Origins of Academic Dishonesty

8/7/2019 Origins of Academic Dishonesty

http://slidepdf.com/reader/full/origins-of-academic-dishonesty 1/24

OrigisofAcademicDishoesty:EthicalOrietatiosadPersoality

FactorsAssociatedwithAttitudesaboutCheatigwithIformatioTechology

StephanieEtter Mt. Aloysius College

JackieJ.CramerUniversity o Pittsburgh, Titusville

SethFinnRobert Morris University

Abstract Growing perceptions that students exploit inormation technology to evade academic assign-ments prompted surveys o student attitudes about unethical uses o inormation technology (e.g., cutting and pasting excerpts rom Web sites without attribution) at two institutions.Students at a private church-afliated college rated cheating behaviors as more oensive thantheir counterparts at a regional campus o a major research university. However, ordinal rankings o academically dishonest behaviors at both institutions were surprisingly similar (rho = .90). Further, students who rated such behaviors as being more serious, typically valued idealism, the ethical principle o doing no harm to others, and disapproved o highsensation-seeking activities involving alcohol, drugs, and sex. (Keywords: academic integrity,cheating, disinhibition, EPQ, ethics, inormation technology, sensation-seeking, technology acceptance model, TAM.)

InTRODUCTIOn

Many educational experts have long anticipated that computer technology would serve as a catalyst or changes in teacher practice Bull, Knezek, Roblyer,

Schrum, & Tompson, 2005; Dexter, Anderson, & Becker, 1999. O equalimportance more recently, however, has been its perceived eect on studentpractices that threaten academic integrity. or example, an ongoing survey con-ducted by the Center or Academic Integrity has noted a our-old increase 10-40% over the past fve years in the number o college students who have usedthe Internet to construct papers based on unattributed text excerpts rom online Web sites McCabe, 2005. urther, a Pew study Levin & Araeh, 2002 ocus-ing on the discrepancy between teachers’ and students’ Internet competence,reported that Internet-savvy high school students, who were critical o their

instructors’ reticence in using the Web to enhance learning, also “admit[ted]to knowing students who plagiarized Internet resources or use[d] other onlinetools to cheat outright” Levin & Araeh 2002 p 11 Although the public

JOURNAL O RESEARCH ON ECHNOLOGY IN EDUCAION, 39 (2), 133–155



Large multi-institution studies o students’ attitudes toward cheating haveestablished the critical role that situational actors, such as peer behavior andpeer disapproval, have on student dispositions about cheating (Bowers, 1964;Leming, 1980; McCabe & revino, 1993, 1997), yet the newest environmentalactor, the diusion o inormation technology, has escaped systematic study. A cursory review o situational actors that acilitate acts o academic dishonesty suggest that inormation technology has lowered barriers to cheating. Previousresearch has shown that cheating occurs when opportunities are enhanced (Mi-chaels & Miethe, 1989; Perry, Kane, Bernesser, & Spicker, 1990), surveillancecan be avoided (Concoran & Rotter, 1987; Covey, Saladin, & Killen,1989),chances or success have been improved (McCabe & revino, 1993), and risk o punishment is lowered (Leming, 1980). All are consistent with anecdotalevidence about how inormation technology has enabled students to engage inacademically dishonest behaviors.

PreviousResearchonStudentCheatingStudies o academic dishonesty among college students date back 70 years or

more. Te two most recent reviews o empirical research on student cheating(Crown & Spiller, 1998; Whitley, 1998) were undertaken in the mid-1990s pri-or to the Internet explosion. Nevertheless, to the extent that a new technology is rst adopted to perorm conventional activities in an expeditious mode, thesetwo reviews set useul ground rules or systematically studying the infuenceo inormation technology on student attitudes and behaviors. Both reviews

sought to separate studies o academic dishonesty into just two domains—stud-ies o situational actors, which we have already alluded to, and studies o indi-vidual actors, which identied sociological or psychological characteristics ascorrelates o cheating.

Over the 25-year period (1970–1995), which Crown and Spiller (1998) re-viewed, the most notable development they reported about individual actors was that gender dierences appeared to have attenuated over time as sex-rolesocialization o male and emale students converged (Ward & Beck, 1990).However, Whitley and his colleagues in a ollow-up meta-analysis (Whitley,

Bichlmeier Nelson, & Jones, 1999) and then Whitley (2001) himsel ocusedon gender dierences alone, nding that women students demonstrated signi-cantly more negative attitudes towards cheating than male students even thoughreported requencies o cheating behavior or both sexes were nearly the same. Among other commonly studied demographic variables, cited by Crown andSpiller (1998), students with lower GPAs and business majors were ound tocheat more (McCabe & revino, 1995), but studies o age and class standinggenerated inconsistent results. Among the most requently tested personality variables, external locus o control and moral obligation were linked to

cheating.In his review, Whitley (1998) covered nearly the same 25-year period, but



“hold[ing] avorable attitudes towards cheating” (p. 23)—which exhibited largeeects in a minimum o fve studies. As perceptions and attitudes, they both allin line as individual actors. Similarly, when it came to actors that exhibitedmoderate eects in relation to cheating, individual actors alone—age (young-er), course task perormance (worse), and deviant behavior (more)—weredemonstrated in sucient studies to make a convincing case. Whitley’s resultsserve then, as do Crown and Spiller’s (1998) observations, as a prelude to recentstudies in which individual actors are more thoroughly studied than situationalones.

In act, in reviewing the empirical literature on college cheating over the last10 years, research on situational actors has been narrowly ocused on the e-cacy o instituting honor codes to reduce academic dishonesty (Brown & How-ell, 2001; McCabe & revino, 1997, 2002; McCabe, revino, & Butterfeld,2001; Zabihollah, Elmore, & Szendi, 2001) along with an assortment o relatedtopics, such as social acceptability (Smyth & Davis, 2003; Strike & Moss,1997), probability o being caught (Buckley, Wiese, & Harvey, 1998), andcross-cultural dierences (Salter, Guey, & McMillan, 2001). As yet, however,studies describing the situational opportunities or cheating aorded by digitaltechnology (Auer & Krupar, 2001; Campbell, Swit, & Denton, 2000; Ross,2005, Szabo, 2004) have not attempted to enumerate the ull range o unethicalactivities that students are aware o or their perceptions about these behaviors.Tis defciency is all the more important because studies o traditional orms o academic cheating suggest that students are generally conused about what con-stitutes plagiarism and other questionable short-cuts to completing academic as-signments (Allmon, Page, & Roberts, 2000; Roig, 1997; Roig & Deommaso,1995). Other concerned academics have challenged the notion that technologi-cal remedies can address new modes o cheating (McLaerty & oust, 2004;ownley & Parsell, 2004).

