Can Demography Predict Academic Dishonest Behaviors of Students ? A Case of Pakistan

Academic dishonesty has been a matter of great concern in higher education for last few decades. The dishonest behavior of students at graduate and undergraduate level has become a severe issue for education and business sector, especially when the students exercise same dishonest practices at their jobs. The number of private and public sector universities is increasing; therefore, the effects of academic dishonest behavior on potential professionals need to be carefully investigated and appropriate policies must be formulated by academicians in order to resolve this issue. The present research addresses this matter by investigating into the relationship of student’s demographics such as age, gender, academic program, business/non-business major and CGPA with academic dishonesty. The results have reported that male students of less age studying at the undergraduate level have their own code of conduct and they are less concerned about the academic ethics. Moreover, business students showed a serious behavior regarding academic integrity, whereas, more intellectual students are also very much conscious towards academic integrity. The study found students’ demographics to have a significant impact on their attitudes towards academic dishonesty. The results provide a strong implication for academicians to develop the moralities and ethics in students so that institutions may provide ethically cultivated professionals to the business community.


Introduction
Academic Dishonesty has been a matter of great concern in higher education during the last few decades.The issue of dishonest behavior of students at graduate and undergraduate level has become very severe, particularly when students continue to exercise the same practices at the workplace.The worst scandals of world top companies of World Com and E-toyes, Enron & Adelphia have forced the researchers to focus their attention on the role of college and universities in ethical training of tomorrow's business leaders.The cheating students have strong tendency to practice same unethical and dishonest behaviors at the workplace which they had exhibited during their education (Grimes, 2004;Rakovski and Levy, 2007;Hardling et al;2004;Lawson, 2004).The number of private and public sector educational institutions is increasing day by day; therefore, the impact of academic dishonest behavior on the life of potential professionals needs to be carefully analyzed and appropriate policies must be formulated in order to minimize these unethical practices in the business and education sectors.
The present study sheds some light on this issue by investigating the relationship of demographics with the dishonest behavior of students at university and college levels in Pakistan.Much of the earlier work on the topic has been carried out in the developed economies, mostly in USA.However, the preset study is focusing on the developing economy of Pakistan where entire culture of education is different from the western world.In Pakistan, education system is divided into secondary, college (higher secondary) and university (bachelor and masters) level.Most of the modern and applied programs are offered in the universities at undergraduate and graduate level which is usually attended by the students at the age of above 18 years.Furthermore, the awareness of female education in Pakistan at higher level of studies is also increasing which is also apparent from the proportion of female respondents of the current study (i.e.approx.30% of the total sample).
Thus, in the different education culture of Pakistan, this research is expected to contribute a better understanding of the ethical decisions of students helping the academicians and business professionals to look into and formulate some policies to refrain from this behavior.The remainder of the paper is organized as follows; the next section reviews some significant studies, third part develops the methodology followed by results and discussion in the next.Final section concludes the study by suggesting some implications for educators and future avenues for researchers.

Literature Review
Much has been written and researched about the students' ethics in higher education (Rakovski & Levy, 2007).Most researchers have focused on assessing the different types of academic dishonest acts and their relation with the demographic factors of students such as; gender, age, program level, subject major and academic performance.In this regard, some earlier work of William and Bowers in 1964 has provided a strong base, which is further explored by Donald McCabe in 1990s.However, these concepts came from the developed world and researches were also carried out in these countries.In general, research concluded that dishonesty in education is rampant which needs to be carefully analyzed in other countries as well along with its relation with the demographic factors of the students.

