ABSTRACT

First published in 2001. Routledge is an imprint of Taylor & Francis, an informa company.

chapter |7 pages

REFERENCE

chapter |12 pages

thermore, a formal code is not the only way to achieve the desired result. As sug-gested earlier, a strong culture of academic integrity can exist at an institution that has no formal code but communicates the importance the community places on in-tegrity in other ways. McCabe, Treviño, and Butterfield's (1996) study of 318 alumni of two private liberal arts colleges suggested honor codes can have long-term effects on behavior. The study focused on alumni who had graduated from their respective colleges be-tween 1962 and 1989, allowing the researchers to test hypotheses about the long-term effects of collegiate honor codes as well as the effect of codes of ethics at their current work organizations. The results supported previous work by show-ing that dishonest behavior in the workplace can be reduced by an organizational code of ethics. The results also show that dishonest behavior in the workplace var-ies inversely with the strength of implementation of an organizational code of eth-ics (i.e., the degree of managerial commitment to the code and the degree to which an organization attempts to communicate its code to employees and to ensure com-pliance) and the degree to which a code of ethics is deeply embedded in the organi-zation's culture (i.e., the degree to which the code is understood and accepted by employees and guides their day-to-day interactions and activities). The results also indicate that college honor codes can have an enduring effect: Dishonest behavior in the workplace was lowest for participants who had experienced an honor code environment in college and who currently worked in an organization that had a strongly implemented code of ethics. Overall, this work suggests that participation in multiple honor code communities can play a part in reducing dishonest behav-ior, particularly if the honor codes are well implemented and strongly embedded in the organizational culture. Faculty Views of Academic Integrity Policies Faculty members' views of academic integrity policies, and how these views differ across code and noncode schools, was the subject of McCabe's (1993) study of 800 faculty at a geographically diverse sample of 16 U.S. colleges and universities. This study showed that faculty at code schools were more likely to rate their school higher than noncode schools on factors such as students' understanding of aca-demic integrity policies, faculty support of these policies, and the overall effective-ness of the policy. Faculty at code schools were also more likely to believe that stu-dents should play a significant part in the judicial process associated with academic cheating. This study also revealed that faculty at both code and noncode schools are reluctant to report cheating and prefer to handle suspected cases of cheating on their own rather than appeal to institutional policies and procedures. Furthermore, this study confirmed student perceptions that many faculty do not treat cases of aca-demic dishonesty very harshly. For example, more than half of the noncode faculty reported that their most likely reaction to an incident of cheating would be failure

chapter |1 pages

cheating behavior among college students. However, unlike most studies with col-lege students, these factors were related to actual cheating frequencies across the multiple courses that students took during a target semester. METHOD Participants Participants attended a small, private liberal arts college that has had a formal honor code in effect since 1965. Anonymous surveys were mailed to a random selection of 25% of the student body in the spring semester. One hundred seventy-five stu-dents (representing approximately 9% of the student body) completed and returned the surveys (11 additional surveys were returned but were unusable), yielding a re-turn rate of 35%. Women were slightly overrepresented in the sample, at 68%, compared to 51% in the college. Participants were predominately White (90.3%). All class years were represented (26% of the sample were lst-year students, 22% were sophomores, 19% were juniors, and 33% were seniors). Measures Cheating rates. Participants reviewed 17 different cheating behaviors and indicated how many times they engaged in each behavior during the previous se-mester. The behavior list was a modified version of lists used by Gardner and Melvin (1988), Newstead et al. (1996), and Sutton and Huba (1995). It included a range of violations, such as copying from another student's exam, plagiarism, and inventing laboratory data. However, in contrast to previous studies, participants in this study reported cheating behaviors course by course. Thus, if a participant was enrolled in four courses during the target semester, the participant filled out the sur-vey four times, once for each course (to protect identities, department areas, not course names, were requested on the survey). In addition, participants indicated the frequency of each behavior by course. Motivation. Measures of mastery and extrinsic motivation were adapted from scales used by Midgley et al. (1998) and Anderman et al. (1998). These scales included measures of personal mastery motivation, personal extrinsic motivation, course mastery motivation, and course extrinsic motivation. The original scales were worded for middle school students and specified a particular subject (English or science). Our version replaced the subject indicator with a more generic descriptor, such as "course," and replaced the word "teacher" with "professor." As with the list of cheating behaviors, participants filled out a motivation scale for each course taken in the previous semester. Response options ranged from 1 (strongly disagree) to 5 (strongly agree).