aced with a moving situational target, then, it is understandable that mucho the recent research is oriented toward the interace between acceptable aca-demic behavior and individual actors, primarily psychological measures. orexample, the study by Buckley, Wiese, & Harvey (1998), previously cited or itssingle situational actor—probability o being caught—measured fve individualactors o which aggression/hostility and male gender were linked to unethicalbehavior. Whitley (1998) too apparently ollowed up on one o his reportedmoderate eects—deviant behavior. Blankenship and Whitley (2000) reportedthat minor orms o deviance, such as engaging in risky driving behaviors andbeing an unreliable riend or worker, were linked to cheating on exams or mak-ing alse excuses to avoid taking exams. Wryobeck and Whitley (1999) exam-ined peer perceptions o cheaters and their accomplices, fnding that students with a high orientation towards learning would recommend a more severe pun-ishment while students with a high orientation towards grades were more likely to emulate the cheater’s and accomplice’s actions. inally, Caruana, Ramase-



PreviousResearchonEthicsUsingStudentSamplesInsight into college cheating has also beneted rom new approaches used in

general studies o ethical behavior in which the interace between individualactors and behaviors (or attitudes about them) are assumed to be mediated by ethical philosophies. Although developed more than 25 years ago, researcherscontinue to use orsyth’s (1980) Ethical Position Questionnaire (EPQ) to pro-vide a measure o ethical ideology on two orthogonal scales labeled idealism andrelativism. orsyth conceived o these two concepts as the essential componentso human ethical principles. Idealism refected the notion that ethical humanbehavior requires one to “do no harm.” Relativism refected the notion that within any diverse group o individuals ethnical norms may span a continuumrom a set o absolute rules to completely ad hoc situational determinations.

Allmon, Page, and Roberts (2000) used the EPQ in a study o attitudes aboutclassroom cheating that included age, gender, religiosity, country o origin, andthe perception and judgment dimensions o the Myers Briggs ype Indicator(Myers & McCaulley, 1985) as predictors. While Allmon et al. (2000) weresurprised to nd that increasing age was overwhelmingly the best predictor o negative attitudes towards two orms o classroom cheating, “getting a classmateto write a term paper” or “do the work on a computer project,” increasing age was also related to lower scores on orsyth’s (1980) relativism actor. Te lack o any direct association between the EPQ and ratings o cheating behavior inthis study, however, may be attributed to peculiarities in Allmon et al.’s (2000)application o orsyth’s (1980) ourold typology as an analytical device. Davis, Andersen, and Curtis (2001) have, in act, argued against implementing thetypology, which creates our categories by variously grouping high and low scor-ers on the idealism and relativism scales, precisely because the loss o statisticalinormation may articially attenuate otherwise signicant relationships.

In Davis et al.’s (2001) own study, which used conrmatory actor analysis tovalidate EPQ’s psychometric properties, the idealism and relativism scales wereound to be correlated, respectively, with similar constructs that measure empa-thy and dogmatism. Te idealism scale, in particular, was shown to be a signi-cant predictor o ethical judgments in ve dierent scenarios. As Davis and hiscolleagues (2001, pp. 42-43) described the results, “Subjects high in idealism were morally opposed to actions potentially harmul to others.” Relativism wasound to play primarily a mediating role in the relationship between idealismand the ormation o moral judgments.

As in the Davis et al. (2001) study, the EPQ has been used in a number o studies o ethical disposition outside the domain o college cheating but withcollege student samples nevertheless. McIntyre, Capen, Minton (1995) oundthat cognitive style dimensions o the Meyers-Briggs ype Indicator infuencedthe EPQ measures o idealism and relativism, with a direct link between rela-tivism and the acceptance o ethically questionable decisions. Barnett, Bass,and Brown (1996) ound that students who scored high in idealism and low



ences in idealism and relativism scores. inally, Wilson (2003) ound a negativecorrelation between social dominance orientation and idealism, leading him toconclude that so-called “ruthless” individuals, when seeking their personal goals,can be indierent to moral issues.

AnExploratoryStudyofAcademicallyDishonestUsesofInformationTechnology

Each o these studies, taken on its own, establishes interesting insights andpossibilities about the origins o academic dishonesty. Unortunately, though,none is so theoretically powerul or closely aligned with the domain o digitaldishonesty among college students that it oers an ideal approach to the study o why computers are used to circumvent the learning process. Tus, or guid-ance in structuring this exploratory study, we turned to the technology acceptance model (Davis, 1989), a mainstay o inormation systems research. Te essentially

linear AM model was devised to trace back to their origins the actors thatlead to user acceptance o inormation technology. As unorthodox as it may seem, deciding to exploit a orm o inormation technology to evade academiceort, even when unethical, diers little as a rational process rom deciding touse a orm o inormation technology to generally enhance learning or acilitatecompletion o academic assignments.

Te core o Davis’s (1989) research strategy was to measure what he calledthe perceived usefulness and perceived ease of use o an inormation system in or-der to “explain and predict uture user behavior. . . ater a very brie period o

interaction with the system” (p. 983). Davis argued that these two actors wereinstrumental in determining a user’s attitude toward a system and behavioral in-tention to use it, which ultimately led to actual system use. Davis’s AM modelprovided one more critical component. Te AM model acknowledged thatboth perceived useulness and perceived ease o use—a somewhat perverse anddicult to measure concept when unethical behaviors are contemplated—may be infuenced by a set o antecedent variables (labeled external variables in themodel).

Tese external variables opened the door to study the role o individual di-

erences in technology acceptance (inn & Korukonda, 2004). Tus, the AMmodel was adopted to structure a study o the origins o academic dishonesty in which relationships between individual dierences in ethical principles andpersonality were evaluated as correlates o attitudes about the unethical use o technology in an academic setting. One objective the model made eminently clear was the need or a systematic assessment o student attitudes about usingtechnology unethically. On that topic we ound a huge gap in the research liter-ature (see, or example, Newstead, ranklyn-Stokes, and Armstead’s (1996) listo 21 conventional cheating behaviors), thereby obligating us to plan two stud-

ies—one to develop a list o questionable activities and a second to validate it.In the rst study, we conducted two ocus groups to develop a list o tech-



own ethical principles as measured by the Ethics Position Questionnaire. Tepurpose o the second study was to test the generalizability o Study 1 results by administering the same questionnaire to a dierent sample o college students while adding depth to our understanding o the phenomenon by including apersonality measure conceptually divorced rom the participants’ ethical prin-ciples. Our choice was Zuckerman’s (1971, 1979) sensation-seeking scale. oour knowledge, the Zuckerman scale had not been previously used in ethicsresearch. Yet two components o it, the disinhibition and the thrill and adven-ture seeking scales, had conceptual components that were closely related to ourindividual actors—deviant behavior (Blankenship & Whitley, 2000), socialdominance (Wilson, 2003), lack o empathy (Davis, Andersen, Curtis, 2001),and anomie (Caruana, Rameseshan, & Ewing, 2000)—previously identied ascorrelates o unethical behavior. urthermore, Zuckerman’s sensation-seekingscales provided a conceptual link to ve-actor models o personality (Zucker-man, 2004, 2002; Zuckerman et al., 1993), a widely accepted, comprehensiveramework or personality research (John & Srivastava, 1999).

Accordingly, we jointly designed the two studies to investigate critical externaland attitudinal actors described in the AM model by pursuing the ollowingresearch questions:

1. What are the current methods by which students put inormationtechnology to dishonest academic use?

2. How do students ethically evaluate these academically dishonestbehaviors?

3. Are students’ evaluations infuenced by aspects o their ethical principles?4. Are their ethical principles related to innate personality characteristics?

