Frequency, Methods, and Causes

Academic cheating:

frequency, methods, and causes.

Mikaela Björklund and Claes-Göran Wenestam

Åbo Akademi University

Department of Teacher Education,

Vasa, Finland

email: or

Paper presented at the European Conference on Educational Research, Lahti, Finland 22-25 September 1999.

Abstract

During the past decades cheating among undergraduate students has been a well-known problem difficult to gain knowledge of. European research in this field of research is scarce. The aim of this paper is to present a study, investigating the frequency of cheating, the cheating methods used and the students’ motives for cheating or not cheating in a Swedish-Finnish university context. Comparisons with other higher education contexts were possible since an anonymous questionnaire, worked out and used by Newstead, Franklyn-Stokes and Armstead (1995), was translated into Swedish and used in the study. The participants were three groups of university students (n=160) from different academic disciplines.

The findings implicate that cheating among undergraduates is common and mainly is a problem of ethic character. The paper also discusses consequences of student cheating for the university staff, legislators, and society. Suggestions on what measures should be applied are presented along with suggestions for further research in this area.

1.Background

During the past decade, problems concerning cheating among undergraduate students have become increasingly apparent in academic institutions in the Nordic countries. Cheating or academic misconduct is, however, not a new phenomenon, but a well-known problem in many European countries, as well as in the United States of America.

Because of the ethical and moral character of the problem it is not easy to do research in this field. Obvious problems are i.e. student integrity. Thus, academic dishonest behaviour and cheating is a familiar problem for any university, but it is often not very well known and sometimes the university authorities do not even want to know of it. Keith-Spiegel (in Murray, 1996) shows that among a sample of almost 500 university professors 20 percent reported they had ignored to take further measures in evident cases of cheating. Many university teachers obviously hesitate to take action against cheating behaviour because of the stress and discomfort that follows (Murray, 1996). Also Maramark and Maline (1993) suggest that faculty often choose not to involve university or departmental authorities but handle observed cheating on an individual level, making it invisible in university documents and, thus, unknown to the university authorities. Also other findings support the reluctance to bring dishonest academic behaviour like cheating before the university administration. Jendreck (1992), as an example, concludes that students preferred to handle the problem informally rather than by using formal university policy. Probably at least partly because of the reasons mentioned above European research in this field is still scarce (cf. Newstead, Franklyn-Stokes & Armstead, 1995 and Ashworth et al., 1997).

Nevertheless, we feel that it is of the utmost importance that this area of research is further developed in the near future, not the least since students tend to see cheating as a more or less normal part of their studies, which is illustrated in the quote below:

Students beliefs that ”everyone cheats” (Houston, 1976, p. 301) or that cheating is a normal part of life (Baird, 1980) encourage cheating. The adage ”cheaters never win” may not apply in the case of academic dishonesty. With cheating rates as high as 75% to 87% (e.g., Baird, 1980; Jendreck, 1989) and detection rates as low as 1.30% (Haines et al., 1986), academic dishonesty is reinforced, not punished. (Davis, Grover, Becker & McGregor, 1992, p. 17)

With detection rates as low as 1,3 % it is hardly surprising that students to a great extent perceive academic misconduct as worth while and even approved of. As an illustration of the low detection rates; during a five year period (1991-1995) only 24 students were brought to the disciplinary board for cheating at one Swedish university (Grahnström, 1996).

It is, hence, of importance to university staff and administrators, as well as to legislators and society as a whole to gain insight in this matter, in order to be able to do something about it.

The aim of the study

The main aim of the study presented in this paper is to provide a first step in a survey over university students’ cheating, i.e. to investigate the overall frequency, different methods and main reasons for cheating and not cheating among students in a Finnish context. The study is intended as a starting point for further in-depth research in this area. In order to get a better understanding of the problem, the aim is also to relate the outcome to subjects’ backgrounds in terms of sex, age, academic experience (number of study years), faculty belongings, level of difficulty, level of study success and main reason for study at the university, in order to get a deeper understanding of student cheating behaviour.

To make comparisons with other contexts possible an anonymous questionnaire, worked out and used by the British researchers Newstead , Franklyn-Stoked and Armstead (1995), was translated into Swedish and used in the study, which was carried out on 160 university students during the spring of 1996.

I this presentation we focus attention on:

a) the frequency of admitted cheating,

b) what kinds of cheating is most frequent in relation to the British results,

b) the relationship between frequency of admitted cheating/not admitted cheating and sex,

c) the reasons selected for or against cheating in relation to the British results, and

d) the relationship between the reasons selected and sex.

2.Theoretical introduction

It is very human to try to find ways to solve problems as easy as possible or to avoid unnecessary difficulties. Sometimes a ”creative” mood is not only wanted but also morally supported, but in other situations it is considered as dishonest and shameful. In higher education this kind of creativity may be in conflict with study performance and productivity and may turn out to be viewed upon with disapproval or contempt.

How define cheating?

Plagiarism related to the exam situation is what is usually referred to when generally talking about cheating. It is also this kind of behaviour that has received most attention in research on cheating. Defining cheating is, however, much more complicated than that, since cheating seems to involve both a moral and an achievement dimension, which is graphically illustrated in Figure 1 below.

Immoral

behaviour

LowHigh achieve- achieve-

mentment

Moral

behaviour

Figure 1. Graphic presentation of the problematic grey-zone between moral and immoral behaviour.

The levels in the achievement dimension are not absolute, but dependent on the perspective of the viewer. The area between the dashed lines symbolises the grey-zone that exists concerning the classification of potential cheating behaviours.