chapter |5 pages

Perceived social norms. Participants also responded to items soliciting their perception of cheating by other students at the college. Specifically, partici-pants were asked to estimate the frequency of cheating behavior by friends, by members of their residence hall and, more generally, by students at the college. Attitudes. Three items measured the extent to which participants believed cheating in college is sometimes justified, is justified to pass a course (to stay in school or to graduate), or is justified if a close friend asks for help. Knowledge. A final set of items asked participants about their knowledge of the honor system used at the college. These items were designed to measure the ex-tent to which students had received, read, and understood the colleges' honor code. RESULTS Cheating Rates For purposes of this study, participants were classified as cheaters if they responded affirmatively to any of the 17 cheating behaviors in any course during the target se-mester. Noncheaters were those who reported no cheating in any course during the same period. In our sample, 31.4% reported cheating on at least one paper or exam during the target semester, and an additional 23.5% cheated in other contexts, such as on homework or laboratory work. Thus, 54.9% of participants were classified as cheaters and 45.1% as noncheaters. Demographics. Initial analyses indicated that cheaters and noncheaters were not distinguishable based on program area of study, gender, or GPA. These variables were dropped from further analysis. In addition, the numbers of students in each class who cheated were not reliably different. However, differences by class year were found for the number of cheating acts committed. A one-way analysis of variance (ANOVA) was conducted with four levels of class year (lst-4th year) as the independent variable, and cheating score as the dependent variable. The cheating score represented the total number of times a student cheated in the target semester. This analysis was statistically sig-nificant, F(3, 170) = 5.64, p = .01. Post hoc comparisons of class year using the Tukey approach indicated that the mean cheating score for lst-year students (5.31) was significantly more than mean cheating score for juniors (1.07) or seniors (2.27). The mean cheating score for sophomores (4.12) was not significantly dif-

chapter |7 pages

Among these, knowledge of institutional policy was the best predictor of cheating rates, followed by mastery motivation and attitudes about cheating. This finding is important because it reminds college administrators that having an honor system, rules about cheating, or both, and effectively administering that system to students, are not one and the same. Participants in this study who cheated reported signifi-cantly less understanding of institutional policy regarding cheating than did noncheating participants. In addition, this study advances our understanding of the role of motivation in cheating behavior. In particular, mastery and extrinsic motivation do not appear to be uniform across all courses, and this variability is related to cheating. Participants who cheated had lower mastery motivation and higher extrinsic motivation in the courses in which they cheated than in courses in which they did not cheat. Cheaters, in courses in which they cheated, also differed on these two motivation variables from noncheaters. Mastery motivation was lower for cheaters, and extrinsic motiva-tion higher, as compared to noncheaters. However, in courses in which cheaters did not cheat, cheater and noncheater scores were not reliably different. Thus, cheaters reported increases in extrinsic motivation and simultaneous decreases in mastery motivation, but only in courses in which they cheated. Anderman etal. (1998) found a similar relation between mastery and what they called performance goals with middle school students. This study confirms that motivation is an important factor in the cheating behavior of college students as well. Finally, as expected, cheating behavior was related to perceptions of the behav-ior of peers and to attitudes about cheating. Both factors were significant predic-tors of cheating rates. Cheaters believed that more students engaged in cheating behaviors than did noncheaters. Cheaters also justified cheating behavior to a greater extent than did noncheaters. In addition, the more the participants cheated, the higher their estimate of cheating on campus. These findings replicate previous research on the importance of student attitudes and peer norms for understanding, and perhaps influencing, cheating behavior (see Whitley, 1998, for a review). Of course, these results do not indicate causality. Once students begin to cheat, their motivation levels and their perceptions and experiences of others may change. Alternatively, perhaps low mastery motivation in a course increases a stu-dent's risk for cheating in that course and increases the cheater's tendency to cheat repeatedly. A student who is uninterested in a course may look for ways to com-plete the course with the least effort. In addition, high extrinsic motivation may also increase student vulnerability to temptations to cheat. If a student's purposes for taking a course have little to do with the course and more to do with extrinsic goals, such as grades or career opportunities, cheating may serve those goals. In ei-ther case, motivations appear to be course specific. With respect to perceived norms, cheaters may estimate higher rates of cheating by others, as compared to noncheaters, as a way of preserving their self-image (e.g., false consensus effect; Ross, Green, & House, 1977). They also may be more