Questions 1 and 2 ocused on the AM model’s attitude toward using compo-nent while Questions 3 and 4 ocused on individual dierences that would bedened as external actors in the model.

While Davis’s (1989) AM model specied perceived usefulness and perceived ease of use as intervening components in this reasoning process, there were sev-eral reasons to orego measurement o these variables in this exploratory study.irst, Davis’ model assumed these measures would refect perceptions based onbrie interactions with the technology. In a college, as opposed to an organi-zational setting, it could not be guaranteed that all or even most participantshad been personally exposed. Second, those who had successully engaged inunethical activities would be unlikely to share their perceptions truthully. Tirdand most signicant, even i valid measures could be obtained, their infuence would be limited because variance in student perceptions would be highly mod-erated by similar campus situations. Perceived useulness would be constrainedby campus-wide honor codes and disciplinary procedures (McCabe & revino,1993, 1997, 2002), and perceived ease o use would be infuenced by available



and ease o use o technology in a study o academic honesty appeared bettersuited to experimental designs.

METHODOLOGY

SamplingandDataGatheringStudents who participated in these two studies were enrolled at two smallacademic institutions—a church-aliated liberal arts college and a regionaltwo-year campus o a major research university. An institutional review boardapproved the successive studies, and we inormed volunteer students o theirrights to withdraw rom the study at any time as they began their participation.

Study 1. One o the authors at the church-aliated college assembled twoocus groups, comprised o seven and eight volunteers, during the 2003 sum-mer term. Te ocus group script was designed to investigate which types o

inormation technology students were using to assist in their course-related work and which ones the students believed to be academically dishonest. Teocus groups’ questions were sequenced to neutrally probe personal experiencesregarding the uses o inormation technology beore raising the issue o whetherthese technologies might be used in an academically dishonest ashion. Te stu-dent participants actively identied 24 questionable behaviors during the twoocus group sessions.

Ater reviewing the ocus group transcripts or clarication, we developedcomplete verbal descriptions o the 24 behaviors or use as the primary com-

ponent o the Study 1 survey instrument. Te instrument was administered to237 students enrolled in an undergraduate computer applications course at thesame institution. Te students who completed the survey instrument were reg-istered in 16 o 22 dierent sections o the course oered during the all 2003and spring 2004 semesters. Because the course was required or graduation, weexpected the selected sections would approximate a representative sample o all students (about 1,250) currently enrolled. Te demographic data collectedrom questionnaires showed that 70% o the respondents were emale and thatalthough 25 dierent majors were represented, nursing majors represented

29% o the sample. In both cases, these percentages refected the undergraduatepopulation o the college, o whom 74% were emale and 25% were nursingmajors.

Study 2. In the ollow-up, Study 2, which occurred during the spring 2005semester, only survey data were collected. wo-hundred two students out o approximately 500, who were enrolled at the two-year campus o the researchuniversity, participated. Students were recruited by asking their instructors,regardless o course type, to devote class time to administration o the survey during the nal two weeks o the semester. Te students who participated were,

nevertheless, inormed that their participation was voluntary.SurveyInstruments



(orsyth, 1980). In addition, the questionnaire developed or Study 2 includedtwo subscales o the Sensation-Seeking Scale (Zuckerman, 1979).

Background Variables. Students were asked to provide their class standing,their major, degree program, gender, year o birth, and estimates o how re-quently they relied on the use o e-mail, the Internet, and sotware applicationsor coursework. or these three estimates, a 6-point scale was adapted rom theongoing Pew Internet and American Lie surveys. Te choices ranged rom sev-eral times a day (6) to every few weeks (2) ollowed by less often/never (1). Becausea year to 18 months elapsed between the two studies, participants’ ages werecalculated separately by subtracting year o birth rom date o survey adminis-tration.

Ratings of Academically Dishonest Uses of Information Technology. wenty-our items, based on the ocus group descriptions provided by studentsin Study 1, were rated on a 6-point scale that included the ollowing options:(0) Not Dishonest , (1) Not Serious , (2) Somewhat Serious , (3) Moderately Serious (4) Quite Serious , and (5) Very Serious . Previous studies on academic dishonesty (see Aggarwal, Bates, Davies, & Kahn, 2002; urrens, Staik, Gilbert, Small, &Burling, 2001) have used a similar scale. However, based on evidence that stu-dents might be lax in their assessment o unethical acts involving inormationtechnology (Siegried, 2004), we added options to designate the activity as Not Dishonest or respond Don’t Know . Te Don’t Know response was needed as wellbecause some items were technical in nature refecting the specialized knowl-edge o some ocus group participants.

Ethical Position Questionnaire. orsyth’s EPQ was comprised o two 10-item scales, which are evaluated using a 9-point Likert-like response set thatranged rom (1) Completely Disagree to (9) Completely Agree . Te idealism scaleoperationalized the concept that ethical behavior means doing no harm, and, inact, our o the 10 items use the word “harm” and three more reer to anotherindividual’s “welare” (we substituted the term “well-being” ater student com-ments during a pre-test). Te relativism scale operationalized the concept thatthere are no hard and ast ethical rules to be applied in every situation. Tisconcept was explicitly stated in the rst eight items o the scale with the last twoocused on the morality o lying, permitting the construction o an alternateveracity subscale (Davis et al., 2001). Idealism and relativism resulted in scalesranging rom nine to 90. Te two-item veracity scale ranged rom two to 18.

Sensation-Seeking Scale. Te Sensation-Seeking Scale (Zuckerman, 1979) iscomprised o 40 items or which respondents must choose one o two possibleoptions. Each o the our subscales—thrill and adventure seeking, disinhibi-tion, experience-seeking, and boredom susceptibility—is comprised o 10 o these two-option choices. or Study 2, we selected only the 20 items rom thethrill and adventure seeking and the disinhibition scales as being personality dimensions similar to other individual actors associated with academic dishon-esty in previous research.



usually ruins a party because some people get loud and boisterous versus Keeping the drinks full is the key to a good party . Selection o an option characteristic o theconcept being measured (the second item in both these examples) counts onepoint. Tus, a participant’s score on each subscale may range rom 0 to 10. Analysis. Although both samples represented college students at rural institu-

tions in the same region o the country, we chose to analyze them separately because one o the chie purposes o Study 2 was to provide validation or re-sults reported in Study 1. Also, events occurring in the ast-changing world o inormation technology during the 12 to 18 month period that elapsed betweenadministrations o the two surveys could have infuenced student attitudes andpractices. Tus, each data set was analyzed individually with comparisons madein regard to summary statistics. Concerning background variables, summary statistics or gender, birth year, and application sotware use were signicantly dierent between the two samples, but means or age, Internet use, and e-mailuse were similar (see able 1).

Table1:ComparisonofStudy1andStudy2SummaryStatisticsforDemographicVariables

DemographicVariable

Survey 1Sample

Survey 2Sample

Statistical est o Dierence

Gender

Proportion70% emale 49% emale X 2(d=1) = 19.03, p < .001

Mean Birth Year 1977.7 1980.0 (1,434) = 7.84, p = .005

Mean Age 26.3 25.2 (1,434) = 1.63, n.s.