Gender & Age Effect
Research has found mixed evidence on the gender effect on moral values of students.Although, some earlier studies reported inconclusive findings on gender differences and academic dishonesty (Thoma, 1986); however, recent studies noted a link is prevailing (Shaub, 1989;Sweeney, 1995, Cohen et al, 1998).As per Malone (2006), attitude of male and female students differs on some dishonest acts but for most of the issues of dishonesty, they behave in same way.Cohen et al. (1998) developed a Multidimensional ethics Scores (MES) to evaluate the ethical evaluation and intention aspects of honest behaviors, and found that males and females had significantly different set of judgments on their perception of ethical behavior.Some other studies reported that male students are more frequently engaged in dishonest acts than females (Bower, 1964;McCabe et al, 1997;Whitley et al, 1999).Moreover, this is also confined by a literature review paper of Crown and Spiller (1998) who reported more involvement of male students in cheating than females.So, we can also expect a significant relation between the gender difference of students and their involvement into academic dishonest acts.
Different studies have addressed the students' dishonest behaviors on the basis of age as well.It is reported that younger students engaged more oftenly in cheating than their older counterparts (Haines et al., 1986;Graham et al., 1994;Diekhoff et al., 1996).Another point of view came into consideration i.e. in younger age, they have their own code of ethics to behave in society but as they grow up, they show moralities in their behaviors and become more philosophical (Auerbach and Welsh, 1994;Barger et al., 1998).Younger and unmarried students are more tolerant to cheating behavior than older and married students (Whitley et al., 1998).This notion is also supported by Coombe and and Newman (1997) that the individuals at younger age, are found to be less ethical than the older ones.

Subject Majors and Level
Regarding the subject majors and program levels of students, researchers are confused.Many studies provided evidence that, at the college and university level, the business students are among the most dishonest ones (Caruana et al., 2000;Clement, 2001;Smyth and Davis, 2004).Business students provided the highest cheating rate 87% while comparing it to the other non-business majors (Caruana et al., 2000).Harris (1989) reported that, most business students have low ethical values than their peer students in other majors.Recently, Christine and James (2008) analyzed the academic behaviors of students and showed that subject major significantly influences the students' choice for academic dishonesty.Contrary to these studies, Beltramini et al., (1984) provided a very weak precedent that despite the gender effect business students are ethically sounds that the students opting for non-business subjects.
Prevalence of academic integrity and dishonesty has also been studied across the different levels of the students.Zastrow (1970) has concluded that the frequency of cheating in students at the graduate level was at least as extensive as for the undergraduate students.Rakovski and Levy (2007) noted that undergraduate students are involved more oftenly and extensively in dishonest acts than the graduate students; however, Christine & James (2008) provided that there is no significant difference between the attitudes of students towards academic dishonesty at the graduate and undergraduate level.Finally, academic performance of students has also been an important predictor which reflects the negative relationship.Smith et al., (2002) summarized the results of various studies and concluded that students with greater academic performance are engaged in cheating less often than the students with lower performance.
From the above discussion, we can infer that the studies to analyze the perceptions of students towards academic dishonesty on the basis of students' demographics are conducted mostly in developed countries.This issue, yet, has not been explored in the context of higher education in developing countries like Pakistan.To fill this gap a structured questionnaire has been administered to assess the determinants of academic dishonesty among the students of professional education in public and private sector universities of Pakistan.

Methodology
The study gathered the data from the respondents on a well-structured and self-administered questionnaire.The questionnaire was divided into two parts.First part was focusing on the demographic factors of the students responding.There demographic factors were gender, age group of respondents, program level in which the student is enrolled, subject majors taken, academic performance measured in terms of CGPA of the student earned.In the education system of Pakistan, the students complete their undergraduate level maximum at the age of 21 years and then they entered into the masters program which is expected to be completed at the age of 23 or 24 years.Thus, the age classifications have been grouped on the basis of their graduate and undergraduate level of education that at the undergraduate level, they will fall in the 1 st category of age group and graduate level students will be in the 2 nd category, hence, students of different ages studying at different education levels show different behaviors towards academic ethics.Some executive and senior students may fall in the third category of age i.e. 26-30 years, however, this ratio is very less (approx.1.3% of the total sample).
Second part consists of most commonly researched thirteen unethical academic acts the students may involve in.The responses were arranged on a five point Likert Scale which receive responses for every dishonest act of the students in always or never (i.e. 1 stands for never and 5 for always involved).The frequency of academic dishonest behaviors has been measured by second part of the questionnaire taken from the literature (Cohen et al., 1998;Davis and Welton, 1991;Rakovski and Levy, 2007).Before filling out the questionnaire, the respondents were briefed about different terminologies used in the questionnaire (rarely, occasionally, never, mostly etc.) as well as about the confidentiality of the data to be gathered.The data was collected by a questionnaire discussed above from the graduate and undergraduate students studying at the various universities of Pakistan.The questionnaires were distributed in the classes and students took approximately 15 minutes to complete each questionnaire.
For the sake of generalization and fruitfulness of the study, students were selected from the senior most classes of the professional fields only like the business, engineering, public administration and commerce.There were 1000 questionnaire distributed among the respondents out of which 958 were found complete and useful questionnaire returned having 95.8% effective response rate.The internal consistency of the Scale and data collected was tested using Cronbach's alpha which produced a co-efficient of 0.85.The research has proved that the value of this alpha is greater than 0.5 and is acceptable in social sciences (Nunally, 1978).Descriptive statistics and different measurements of association have been used to investigate the relationship of students' demographics and their academic dishonest behaviors using SPSS 16.