chapter |3 pages

substantial numbers of students cheat not just once, but repeatedly (Hollinger & Lanza-Kaduce, 1996; McCabe & Treviño, 1995; Moffatt, 1990). Given the prevalence of academic dishonesty among college students, it is not surprising that considerable research has been conducted on its causes and corre-lates, with more than 100 studies having been published on the topic during the past 3 decades (Whitley, 1998). What is more surprising is the relative lack of at-tention that researchers have paid to gender differences in academic dishonesty given the important role gender plays in theories of in moral reasoning (e.g., Lapsley, 1996). Theorists such as Chodorow (1989) and Gilligan (1982) proposed that differential childhood socialization processes lead to different moral reason-ing orientations in men and women. These theorists proposed that gender differ-ences in moral orientation result in gender differences in behavior, with women being less likely than men to violate social norms because of the negative effects that such violations could have on other people and the potential of such violations to impair fulfillment of women's nurturant role obligations (Robbins & Martin, 1993). Thus, women are less likely than men to engage in minor criminal behavior (e.g., Tibbetts & Herz, 1996), excessive alcohol consumption (e.g., Robbins & Martin, 1993), and unprovoked aggression (e.g., Bettencourt & Miller, 1996). One would therefore expect to find similar gender differences in violations of academic integrity norms, especially given that engaging in academic dishonesty is corre-lated with engaging in other forms of minor deviance (Blankenship & Whitley, 2000; Whitley, 1998). However, in a meta-analysis of research on gender differences in cheating, Whitley, Nelson, and Jones (1999) found a mean difference of only 0.2 standard deviations between men's and women's self-reports of having cheated. Although this difference was statistically significant due to the large cumulative sample size in the meta-analysis, in absolute terms it just met Cohen's (1992) criterion for a nontrivial effect size. In contrast, Whitley et al. found a mean gender difference of about 0.5 standard deviations for attitudes toward cheating, with women reporting more negative attitudes. Thus, women hold more negative attitudes toward cheat-ing than do men but are about equally likely to cheat. Cognitive dissonance theory holds that such attitude-behavior inconsistencies lead to a negative emotional state called cognitive dissonance (Elliot & Devine, 1994; Festinger, 1957). Because, on the average, women hold more negative atti-tudes toward cheating than do men, one would expect women who cheat to experi-ence more cognitive dissonance and so to have more negative affective reactions to having cheated than would men. Although little research has been conducted on such gender differences, Smith, Ryan, and Diggins (1972) found that women re-ported experiencing more guilt over having cheated than did men. In addition, Tibbetts (1997) found that women reported more shame concerning intentions to cheat, and Cochran, Chamlin, Wood, and Sellers (1999) found that women re-ported expecting to feel more shame and embarrassment if they cheated. Further-