Internet Use 3.16 2.93 (1,435) = 2.59, n.s.

E-mail Use 4.07 4.27 (1,436) = 1.90, n.s.

Sotware Applications Use

3.50 2.80 (1,436) = 28.09, p < .001

RESULTSRatingsofAcademicDishonesty

Te survey or Study 1 was completed by 237 students, who were enrolled ina required inormation technology course during the last weeks o the all 2003semester or rst weeks o the spring 2004 semester. Tey rated the 24 descrip-



activities in rank order by mean score; non-technical items are italicized; andkeywords written in all capital letters designate short titles or unethical behav-iors listed in ables 4 and 6.) Te our highest scoring items were activities thatinvolved submitting as one’s own an assignment completed by someone else.Teir mean ratings placed them in the highest possible range, rom (4) quite serious to (5) very serious , as orms o academic dishonesty. Te next seven items, which ell between (3) moderately serious and (4) quite serious were more dicultto categorize. Tree o the behaviors involved possibly unauthorized assistanceon an exam. wo others involved questionable delay tactics, alsely claiming tohave attached an assignment to an e-mail to gain extra time to complete the work and a non-technical analogue, giving a alse excuse to delay an exam orthe deadline or a paper.

Except or the least serious behavior, reormatting a paper to increase itslength, the means or all 13 remaining items were rated (2) somewhat serious to(3) moderately serious . Tis set o behaviors also matches at the top an inorma-tion technology activity, copying one sentence rom an online source withoutacknowledging it, with its non-technical analogue, copying two lines rom aprinted reerence with no citation. Similarly, near the bottom o this group,reading an online summary or review o a book is paired with reading a con-densed version o a novel.

We gathered survey data or Study 2 approximately 16 months ater Study 1rom 202 students enrolled in a broad sample o courses at a two-year campuso a major research university at the end o the spring 2005 semester. able 2lists their ratings and rankings or the same 24 behaviors and provides a sta-tistical test o dierences between the means generated by the two samples.Dividing the Study 2 ratings into one-point segments, a number o dierencesoccur. irst, only three rather than our items all in the highest range between(4) quite serious to (5) very serious while at the other end o the spectrum, fvebehaviors, instead o just one, all in the range between (1) not serious and (2)somewhat serious . Te our technical items in this latter group are seemingly characterized by their unctional status as sotware applications that automatethe organization o verbal data. Consistent with this overall trend o partici-pants in Study 2 generally rating the seriousness o all 24 behaviors less criti-cally, nine o the 24 paired means in able 2 exhibit statistically signifcant di-erences. In each case, the participants in Study 1 rated the inractions as beingmore serious. Study 1 was conducted 16 months earlier at a church-aliatedcollege.

On second glance, however, there is also a striking similarity between the twosets o scores. I one disregards the absolute values o the ratings and considersthe relative rankings instead, then the two sets o responses are phenomenally similar. Tat is, the Pearson correlation coecient or the two sets o means isa remarkable .96 (df = 22, p < .001). Te more conservative Spearman rho (orordinal level data) is .90. Tus, it may be instructive to consider the two types



riend’s assignment and submitting it as one’s own, receiving and using an e-mailrom a riend about questions on an exam just completed, using Internet chatrooms to ask homework related questions, and having a riend e-mail a copy o acompleted assignment to use as a ramework or one’s own work. In these cases,it would seem that the judgments o the Study 1 sample are more critical, basedon some general group dierence. or the three remaining items, though, thedierences in means are refected in dierences in rankings as well. Tese behav-iors include carrying on an instant message conversation while taking a comput-erized exam, using a Web site or sotware to ormat a bibliography, and usingree Internet sotware programs to complete an assignment. Te dierences inmeans between Study 1 and Study 2 or these three high-tech activities rangedrom .62 to .77 as opposed to .27 to .40 or the other six, thereby eecting a six-place change in each o their rankings as well. While inormation technology ap-pears to have only acilitated an exchange or transer o inormation or the rstsix items, or the remaining three the behaviors would be impossible to perorm without the diusion o sotware innovations. Regardless, the high overall cor-relation between Study 1 and Study 2 ratings suggests that the behaviors listed were reliably evaluated by the two student samples and provides an acceptableramework within which to examine the origins o these ratings.

PREDICTORSOFSTUDY1RATInGS

As a measure o individual dierences, participants in Study 1 were asked to

complete the Ethical Position Questionnaire (orsyth, 1980). On the basis o Davis et al.’s (2001) psychometric analysis, we have generated three scales romthe 20-item instrument, representing idealism, relativism, and veracity (as atwo-item subscale o relativism). As presented in able 3, the means or thesethree scales were 72.7 or idealism, 60.0 or relativism, and 10.72 or veracity, with all three exhibiting acceptable levels o reliability. In addition, they exhibitan interesting set o intercorrelations. Not surprisingly, veracity and relativismare highly correlated (r = .70, df = 208, p < .001) inasmuch as the veracity scaleis derived rom the ninth and tenth items o the relativism scale. However, it

is interesting to note, that while relativism and idealism are also signicantly correlated (r = .30, df =206, p < .001), the correlation between veracity and ide-alism is virtually nil (r = -.02, df =219, n.s.), an indication o its psychometricindependence.

Te ecacy o adding the veracity scale to the analysis is borne out by theset o bivariate correlations (see able 4) between the ratings o the 24 ethically suspect behaviors and the scores generated by the EPQ. While there is no con-testing the primacy o the idealism measure as a signicant actor in determin-ing the orientation o the participating students towards these 24 behaviors,

the two-item veracity scale scores were signicantly correlated with almost hal the behavior ratings while the traditional 10-item relativism scale exhibitedrelationships with only two o the 24 ethically suspect behaviors Even then the



Behavio

rsUs ingInormationTechnolo

gy

Survey1

Survey2

Ratings

Ranks

Ratings

Ranks

- ratio

d.f.

sig.

aPAPERONLINEandsubmittingitasyourown.

4.65

1

4.33

1

12.62

1,

432

.001

GAND

pastinganessayromtheInter

netandSUBMITtingitASyour

4.61

2

4.24

2

15.88

1,

434

.001

ILEo

ndiskcontainingacompleteassignmentROMARIEND

and

thewo

rkwithyourname.

4.33

3

4.10

3

5.38

1.

435

.021

ariend

’sfleromadisk, replacinghis

namewithyoursandsubmitting

nWOR

KitheassignedmaterialissomethingthatYOUalreadyKN

OW

4.06

4

3.84

4

3.56

1,

424

.060

gtohaveattachedanassignmenttoan

e-mailwhenyoudidnotinorderto

RATIM

Etocompletethework.

3.66

5

3.60

5

0.24

1,

427

.624

nanInstantMessageCONVERSATIONwhileTAKINGacompute

rized

3.59

6

2.97

12

16.20

1,

416

.00 1

NGand

usinganE-MAILromarien

dABOUTthequestionsonan

justcom

pleted.

3.55

7

3.28

8

3.89

1,

429

.049

Gtak in

ganexaminationorturningin

apaperUSINGaALSEEX-

3.50

8

3.30

7

2.44

1,

427

.119

GanE-MAILtoariendtoletthemknowABOUTthequestionson

an

ujustco

mpleted.

