Bilig 49. sayı

bilig, Spring / 2009,  number  49 
 
192 
reported that cheating rates among college  students  were  as  high  as  90% 
(Graham et al. 1994). 
Academic cheating behavior can be seen at the every level of academic life. 
It has been seen as one of the most serious academic problems, therefore 
many studies were carried out to understand the reasons of student cheating 
behavior in exams. It is possible to classify these studies under three groups. 
One of these studies examined the effects of situational variables such as 
intelligence and sex on cheating. Research findings indicated that students of 
lower intelligence, having more to gain with regard to grades, would cheat 
more compared to more intelligent students (Johnson et al. 1972, Vitro 
1971, Kelly et al. 1978). The results of the research that examined the 
relationship between sex and cheating behavior indicated inconsistent 
results. At one hand, some studies found that female students were involved 
in more academic dishonesty than male students (Graham et al. 1994, 
Jacobson et al. 1970). On the other hand, some studies found that male 
students cheated more than female students (Baird 1980, Cochran et al. 
1998, Davis et al. 1992, Kelly et al. 1978, Roth et al. 1995, Newstead et al. 
1996). Besides, some studies indicated that there was no sex related 
difference (Vitro et al. 1972, Houston 1977, Karabenick et al. 1978, Tibbetts 
et al. 1999, Ward et al. 1990). Further, studies that examined cheating rates 
in relation to age are similarly inconsistent. Some researchers reported that 
younger students cheated more than did older students (Baird 1980, 
Cochran et al. 1998, Haines et al. 1986, Newstead et al. 1996), but at least 
one study reported higher rates of cheating for older students (Tang et al. 
1997). 
The second group of studies examined the effects of the performance goal 
on cheating behaviours. They suggested that there was a relationship 
between the pressures of performance, fear of failure, the goal of getting 
better grates and cheating behaviour (Calabrese et al. 1990, Michales et al. 
1989, Ames et al. 1988, Newstead et al. 1996). The third group of studies 
focused on the relationship between social factors and cheating behaviour. 
They found positive relationship between students self reported cheating, 
dislike of school and views of teachers and schools as unfair in samples 
student. It means that social factors had relationship with cheating behavior 
(Juvonen et al. 1996, Goodenow 1993, Midgley et al. 1996). Finally, a few 
studies were conducted in literature to understand the relationship between 
cheating behaviour and reported high grade point average. Findings have 
shown that students who report comparatively high levels of cheating have 
lower grade point averages (GPAs) (Baird 1980, Graham et al. 1994). There 
are no sufficient studies over the relationship between cheating behaviour 

---

Tayfun, Is There a Relationship Between Grade Average Point and Students 
 
193 
and reported grade point in literature. The reason for this may be due to the 
perception of being a successful student, because successful students usually 
are thought that they have high grade point average and they don’t need 
cheating. At this stage, it should be examined whether the grade point 
average has an effect over cheating behavior. The present study aimed to 
examine the effect of high grade point average over the cheating behavior 
and the cheating perception of cheater students to help researchers, 
educators and university’s managements to meaning accurately the cheating 
behavior.  
Method  
Participants 
In total, 493 students participated in this study from Department of The 
Faculty of Commerce and Tourism at Gazi University, located in the capital 
city of Turkey, and has second high student population in Turkey. Data were 
collected in the autumn of 2007. All students at the tourism department 
participated in this study. The questionnaires were prepared in two parts. 
One part of the questionnaires was related to the demographic information 
of the students and the other part was related to the cheating perception of 
students. A questionnaire was comprised of 37 questions. The demographic 
features of the participants and their academic terms are exhibited below. 
Table 1: Demographic Information About The Participants 
Characteristics N 
% 
Gender 
Female 
Male 
Total 
 
225 
265 
490 
 
45,9 
54,1 
100,0 
Ages  
17 and below 
18-19 
20-21 
22-23 
24 and more 
Total 
 
4 
51 
213 
182 
37 
487 
 
,8 
10,5 
43,7 
37,4 
7,6 
100,0 
The terms  
Term II  
Term IV  
Term VI  
Term VIII 
Total 
 
106 
136 
135 
114 
491 
 
21,6 
27,7 
27,5 
23,2 
100,0 
 

---

bilig, Spring / 2009,  number  49 
 
194 
Information Gathering and Analysis 
The data reported in this study is based on students’ self-reported survey 
data. All questionnaires were filled during the courses by students, under the 
supervision of the faculty members and research assistants. First of all, 
students were informed about questionnaire and instructed to respond to all 
using a items a 5- point Likert –type scale ranging from 1 = disagree to 5= 
full agree. In order to reduce data and to classify variables, factor analysis 
was applied. The main applications of factor analysis techniques are: (1) to 
reduce the number of variables and (2) to detect structure in the 
relationships between variables, that is to classify variables. Before factor 
analysis, the adequacy of data for factor analyze should be examined. For 
this purpose, Kaiser-Meyer-Okin (KMO) and Bartlett test was conducted.  
KMO value is calculated as 0,808 for adequate of sample. The KMO value 
shows that data are suitable of factor analysis. According to the results of 
Bartlett test, Approx. Chi-Square was calculated as 4717, 37 and significant 
level was p=000. The results show that sample and data are adequate for 
factor analysis.  
  As a result of the factor analysis, the nine factors were determined. The 
nine factors and their variances were given in the table 2. According to the 
table 2, the nine factors explained the 60,7% of the total variance. It means 
the nine factors can represent 34 variables.  
Table.2: Total Variance Explained 
 
Initial Eigen Values 
Extraction Sums of Squared 
Loadings 
Rotation Sums of Squared 
Loadings 
Component Total  % 
of 
Variance 
Cumulative 
% 
Total % 
of 
Variance 
Cumulative 
% 
Total % 
of 
Variance 
Cumulative 
% 
1 5,015 
14,327 
14,327 
5,015  14,327 14,327 4,415 
12,615 12,615 
2 4,570 
13,057 
27,384 
4,570  13,057 27,384 3,292 
9,406 22,021 
3 3,203 
9,152 
36,537 
3,203 
9,152 36,537 
2,559 
7,312 29,333 
4 1,919 
5,483 
42,020 
1,919 
5,483 42,020 
2,496 
7,132 36,465 
5 1,901 
5,432 
47,452 
1,901 
5,432 47,452 
2,215 
6,328 42,793 
6 1,252 
3,577 
51,029 
1,252 
3,577 51,029 
2,030 
5,800 48,593 
7 1,235 
3,527 
54,557 
1,235 
3,527 54,557 
1,670 
4,771 53,364 
8 1,125 
3,213 
57,770 
1,125 
3,213 57,770 
1,334 
3,812 57,176 
9 1,041 
2,974 
60,744 
1,041 
2,974 60,744 
1,249 
3,568 60,744 
Extraction Method: Principal Component Analysis. 
To describe the relationship between factors and 34 variables, Principal 
Components Analysis was conducted. As a result of the component analysis, 
rotated component matrix table was formed. Table 3 shows the variables 
and their related factor. 

---