chapter |10 pages

ley, 1999). The impetus for understanding the underlying dynamics of dishonest behavior among students stems from the conviction that, apart from assuming the role of an educational and credentialing agency, the primary focus of an academic institution is to provide an environment for personal development of our youth in the moral, cognitive, physical, social, and aesthetic spheres. An atmosphere that promotes academic honesty and integrity is a precondition for generating, evaluat-ing, and discussing ideas in the pursuit of truth, which are at the very heart of aca-demic life. Research has shown that dishonesty in college, cheating in particular, is a predic-tor of unethical behavior in subsequent professional settings (e.g., Sierles, Hendrickx, & Circel, 1980). More recently, Sims (1993) also found academic dis-honesty to be significantly related to employee theft and other forms of dishonesty at the workplace. Sim's findings suggest that people who engaged in dishonest behav-iors during their college days continue to do so in their professional careers. Further-more, Sim's findings indicate that people who engaged in dishonest behaviors during college are more likely to commit dishonest acts of greater severity at work. Existing research on academic dishonesty has largely been conducted in Eu-rope and North America. The results of these studies suggest that a large percent-age of university students indulge in some form of cheating behaviors during their undergraduate studies (e.g., Newstead, Franklyn-Stokes, & Armstead, 1996). Sur-vey findings also suggest that not only is student cheating pervasive, it is also ac-cepted by students as typical behavior (e.g., Faulkender et al., 1994). Although the research conducted in the Western context has increased our under-standing of academic dishonesty among students, the relevance of these results to the Asian context is questionable. Differences in sociocultural settings, demo-graphic composition, and specific educational policies may render some compari-sons meaningless. Different colleges also vary widely in fundamental ways, such as size, admission criteria, and learning climate. These factors render the comparabil-ity of results obtained from different campuses difficult. Cross-cultural studies con-ducted to examine students' attitudes toward academic dishonesty have found evidence that students of different nationalities and of different cultures vary signifi-cantly in their perceptions of cheating (e.g., Burns, Davis, Hoshino, & Miller, 1998; Davis, Noble, Zak, & Dreyer, 1994; Waugh, Godfrey, Evans, & Craig, 1995). For example, in their study of U.S., Japanese, and South African students, Burns et al. found evidence suggesting that the South Africans exhibited fewer cheating behav-iors than the Americans but more than the Japanese at the high school level. How-ever, at the college level, the cheating rates for South African students were lower compared to both their American and Japanese counterparts. In another cross-national study on academic dishonesty, Waugh et al. (1995) examined cheating behaviors and attitudes among students from six countries (Australia, the former East and West Germany, Costa Rica, the United States, and Austria) and found significant differences in their perceptions of cheating. Stu-

chapter |2 pages

as you may be the

chapter |4 pages

LIM AND SEE REFERENCES

chapter |2 pages

In summary, the academic integrity literature suggests a possibility that SEPIB may have broader reaching effects than the traditionally examined learning out-comes and may actually influence academic integrity decisions. Based on the SEPIB and academic integrity literature, I hypothesized that students who commit academic dishonesty will have lower evaluative perceptions of their instructors than students who have not committed academic dishonesty. METHOD Participants The respondent pool, taken from 1,390 students, were students enrolled in a variety of courses at a large (approximately 18,000 undergraduate students) regional col-lege in the West. Of the original pool, 1,369 had participated in a test or other type of graded assignment when the survey was conducted, thereby having had an opportu-nity to have committed acts of academic dishonesty. Of these 1,369 students, 281 (20.5%) reported committing an act of academic dishonesty in the course in which the survey was conducted. Ages of the students ranged from 16 to 65, with the majority (70.2%) being 18 to 22 years of age. The sample included 564 men (43.3%) and 737 (56.6%) women. All college class levels were represented: 26.0% were lst-year students, 20.4% were sophomores, 28.3% were juniors, 23.7% were seniors, and 0.8% were graduate students. In an open-ended response question, students reported 38 dif-ferent majors. Measures Academic integrity. Items regarding types of academic dishonesty were compiled from previous surveys (with particular reliance on Barnett & Dalton, 1981; Bowers, 1964; Jendrek, 1992; Oaks, 1975; Singhal, 1982; Stevens, 1984; Wright & Kelly, 1974). Nineteen types of dishonest acts were selected and divided into two categories: tests or quizzes (see Table 1) and other activities (see Table 2). A determination of whether a student committed academic dishonesty in the course in which the survey was administered was based on whether the student responded yes to any of the 19 items listed in Tables 1 and 2. If students responded yes to a spe-cific behavior, they were also asked to indicate how frequently they had engaged in this behavior in the course. All students who answered no to all of the items were placed in the "not admitting to academic dishonesty in the course" category. SEPIB. Because SEPIB was assessed using four 5-point scales ranging from 1 (good, fair, worthless, negative) to 4 (bad, unfair, valuable, positive; see