3.49

9

3.24

9

2.96

1,

430

.086

WEBSIT

ESthatyouDIDNOTUSEtocompleteanassignmentinthe

hyoan

assignment.

3.40

10

3.21

10

1.93

1,

423

.165

wolines

oftextdirectlyFROMaPRINT

EDREFERENCEsourcewithout

ingthesource.

3.32

11

3.49

6

1.50

1,425

.221

andpastingoneSENTENCEroman

onlinesourcewithoutackNOwl-

SOURCE.

2.90

12

2.93

13

0.03

1,

431

.856

T a b l e 2 : C o m p a r i s o n o f M e a n R

a t i n g s i n S u r v e y s 1 a n d 2

o f U n e t h i c a l B e h a v i o r s U

s i n g I n f o r m a t i o n T e c h n o

l o g y



Behavio

rsUsingInormationTechnolo

gy(Cont’d.)

Survey1

Survey2

Ratings

Ranks

Ratings

Ranks

F-ratiod.f.

sig.

NGafEWWORDSoaparagraphcopiedandpastedromtheInternet

materialdoesNOthavetobeCITEd,

2.84

13

3.03

11

1.56

1,424

.213

iesoparagraphsthathavebeencopiedandpastedromaVARIETY

Of

ETSITESt ocreateapaperwithACKN

OWLEDGMENTtothesitesin

graphy.

2.78

14

2.53

14

2.08

1,420

.150

ternetC

HATroomsTOASKHOME

WORKrelatedquestionsonassign-

cribedasindividualassignments.

2.77

15

2.37

16

7.32

1,420

.007

SUMMARYbasedONanONLINEA

BSTRACToajournalarticle

readingthearticleitsel.

2.50

16

2.43

15

0.30

1,424

.586

ebsiteo

rSOfTWAREthatcreatesaB

IBLIOGRAPHYoryoubased

on

nyouenter.

2.50

17

1.73

23

20.40

1,400

.001

sotwarePROGRAMavailablefORr

eeontheInternetTOCOMPLETE

ment(orexample,usingaWebsitetotranslatetextwritteninEnglishto

2.47

18

1.70

24

23.20

1,403

.001

RIEND

E-MAILyouacopyohisco

mpletedassignmentjusttogiveyou

WORK

tobaseyourworkon.

2.46

19

2.06

19

6.60

1,422

.011

anonlineSUMMARYORREVIEWoabookratherthanreadingth

eas-

-length

version.

2.26

20

2.14

17

0.76

1,432

.383

aCOND

ENSEDversionofaNOVELratherthantheassignedfull-lengthver-2.23

21

1.99

20

3.18

1,426

.075

afewlinesofapaperyouhavewrittenforoneclassandSUBMITtin gthisnew

aDIFF

ERENTCLASS.

2.07

22

2.14

18

0.21

1,419

.648

OfTW

AREPROGRAMselecttheim

portantpointsoapaperyouhave

dorganizingthemintoaSUMMARY.

2.04

23

1.80

22

2.56

1,397

.110

actersp

acing,linespacing,andmargin

sizingoptionsinaSOfTWAR

E

oINCR

EASEtheLENGTHoapaper.

1.80

24

1.91

21

0.68

1,426

.408

t a l i z e d

w o r d s a r e u s e d a s a b b r e v i a t e d d e s c r i p t o r s i n s u c c e e d i n g t a b

l e s . I t a l i c s d e n o t e a b s e n c e o f t e c h n o l o g i c a l t e r m i n o l o g y .



Table3:Study1:IntercorrelationsandReliabilitiesofEthicsMeasures

PredictorsUnivariate Statistics Correlation Coefcients

Mean Std. Dev. N Idealism Relativism Veracity

Idealism 72.66 12.55 228 (.86)Relativism 60.00 14.21 210 .30*** (.86)

Veracity 10.72 4.24 225 -.02 .70*** (.78)

Notes: Coefcients o reliability are presented on the diagonal axis.

Table4:Study1:CorrelationsofUnethicalBehaviorsandEthicsMeasures

Unethical Behaviors UsingInormation Technology

Correlation Coefcients

192 ≤ n ≤ 227 Idealism Relativism Veracity

BUYING PAPER ONLINE .20***

COPY AND SUBMIT AS OWN .24***

COPY ILE ROM A RIEND .26*** -.12*

COPY WORK YOU KNOW WELL .33*** -.18**

CLAIM EXTRA TIME .31*** -.13*

IM CONVERSATION TAKING EXAM .20***

RECEIVING E-MAIL ABOUT EXAM .28***

DELAY USING ALSE EXCUSE .34*** -.11*

SENDING E-MAIL ABOUT EXAM .23***

LIST WEB SITES DID NOT USE .37*** -.12*

COPY ROM PRINTED REERENCE .30*** -.14*

COPY SENTENCE NO SOURCE .22***

CHANGING EW WORDS NO CITE .26*** -.13*

VARIETY O INTERNET SITES -.17*

USING CHAT TO ASK HOMEWORK .21***

SUMMARY ON ONLINE ABSTRACT .35*** -.16**

SOTWARE BIBLIOGRAPHY .13*

USING PROGRAM TO COMPLETE .21***

RIEND E-MAIL RAMEWORK .27*** -.13*

READ SUMMARY OR REVIEW .32*** -.12*

READ CONDENSED NOVEL .27*** .13*

SUBMIT OR DIERENT CLASS .17**

SOTWARE PROGRAM SUMMARY 13*



STUDY2RESULTS

Study 2 was completed by 202 students on the two-year remote campus o a major research university during the last two weeks o April, 2005. able 5provides the means, standard deviations, and intercorrelations between the ide-

alism, relativism, veracity, disinhibition, and thrill and adventure seeking scales with their reliability coecients on the diagonal axis. As beore, the correlationbetween relativism and veracity is very strong while the correlation betweenidealism and veracity is near zero. In addition, the correlation between idealismand relativism is signifcant but much weaker or these students. able 5 alsoprovides our frst look at the relationship between idealism and the two sensa-tion-seeking scales. Not surprisingly, they are both negative, but disinhibitionis more strongly correlated with idealism than thrill and adventure seeking is.urther, relativism is correlated only with disinhibition.

Table5:Study2:ItercorrelatiosadReliabilitiesofEthicsadPersoalityMeasures

PredictorsUnivariate Statistics Correlation Coecients

MeanStd.Dev. N Idealism

Relativ-ism Veracity

Dis-inhibition AS

Idealism 67.93 15.01 194 (.88)

Relativism 58.29 15.25 194 .12* (.86)

Veracity 10.86 4.36 201 -.02 .74*** (.68)

Dis-

inhibition4.03 2.53 178 -.35*** .24*** .35*** (.72)

Trill and

Adventure

Seeking

5.93 2.72 196 -.13* .05 .11 .17** (.76)

Note: Coefcients o reliability are presented on diagonal axis.