Results and Discussion
The cross tabulation were shown in Table 1 in CGPA, Academic Program, Age Group and Major were compared with Gender of the respondents.The sample was collected from different universities which are representative of population.Out of 958 respondents 638 (67%) are male and 320 (33%) are female.In terms of academic performance, a vast majority of male students secured a CGPA ranging from 2.50 to 3.50 while out of 320 female respondents 247 (77%) were found to have secured CGPA of 3.00 to 4.00 which reflects that female students are more hard working and earn good grade than their male counterparts.Number of respondents studying in undergraduate classes is more than that of graduate classes.A heavy majority of respondents were found in age group of 21 to 25 while no student was found above thirty years of age.There is slight variation in sampling across business and non-business majors with former having small majority both in male and female respondents.
Table 2 reports means and percentage of 13 academic dishonest acts the students usually involve in during their higher education.Most of the students perceive that out of thirteen dishonest acts, copying exam from cheating material, submit another's assignment with their own name and very importantly, stealing the exam material are the most serious dishonest acts in which they have never or rarely involved very often i.e. 90%.However, there are some acts which are less severe and students don't feel any hesitation if they are occasionally or mostly do those actions.These are helping others to copy exam and assignments, copy from internet with mentioning the source of information, receiving and helping others on graded assignment or project report and allow others to use their report for new report preparation.More than 60% of the students usually involved in these dishonest academic acts and they don't feel it against academic integrity.The remaining issues of copying other's exam sheet, copying assignment or project report and copying from internet without mentioning the source are the academic dishonest acts in which the students involved rarely or occasionally.
In order to investigate the impact of the demographic variables on the perception of the students regarding academic dishonesty, mean scores have calculated of each dishonest act and arranged on the basis of demographic factors of the respondents in Table 3. Moreover a composite measure of academic ethics has also been calculated as the means of all thirteen dishonest acts and names as the Ethical Score (ES) of each respondent.Lower this ES, better the ethical values of the students in the sample.The average ES of the respondents is 2.39 which may said to be a good score as it is lying in the lower half of the sample.In accordance with our earlier discussion, the average scores of item 4, 5, 8, 9 & 11 are higher than the average value of ES and these are the acts about which students don't feel seriousness regarding academic dishonesty and the involvement in these acts is somehow greater.It is also evident from Table 3, that ethical values of male students are lower than their female counterparts.Although male are more dishonest than females in terms of academic dishonesty, however; both are behaving in the same way towards individual dishonest acts.These findings are similar with the earlier ones of Malone (2006), Whitley et al., (1999) and McCabe et al., (1997) that males are most frequently engaged in dishonest acts than females.
Mixed evidence is found regarding the age and academic dishonesty of the students.The sample involves only 1% of the students greater than the age of 25 years.The two remaining categories of age are unable to produce any prominent difference between the Ethical Scores of respondents as well as on the individual dishonest acts.However, this phenomenon is very clear when we came to the program level of students.As literature proved that students at undergraduate level tend to be more tolerant towards academic dishonesty by providing higher value of ES as well as greater mean values of approximately all the academic dishonest acts (Haines et al., 1986;Graham et al., 1994;Diekhoff et al., 1996;Coombe and Newman, 1992;Whitley et al., 1998;and Rakovski and Levy, 2007).It is evident that as the students grow up and progress to higher classes, their moral values are developed and tendency of cheating is reduced.The subject major significantly influences the students' choices of academic dishonesty.Contrary to the earlier researches, the present study is finding the non-business students more dishonest academically than the business students.The value of ES is greater for non-business students as well as the mean scores of academic dishonest acts are greater for non-business students.The only exception is copying from internet with mentioning the sources of information where business students scored higher.So it is concluded that business students of Pakistan are less concerned with academic dishonesty as compared to the non-business students.Finally, it is clear from the last panel of Table 3 that students with lower academic performance in their course are most likely to be involved in academic dishonesty.In accordance with Smith et al., (2002), the present study found that students with better academic performance in terms of their accumulated results are engaged less frequently in academic cheating than the students with poor performance.
In order to check the statistical significance of the results found, various tests like one-sample t-test, independent sample t-test and one way ANOVA been conducted and results are reported in Table 4, 5, and 6.For one sample t-test, the mean of the means (i.e.2.54) has been used as test value.The results found the negative mean differences for more than half of the academic dishonest acts and these are statistically significant at 1% level.The items with positive mean difference are not much serious acts as perceived by the students in the earlier discussion of Table 3.Moreover, Independent Sample t-test has been used to investigate the impact of gender, program level and subject major on the academic dishonesty of students.Table 5 reports positive and statistically significant result (at 1% level) regarding the grade for all dishonest acts including the ethical score of respondents.It is now confirmed statistically, that male students are more frequently involved in academic cheating than the females.The results of Table 5 further confirm our earlier findings regarding program level and subject major and report that students at undergraduate level and from non-business field of study are more frequently involved in academic dishonesty as compared to graduate level and business students in Pakistani universities.
One-Way ANOVA has been used to validate the relationship of students' dishonesty and their age & academic performance and results are presented in Table 6.The first column of results represented the F-value of age variable along with the thirteen variables of academic dishonesty and ethical score of respondents.Among all, age is only found significantly related to copying exam from cheating material for which we further run the post hoc test of Lease Significant Difference (LSD) Test.LSD reported that only four pairs are significant for age variable with copying exam from cheating material and these are found for the first category of age i.e. 16-20 years.Finally, approximately, all the academic dishonest acts including ES are found statistically significant either at 1% or 5% level by One-Way ANOVA.The post hoc results of LSD reported that 126 out 280 pairs are found statistically significant at certain level of significance (i.e.1%, 5% or 10%).It has now statistically confirmed the earlier discussions and findings that demography do predict the ethical values of students studying in the universities of Pakistan.