chapter |1 pages

the semester. I approached faculty mem-

chapter |6 pages

A study of 207 graduate business students found that 80% had engaged in at least one of 15 unethical academic practices as a graduate student (Brown, 1995). These students also perceived themselves as more ethical than their undergraduate counterparts, although they had similar rates of academic dishonesty. Because the research conducted among various graduate schools has been lim-ited in scope, we systematically investigated the definition, prevalence, perceived prevalence, and severity of, as well as justifications for and expected responses to, academic dishonesty at the graduate level using the same approach as LaGrange (1992). These issues were assessed and compared from the perspectives of students and faculty representing multiple disciplines within the university. Students', fac-ulty members', and administrators' ideal and realistic expectations of how cheating would be handled were also examined. Finally, the relation between academically dishonest behavior and student demographic variables was examined. METHOD Procedure A sample of students, faculty, and administrators at the graduate and professional school level was obtained from a large, private, religiously affiliated Midwestern university. All programs were invited to participate in this research, and 22 pro-grams agreed to participate. Students, faculty, and administrators all received pack-ets that contained a recruitment letter, a survey, two answer sheets, and an envelope for returning the survey via intercampus mail. Respondents were asked to return the surveys unmarked if they did not want to complete them. Surveys were placed in the campus mailboxes of 2,669 graduate students. One department did not have mailboxes and consequently 83 surveys were distributed via U.S. mail. Surveys were distributed to 387 faculty and 50 administrators via intercampus mail. Participants Survey instruments were sent to 2,752 students, with 246 students returning com-pleted surveys for a return rate of 8.9%. The student group is made up of students representing all year levels, working toward a variety of graduate degrees (MA, MS, JD, MD, and PhD), and representing the social sciences, natural sciences, hu-manities, health sciences, nursing, law, and medicine. Survey instruments were sent to 387 faculty, with 49 faculty returning com-pleted surveys for a return rate of 12.6%. The faculty sample was 61.2% men, and included 34 (69.4%) tenured faculty and 15 (30.6%) nontenure-track faculty. Sev-enty-seven percent of the faculty were either associate or full professors. The ma-

chapter |5 pages

faculty. Follow-up univariate F tests revealed 14 items that differed significantly at the 0.05 level using the Scheffé test. For a number of the items, it appears that there is statistical significance and not necessarily meaningful differences. For ex-ample, the item concerning taking a test for another student was rated 3.97 and 3.84 by students and faculty, respectively, yet was different in statistical signifi-cance. Of the items that are significantly different, 3 stand out as being meaning-ful. Items pertaining to studying from old versions of exams and having the instructors manual that contains test items were rated more severe by students (Ms = 2.75 and 3.53, respectively) than faculty (Ms = 1.64 and 2.91, respectively), whereas faculty rated collaborating on work that was supposed to be done individ-ually as more severe (M = 3.05) than did students (M = 2.70). A broad definition of perceived prevalence of cheating was established by ask-ing faculty and students to rate the percentage of students they believed cheat using a Likert scale with 1 being 0% and 10 being 90% to 100%. Faculty perceived that between 0% to 10% of students cheat (M = 1.89), whereas students perceived be-tween 10% to 20% of students cheat (M = 2.32), a difference that is significant, i(271) = 4.27, p<. 001. Table 1 presents the perceived prevalence results using a narrow definition of the term based on the 40 academically dishonest items. As can be seen from Table 1, both faculty and students perceive that the prevalence of cheating is quite low, with most item ratings having means below 2.00, indicating that 1 % to 10% of stu-dents are perceived as engaging in the behaviors. The behaviors perceived to be most prevalent by faculty were students using old tests without permission, whereas students perceived changing words slightly from an original source as the most prevalent. Although the perception of these behaviors is relatively low, there is greater variance (standard deviations typically above 1.0) in the ratings of per-ceived prevalence when compared to the severity ratings. In general, students have greater variance than faculty in their perceived prevalence ratings. A MANO VA was computed to determine whether significant differences ex-isted between student and faculty ratings of perceived prevalence. Results re-vealed a significant MANO VA, Wilks's A = .70, F(40, 247) = 2.69, p < .001. Follow-up univariate F tests revealed 23 items that differed significantly at the .05 level. Students had higher perceived prevalence ratings on 22 of the 23 items, with faculty rating the use of old tests without permission as more prevalent than stu-dents. Similar to the results regarding severity ratings, many of the significant dif-ferences obtained on perceived prevalence ratings, although statistically significant, do not appear to be meaningful differences. For example, although the item concerning using unauthorized materials such as crib notes is statistically sig-nificant with a student mean of 1.55 and a faculty mean of 1.34, this difference is not a meaningful one.