Te most striking aspect o the Study 2 results, however, is that the disin-hibition scale, used as an independent variable or the frst time in this study,appears to be as eective a predictor o the academically dishonest behavior rat-ings as the EPQ idealism scale (see able 6). Te disinhibition scale correlatessignifcantly with 16 o 24 behaviors. Te idealism scale correlates signifcantly with only one more, 17 o 24 behaviors. Interesting as well are the number o inverse correlations between veracity and the ratings o ethically suspect behav-iors. Tey are signifcant in nine out o 24 cases.

DISCUSSIOnBeore drawing inerences rom the specifc results o these two studies it is



Unethical Behaviors UsingInormation echnology Correlation Coefcients

(177 ≤ n ≤ 194) Idealism Rela-tivism Veracity Disinhi-bition Trill and Adventure

BUYING PAPER ONLINE .18**

COPY AND SUBMI AS OWN .24*** -.13* -.25***

COPY ILE ROM A RIEND .21** -.17** -.30***

COPY WORK YOU KNOW WELL

.21** -.14* -.25***

CLAIM EXRA IME .14* -.17** -.15*

IM CONVERSAION

AKING EXAM.16* -17** - .22*** -.22**

RECEIVING E-MAIL ABOUEXAM

.29*** -.28***

DELAY USING ALSE EXCUSE .19** -.14* -.24***

SENDING E-MAIL ABOUEXAM

.20** -.20**

LIS WEB SIES DID NOUSE

.20** -.20**

COPY ROM PRINED

REERECE .20** -.15*COPY SENENCE NOSOURCE

.20** -.24***

CHANGING EW WORDS NOCIE

.14*

VARIEY O INERNE SIES

USING CHA O ASK HOMEWORK

-.13*

SUMMARY ON ONLINE ABSRAC

.13* -20** -.20** -.25***

SOWARE BIBLIOGRAPHY

USING PROGRAM OCOMPLEE

-.16*

RIEND E-MAILRAMEWORK

-.16* -.14*

READ SUMMARY OR REVIEW .19** -.16* -.21** -.32*** -.14*

READ CONDENSED NOVEL .13* -.20**

SUBMI OR DIEREN

CLASS .17** -.20**SOWARE PROGRAMSUMMARY

Table6:Study2:CorrelationsofUnethicalBehaviorswithEthicsandPersonalityMeasures



Te same norms and individual dierences apply to both. Looking back at theresults rom both surveys, then, there is considerable evidence that both ideal-ism and disinhibition are consistent correlates o attitudes about students usinginormation technology dishonestly in an academic setting. By contrast, thenumber o correlations exhibited by the relativism and the thrill and adventureseeking scales were surprisingly ew. In act, the simple two-item veracity scale,proposed by Davis et al. (2001), proved to be a more requent correlate o rat-ings o unethical uses o inormation technology than its parent relativism scale.

As in every successul exploratory study, we have been able to replicate thendings o previous research, such as conrming in Study 1 the relationshipbetween idealism and descriptions o unethical behavior. We also extended theboundaries o understanding by establishing new correlates in Study 2, such asthe relationship between the personality measure disinhibition and the sameset o unethical behaviors used in Study 1. Nevertheless, the results o the twosurveys have also generated new questions, which we intend to address beoremoving on to limitations o the study and implications or uture research.

Te rst o these questions is how to explain mean dierences in the ratingso unethical uses o inormation technology between the Survey 1 and Survey 2. A number o demographic actors oer possible answers. We know that the in-stitutions where the surveys were administered, though not geographically dis-tant, may draw very dierent student bodies, because one is private and church-aliated while the other is a state-supported, small rural campus o a majorresearch university. Te higher ratings o seriousness occurred at the church-a-liated school. urther, previous research on cheating behaviors (Whitley et al.,1999)has indicated that emale students hold more critical attitudes than malestudents do, and in this case, the church-aliated school sample was heavily weighted with women. inally, there is the tantalizing possibility that the16-month on average period that transpired between administration o the surveysin Study 1 and Study 2 also had a mediating eect on how students viewedsome behaviors. Tis may be especially true as the three exhibiting the greatestchange in rankings were all based on sotware innovations (instant messaging,bibliographic sotware, and ree sotware downloaded rom the Web) which would be dicult to characterize as old orms o questionable behavior usingnew technologies. Demographic data revealed a higher level o sotware use orstudents participating in Study 2 as well.

Te second question that looms large in these results is why the idealism scaleo the EPQ is consistently associated with ratings o academically dishonestbehavior while the relativism scale is not. A careul review o the recent litera-ture, however, suggests that such a result is the norm rather than the exception.Despite orsyth’s theoretical oundations or both measures, the idealism scalealone is the one that dominates results in practice (Davis et al., 1999) whenresearchers ignore orsyth’s our-category typology. In act, the weak results o the relativism scale served as motivation or acting on Davis et al.’s (1999) sug-



tively associated with serious reservations about ethically suspect behaviors threeor more times as requently as its parent relativism scale.

Te results o Study 2 revealed a second interesting dichotomy between twomeasures rom a single psychological instrument, which merits urther discus-sion. When we selected the disinhibition and the thrill and adventure seek-ing subscales o the Zuckerman Sensation-Seeking Scale as possible negativecorrelates o ethical behavior ratings, we chose them because we thought they were the two most promising components o a comprehensive instrument.Conceptually, the disinhibition subscale met the requirements or a measure o deviant behavior (Whitley, 1998), but the thrill and adventure seeking subscaleappeared similarly appropriate because it ocused on a preerence or high risk behaviors. Te rst indication o the disparity between the two subscales was re-vealed by the intercorrelations between the independent variables (see able 5),disinhibition was more strongly correlated with idealism (r = -.35) and veracity (r = .35) than thrill and adventure seeking was with either (-.13 < r < .11).

A review o the items that comprise these our scales, however, suggested thatdierences that exist at the operational level are actually more compelling thanthe conceptual rationale or adopting both the thrill and adventure seekingand the disinhibition scales as possible correlates. Because the items compris-ing the idealism scale were weighted with statements about doing no harm toothers, it is not surprising that a person scoring high on idealism would nd itincompatible to “seek pleasure around the world with the ‘jet set,’” “like to gethigh (drinking liquor or smoking marijuana),” “like to date members o theopposite sex who are physically exciting” or be in the company o “swingers.”Te implied risks in these behaviors are not only to sel, but also to others. By contrast, this conscious disregard or others is almost totally absent rom the 10items that comprise the thrill and adventure seeking scale. Nine o them deal with individualized physically demanding sports—mountain climbing, waterskiing, sur boarding, fying, scuba diving, parachute jumping, high diving, astskiing, and long-distance sailing—“things that are a little rightening” as thetenth item indicates, but do not necessarily require that others take the samerisks. In this regard, the thrill and adventure seeking scale could be as mucha measure o athletic sel-condence as sensation-seeking. A major dierencebetween the thrill and adventure seeking and the disinhibition scales is how narrowly the risk actor is operationalized in the ormer. Given this distinction,it is not surprising that the thrill and adventure seeking scale bore scant relationto ethical evaluations. Zuckerman (2004) has acknowledged in a discussion o the development o his ve-actor model o personality that sensation-seeking asa characteristic has both a bright and a dark side. Some components align withextraversion, others with neuroticism. Bringing sensation-seeking to bear onethical behavior may have inadvertently exposed that division.Limitations. A discussion o the meaning o results makes one well aware

o the limitations o this study. oremost is the diculty o generalizing rom



AM model species that behavior intentions and actual use would ollow. Teimportance o the relationships the data analysis revealed or devising ways tothwart the threat to academic integrity occasioned by technological innovationsis at best hypothetical. inally, it may be that the briefy described unethicalbehaviors devised or this study were devoid o sucient complexity to create aneed to weigh situational actors beore evaluating a behavior. Accordingly, theshort descriptive items used may not have provided a valid test o whether rela-tivism is a salient actor in students’ ethical responses.

