 37(2), 62–66. https://doi.org/10.1097/NNE.0b013e3182461b5c Cohen, A. D. (2011). Strategies in learning and using a second language (1st ed.). Taylor & Francis.
90 Cole, M. T., Shelley, D. J., & Swartz, L. B. (2019). In relaunching a new vision in education and e-learning: Fostering a culture of academic integrity in e-learning. In V. L. Uskov, R. J. Howlett, L. C. Jain, & L. Vlacic (Eds.), Smart Education and e-Learning 2018 (pp. 151–164). Springer International Publishing. https://doi.org/10.1007/978-3-319-92363-5_14 Conzett, J., Martin, M., & Mitchell, M. (2010). Proactively addressing plagiarism and other academic honesty issues with second-language writers. Writing and Pedagogy, 2(2), 293–309. https://doi.org/10.1558/wap.v2i2.293 Costley, J. (2019). Student perceptions of academic dishonesty at a cyber-university in South Korea. Journal of Academic Ethics, 17(2), 205–217. https://doi.org/10.1007/s10805-018-9318-1 Cuadrado, D., Salgado, J. F., & Moscoso, S. (2019). Prevalence and correlates of academic dishonesty: Towards a sustainable university. Sustainability, 11(21), 1– 20. https://doi.org/10.3390/su11216062 Curtis, G. J., & Clare, J. (2017). How prevalent is contract cheating and to what extent are students repeat offenders? Journal of Academic Ethics, 15(2), 115–124. https://doi.org/10.1007/s10805-017-9278-x de Fátima Goulão, M. (2014). The relationship between self-Efficacy and academic achievement in adults’ learners. Athens Journal of Education, 1(3), 237–246. Deckert, G. D. (1993). Perspectives on plagiarism from ESL students in Hong Kong. Journal of Second Language Writing, 2(2), 131–148. https://doi.org/10.1016/1060-3743(93)90014-T Deeming, P., & Johnson, L. L. (2009). An application of bandura’s social learning theory: A new approach to deafblind support groups. Journal of the American Deafness & Rehabilitation Association (JADARA), 42(4), 203–209. Diep, A.-N., Zhu, C., Struyven, K., & Blieck, Y. (2017). Who or what contributes to student satisfaction in different blended learning modalities? British Journal of Educational Technology, 48(2), 473–489. https://doi.org/10.1111/bjet.12431 DiPietro, M. (2010). Theoretical frameworks for academic dishonesty: A comparative review. To Improve the Academy: A Journal of Educational Development, 28(1), 250–262. http://dx.doi.org/10.3998/tia.17063888.0028.018 Doménech-Betoret, F., Abellán-Roselló, L., & Gómez-Artiga, A. (2017). Self-efficacy, satisfaction, and academic achievement: The mediator role of students’ expectancy-value beliefs. Frontiers in Psychology, 8. https://doi.org/10.3389/fpsyg.2017.01193 Dörnyei, Z., & Ryan, S. (2015). The psychology of the language learner revisited. Routledge.