Conclusion
The issue of academic dishonesty has remained a matter of great concern during past few decades.The situation and scandals of popular companies make it more sensitive due to the students' practices of dishonesty at the workplace.The present research has tried to investigate some of the aspects of this issue and explored the impact of students' demographic factors on their inventions and perception of academic dishonesty.A self-administered questionnaire was distributed to the senior students of different business and non-business programs at higher level of study.The students were supposed to respond whether they engaged in any dishonest act always or never.The results of different statistical tests have concluded that the male students are more frequently involved in academic dishonesty than their female counterpart.Moreover, students of less age and studying at undergraduate level are more concerned with the academic dishonest acts.This is the confirmation of the notion that the younger students have their own code of ethics to behave in the society; however, as they grow up and progress to senior classes at graduate level, they show moralities in their behaviors and become more philosophical in attitudes.Contrary to most of the earlier literature, the business students of Pakistan showed higher ethical values towards their academics, and perhaps; this is the reason that Pakistani business community has not faced any popular and major collapse like ENRON, and some other international firms.Furthermore, the brilliant and academically strong students involve less in academic dishonesty as they are competent enough and do not need to be a part of such immoral activities.
The results also put emphasis on the need to have a careful insight by the academicians and policy makers on the ethical and moral values of students at the undergraduate level at a university.This also puts stress on the requirement to impact the course of Ethics in the undergraduate curriculum, especially for non-business students.Finally, the students also left some un-attended areas of this field to be addressed in future.These may include looking this issue in the other regions of Pakistan by increasing the sample and taking into consideration more universities.Different programs, subject major and other academic characteristics can be helpful to further explore the demographical impact on the students' attitude towards academic dishonesty.Different personality traits of students and personality types (A/B) along with locus of control are also some issues of academic dishonesty to be discussed in future researches.Zastrow, C. H. (1970)

Table 3 .
. Cheating among college graduate students.The Journal of Educational Research, 64 (4), Mean Scores bases on Demographic Differences * Significant at 1% Level

Table 5 .
Independent Sample T-test for Gender, Program Level and Subject Major T-statistic is in the parenthesis.*, **, and *** are Significant at 1%, 5%, and 10% Levels, respectively

Table 6 .
One-way ANOVA for Age and Academic Performance *, and ** are Significant at 1% and 5%, Levels, respectively