Future Research. Te implications or uture research are many and diverse. While we were struck during our review o the literature that little eort hadbeen expended to develop an inventory o academically dishonest behaviorsaorded by access to inormation technology, we are now struck by the phe-nomenological nature o the list which we compiled. Using ocus groups to de-velop such a list was an appropriate methodology (Morgan, 1997), but or thepurposes o analysis, a more structured, however limited, list might have beenmeaningully submitted to actor analysis to avoid separate multivariate compu-tations or each item on the list. Some o these behaviors may be old orms o academic dishonesty that have been updated with inormation technology, new orms o academic dishonesty aorded only because o technological innova-tions, and high prole inractions in which the perpetrator claims authorshipor a lengthy nished work completed by someone else. In reviewing oensesthat students rated as highly serious, it struck us that the amount o eort ex-pended—in these academically dishonest behaviors, the eort expended seemedlittle more than inserting one’s name as an author—was a critical actor inorming ethical judgments, and that should be the object o urther study. Anethical imperative emerged, because o a perception o unair treatment o oth-ers who truly worked hard to complete an academic assignment. Tis mode o thinking may be exploited to encourage compliance with honor code systems,including the reporting o inractions.

Another major objective o our research was to make a case or using well-known and well-investigated psychological instruments to make aster headway in understanding the origins o academic dishonesty. In reading the extensivereviews o research on student cheating (Crown & Spiller, 1998; Whitley,1998), one sees that many research results come to a dead end because the in-dividual actors selected or study are incommensurate with those used in otherstudies. Use o the EPQ was an exception in the ethics literature starting in the1990s, and since we wanted to investigate even more undamental antecedentsto ethical attitudes, we used Zuckerman’s Sensation-Seeking Scale as a surrogateor longer instruments developed to document the ve-actor model o per-sonality (John & Srivastava, 1999). As testimony to the ecacy o our strategy, we ound Davis et al.’s (1999) psychometric analysis o the EPQ invaluable inpursuing our own analysis. One implication, we think, o this study is that theEPQ, now more than 25 years old, needs a serious theoretical and operational



ACKnOWLEDGMEnT

We wish to acknowledge Dr. A. J. Grant or his assistance in the developmento the ocus group script or the rst part o this study and or his guidance inraming descriptions o ethically questionable behaviors used in the survey instrument.

Cotributors

Stephanie Etter is an assistant proessor o inormation technology and direc-tor o the itle III Project at Mount Aloysius College in Cresson, Pennsylvania.Her current research interests include computer and inormation security andpedagogical issues, such as distance education, incorporating technology acrossthe curriculum, and academic dishonesty. (Address: Stephanie Etter, DSc, De-partment o Inormation echnology, Main 230, Mt. Aloysius College, Cres-son, PA 16630; [email protected].)

Jackie Cramer is an instructor in accounting at the University o Pittsburgh atitusville. Her doctoral research ocused on how ethical styles and risk-takingbehaviors infuence student perceptions o academically dishonest uses o inor-mation technology. (Address: Jackie J. Cramer, DSc, Accounting and BusinessInormation Systems, Broadhurst Science Center 108, University o Pittsburghat itusville, itusville, PA 16354; [email protected].)

Seth inn is a proessor o communication at Robert Morris University. Hisresearch interests include evaluating alumni perceptions o their participation incollege-wide laptop programs and the relationship between unpredictable verbal

inormation and physiological indicators o arousal. (Address: Seth inn, PhD,School o Communications and Inormation Systems, 6001 University Boule-vard, Robert Morris University, Moon ownship, PA 15108; [email protected].)

Refereces

Aggarwal, R., Bates, I., Davies, J. G., & Khan, I. (2002). A study o academicdishonesty among students at two pharmacy schools. Te Pharmaceutical Jour-nal , 269 , 529–533.

Allmon, D. E., Page, D., & Roberts, R. (2000). Determinants o perception

o cheating: Ethical orientation, personality, and demographics. Journal of Busi-ness Ethics , 23, 411–422. Auer, N., & Krupar, E. (2001). Mouse click plagiarism: Te role o technol-

ogy in plagiarism and the librarian’s role in combating it. Library rends , 49 ,415–435.

Barnett, ., Bass, K., & Brown, G. (1996). Religiosity, ethical ideology, andintentions to report a peer’s wrongdoing. Journal of Business Ethics , 15 , 1161–1174.

Blankenship, K. L., & Whitley, B. E. (2000). Relation o general deviance to

academic dishonesty. Ethics & Behavior, 10, 1–12.Bowers, W. J. (1964). Student dishonesty and its control in college . New York:Columbia University



Buckley, M., Wiese, D., & Harvey, M. (1998). An investigation into the di-mensions o unethical behavior. Journal of Education for Business , 73, 284–290.

Bull, G., Knezek, G., Roblyer, M. D., Schrum, L., & Tompson, A. (2005). A proactive approach to a research agenda or educational technology. Journal of Research on echnology in Education, 37 , 217–220.

Campbell, C. R., Swit, C. O., & Denton, L. (2000). Cheating goes hi-tech:Online term paper mills. Journal of Management Education, 24 (6), 726–739.

Caruana, A., Ramaseshan, R., & Ewing, M. . (2000). Te efect o anomieon academic dishonesty among university students. Te International Journal of Educational Management , 14 (1), 23–30.

Concoran, K. J., & Rotter, J. B. (1987). Morality-Conscience Guilt Scale asa predictor o ethical behavior in a cheating situation among college emales, Journal of General Psychology , 116 , 311–316.

Covey, M. K., Saladin, S., & Killen, P. J. (1989). Sel-monitoring, surveil-lance, and incentive efects on cheating. Journal of Social Psychology , 12 9,673–679.

Crown, D. E., & Spiller, M. S. (1998). Learning rom the literature on col-legiate cheating: A review o empirical research. Journal of Business Ethics , 17 ,683–700.

Davis, . D. (1989). Perceived useulness, perceived ease o use, and user ac-ceptance o inormation technology. MIS Quarterly , 13(3), 319–340.

Davis, M. A., Andersen, M. G., & Curtis, M. B. (2001). Measuring ethicalideology in business ethics: A critical analysis o the Ethics Position Question-naire. Journal of Business Ethics , 32 , 35–53.