chapter 320|1 pages

ROIG represent actual writing practices, then the issue of how paraphrasing is defined within and across disciplines needs to be seriously considered. Most modern manuals of writing that discuss the parameters of correct para-phrasing (e.g., Aaron, 1998; Hacker, 1994; Nadell et al., 1994; Troyka, 1999) are relatively clear on the extent to which the original material must be modified to be considered properly paraphrased. For example, Troyka stated, "Even though a paraphrase is not a direct quotation, you must use DOCUMENTATION to credit your source. Also, you must reword your source material, not merely change a few words" (p. 498). In another manual, Aaron suggested that when paraphrasing a source, "Restate the source's ideas in your own words and sentence structures" (p. 257). Other writers offer even stricter definitions of paraphrasing. For example, consider Howard's (1993; cited in Howard, 1999) definition of patchwriting, a form of writing that she considers plagiarism: "copying from a source text and then deleting some words, altering grammatical structure, or plugging in one-for-one synonym-substitutes" (p. 89). Accordingly, paraphrases such as those supplied by respondents who appropriated strings of words in Studies 2 and 3 are certainly not within the guidelines of paraphrasing outlined by these manuals and would there-fore constitute possible instances of plagiarism. If indeed college professors paraphrase in a manner similar to that observed in the studies in this article, can we conclude that a small but significant proportion of writing by college professors may be classified as plagiarism? Obviously, such a conclusion would depend on a number of factors. For example, has text appropria-tion occurred systematically across various works cited throughout the paper, or is it confined to one or two instances of, say, a description of a complex methodology section of an experimental research report? In addition, how many strings of con-secutive words have been appropriated and of what length are these strings of text? Clearly, these and other issues need to be taken into account when making a deter-mination of plagiarism. One important factor to be considered when reviewing others' work for potential plagiarism is the discipline of the writer. For example, if the writer is a psychology student or professor, then perhaps those paraphrases might be acceptable within the psychology community. Consider how the psychology profession defines para-phrasing and plagiarism. The Publication Manual of the American Psychological Association (APA; APA, 1994), a source used by most psychologists and others in the social sciences (e.g., sociologists, social workers), offers the following guide-lines: "Summarizing a passage or rearranging the order of a sentence and changing some of the words is paraphrasing" (p. 292). A comparison of the Publication Man-ual definition with the definitions of traditional writing manuals outlined earlier in-dicates some obvious differences in the extent to which text should be modified to be considered a proper paraphrase. Unfortunately, the absence of a general operational definition for paraphrasing leaves plenty of room for disagreement as to when a para-phrase might be considered an instance of plagiarism.

Registered in England & Wales No. 3099067
5 Howick Place | London | SW1P 1WG© 2024 Informa UK Limited