Dexter, S. L., Anderson, R. E., & Becker, H. J. (1999). eachers’ views o computers as catalysts or changes in their teaching practice. Journal of Researchon Computing in Education, 31(3) 221–239.

inn, S., & Korukonda, A. R. (2004). Avoiding computers: Does personality play a role? In E. P. Bucy & J. E. Newhagen (Eds.), Media access: Social and psy-chological dimensions of new technology use (pp. 73-90). Mahwah, NJ: Erlbaum.

orsyth, D. (1980). A taxonomy o ethical ideologies. Journal of Personality and Social Psychology , 39 , 175–184.

John, O. P., & Srivastava, S. (1999). Te Big ive trait taxonomy: History,measurement, and theoretical perspectives. In L. A. Pervin & O. P. John (Eds.),Handbook of personality: Teory and research (pp. 102–138). New York: GuilordPress.

Leming, J. S. (1980). Cheating behavior, subject variables, and componentso the internal-external scale under high and low risk conditions. Journal of Educational Research, 74 , 83–87.

Levin, D., & Araeh, S. (2002). Te digital disconnect: Te widening gap be-

tween Internet-savvy students and their schools. Washington, DC: Pew Internet& American Lie Project. Retrieved April, 28, 2006, rom http://www.pewin-ternet org/pds/ PIP Schools Internet Report pd



McCabe, D. L., & revino, L. K. (1993). Honor codes and other contextualinfuences, Journal of Higher Education, 64 , 522–538.

McCabe, D. L., & revino, L. K. (1995). Cheating among business students: A challenge or business leaders and educators. Journal of Management Educa-tion, 19 , 205–218.

McCabe, D. L., & revino, L. K. (1997). Individual and contextual infu-ences on academic dishonesty: A multicampus investigation. Research in Higher Education, 38, 379–396.

McCabe, D. L., & revino, L. K. (2002). Honesty and honor codes. Aca-deme , 88 , 1, 37–41.

McCabe, D. L., revino, L. K., & Buttereld, K. D. (2001). Dishonesty inacademic environments. Te Journal of Higher Education, 72, 29–45.

McIntyre, R. P., Capen, M. M., & Minton, A. P. (1995). Exploring the psy-chological oundations o ethical positions in marketing. Psychology & Market-ing , 12 (6), 569–582.

McLaerty, C. L., & oust, K. M. (2004). Electronic plagiarism as a collegeinstructor’s nightmare—Prevention and detection. Journal of Education for Busi-ness , 79 (3), 186–190.

Michaels, J. W., & Miethe, . D. (1989). Applying theories o deviance toacademic cheating. Social Science Quarterly , 70, 870–885.

Morgan, D. (1997). Focus groups as qualitative research (2nd ed.). TousandOaks, CA: Sage.

Mullin-Marta, J. K., Attia, A., Singhapakdi, A., & Atteya, N. (2003). A com-parison o ethical perceptions and moral philosophies o American and Egyp-tian business students. eaching Business Ethics , 7 , 1–20.

Myers, I. B., & McCaulley, M. H. (1985). Manual: A guide to the development and use of the Myers-Briggs ype Indicator . Palo Alto, CA: Consulting Psycholo-gists Press.

Newstead, S. E., ranklyn-Stokes, A., & Armstead, P. (1996). Individual di-erences in student cheating. Journal of Educational Psychology, 88, 229–241.

Perry, A. R., Kane, K. M., Bernesser, K. J., & Spicker, P. . (1990). ype-A behavior, competitive achievement-striving, and cheating among college stu-dents. Psychological Reports , 66 , 459–465.

Roig, M. (1997). Can undergraduate students determine whether text hasbeen plagiarized? Te Psychological Record , 47, 113–122.

Roig, M., & Deommaso, L. (1995). Are college cheating and plagiarism re-lated to academic procrastination? Psychological Reports, 77, 691-698.

Ross, K. A. (2005). Academic dishonesty and the Internet. Communications of the ACM , 48 (10), 29–31.

Salter, S. B., Guey, D. M., & McMillan, J. J. (2001). ruth, consequences,and culture: A comparative examination o cheating and attitudes about cheat-ing among U.S. and U.K. students. Journal of Business Ethics, 31, 37–50

Siegried, R. M. (2004). Student attitudes on sotware piracy and related is-



Strike, K., & Moss, P. (1997). Ethics and college student lie . Englewood Clis,NJ: Prentice Hall, Inc.

Szabo, A. (2004) Cybercheats: Is Inormation and Communication echnol-ogy ueling academic dishonesty? Active Learning in Higher Education, 5 (2),180–199.

ownley, C., & Parsell, M. (2004). echnology and academic virtues: Studentplagiarism through the looking glass. Ethics and Inormation echnology , 6 (4)271–278.

urrens, J. ., Staik, I. M., Gilbert, D. K., Small, W. C., & Burling, J. W.(2001). Undergraduate academic cheating as a risk actor or uture proessionalmisconduct. In N. H. Steneck & M. D. Scheetz (Eds.), Investigating researchintegrity: First ORI research conerence on research integrity (pp. 49-53). Retrievedebruary 14, 2006 rom http://www.ori.dhhs.gov/documents/proceedings_rri.pd.

Ward, D. A., & Beck, W. L. (1990). Gender and dishonesty. Journal o Social Psychology , 130 , 333–339.

Whitley, B. E. (1998). actors associated with cheating among college stu-dents: A review. Research in Higher Education, 39, 235–274.

Whitley, B. E. (2001). Gender dierences in aective responses to havingcheated: Te mediating role o attitudes. Ethics & Behavior , 11, 249–259.

Whitley, B. E., Bichlmeier Nelson, A., & Jones, C. J. (1999). Gender dier-ences in cheating attitudes and classroom cheating behavior: A meta-analysis.Sex Roles , 41, 657–680.

Wilson, M. S. (2003). Social dominance and ethical ideology: Te end justi-fes the means? Te Journal o Social Psychology , 143, 549–558.

Wryobeck, J. M., & Whitley, B. E. (1999). Educational value orientation andpeer perceptions o cheaters. Ethics & Behavior , 9 (3), 231–242.

Zabihollah, R., Elmore, R. C., & Szendi, J. Z. (2001). Ethical behavior inhigher educational institutions: Te role o the code o conduct. Journal o Busi-ness Ethics , 30 , 171–183.

Zuckerman, M. (1971). Dimensions o sensation seeking. Journal o Consult-ing and Clinical Psychology , 36 , 45–52.

Zuckerman, M. (1979). Sensation seeking: Beyond the optimal level o arousal .Hillsdale, NJ: Lawrence Erlbaum Associates.

Zuckerman, M. (2002). Zuckerman-Kuhlman Personality Questionnaire(ZKPQ): An alternative fve-actorial model. In B. DeRaad & M. Peruginini(Eds.), Big fve assessment (pp. 377-396). Seattle, WA: Hogree & Huber.

Zuckerman, M. (2004). Te shaping o personality: Genes, environments,and chance encounters. Journal o Personality Assessment , 82 , 11–22.

Zuckerman, M., Kuhlman, D. M., Joireman, J., eta, P., & Krat, M. (1993). A comparison o three structural models or personality: Te big three, the bigfve, and the alternative fve. Journal o Personality and Social Psychology , 65 ,757–768.



Documents

Origins of Academic Dishonesty