Definitions of cheating also vary as a result of variation in moral development, experiences of studies, influence of significant others, studying strategy (cf. Miller & Parlett, 1973) and probably also other factors. The result is a wide spectrum of definitions ranging from liberal to conservative. Hence, the need for normative documents is apparent. Evenso they do not seem to exist, at least not in Finland. Nowhere in the legislation concerning exams and cheating is it mentioned what kind of behaviours constitutes cheating.

The examples above illustrate what a complex problem cheating is. In the study presented in this paper all not strictly correct behaviours were classified as cheating for clarity’s sake.

To what extent does cheating occur?

Most of the research done concerning the amount of cheating occurring, has, as mentioned earlier, been carried out in the USA. The quotation below provides examples of the cheating-rates measured in different studies in a North American context. The reader ought to observe that these studies were different in design; concentrated on different behaviours and therefore some of the variation in the percentages might be accounted for in that way, and thus can not only be taken to convey a steadily increasing rate of cheating.

Drake (1941) reported a cheating rate of 23%, whereas Goldsen, Rosenberg, William, and Suchman (1960) reported rates of 38% and 49% for 1952 and 1960, respectively. Hetherington and Feldman (1964) and Baird (1980) reported cheating rates of 64% and 76%, respectively. Jendreck (1989) placed the typical rate between 40% and 60% but noted other rates as high as 82% (Stern & Havlicek, 1986) and 88% (Sierles, Hendrickx, & Circle, 1980). (Davis et al., 1992,s.16)

Davis et al. (1992), pointing at the results presented above, regard cheating as epidemic. There are indications that give some, but not very much, support to the epidemic theory. McCabe and Trevino (1996) found that the tendency to cheat had increased only little, from 63 percent in 1963 to 70 percent in 1990-91 but that the cheating methods had been more developed and the repertoire wider. Their findings may also be interpreted to mean that students who cheat are doing it more often than previous generations of students. Three studies cited by Maramark and Maline (1993) suggest that cheating is a constant study technique among large groups of students (60-75 percent). Also Davis and Ludvigson (1995) found in a more recent study that the individuals who cheat during their university-level studies are the ones that also have cheated earlier in their studies.

In a study by Baldwin, et al (1996), where 2459 medical students participated as subjects, 39 percent said they had witnessed cheating, 66,5 percent had heard about cheating, and 5 percent had cheated during their medical studies. Graham et al (1994) found that among 480 college students 89 percent admitted cheating and in a study by Lord and Chiodo (1995) 83 percent of the undergraduates investigated (n=300) admitted to cheating on significant tests and major projects.

In a European context Newstead et al. (1995) also present high rates of cheating. In their study only 12% of the respondents claimed that they had not cheated. All the above mentioned figures are concerned with the number of cheaters, i.e. the number of students who have at least on one occasion been involved in academic misconduct, they do not tell us anything about to what extent these people do cheat. It is, however, likely that the more cheating is done, the more probable it is that the numbers of behaviours used vary. It is therefore of importance to find out what kind of behaviours students utilise.

What methods are used?

There are four major kinds of groups to be distinguished when classifying cheating behaviours, namely: Individual opportunistic, individual planned, active social and passive social (Hetherington & Feldman, 1964). Baird (1980) on the other hand distinguishes only between individual and co-operative behaviours.

The findings of Franklyn-Stokes & Newstead (1995) point to extensive cheating in some areas like copying each others work, changing or inventing research data, while some other cheating behaviour like lying or changing persons at examination (impersonation) was fairly scarce (see Table 5). Hence, there seems to be a correlation between level of perceived seriousness of the behaviour and its frequency of occurrence – the more serious the behaviour, the less frequent it is. Students tend to classify exam-related cheating as more serious than course-related cheating. These classifications were also confirmed by Newstead et al’s results, where all exam-related items were among the least frequent and course-related items among the most frequent.

McCabe and Trevino (McCabe & Trevino, 1996) summarise their findings in a table showing what kind of cheating and the frequency students admit they are engaged in. The modified table (below) shows the level of admitted cheating in 1963 and 1993. The two tests make a comparison possible.

Table. 2.Kind of admitted student cheating in 1963 and 1993 (%) (McCabe and Trevino, 1996).

19631993DiffTendency

Tests/Examinations

Copied from another student 26 52 +26Increase

Helped another student 23 37 +14Increase

Used crib notes 16 27 +11 Increase

Written Work

Copied material without footnoting 49 54+5Increase

Plagiarised 30 26 -4 Decrease

Falsify a bibliography 28 29 +1 Similar

Turned in work done by another 19 14 -5Decrease

Collaborated on assignments

requiring individual work 11 49 +38Increase

Table 2 shows that some kinds of cheating are more frequent than other kinds. It is also interesting to find that in most of the cases the tendency is an increase of the cheating between 1963 and 1993. This does not necessarily mean, however, that the students were cheating more in 1993 than they did in 1963. Another reasonable explanation is that the students were more prone to admit cheating in 1993 than the students were in 1963.

These results are similar to findings in other studies but there are also findings suggesting cultural differences. Kuehn, Stanwyck, and Holland (Kuehn, Stanwyck, & Holland, 1990), for instance, asked students from Mexican, Arabic and US cultural backgrounds about cheating. The main focus was on three typical cheating behaviours: using crib notes, copying another student’s test, and allowing another student to copy course work. The findings suggest that there were differences between the culturally different groups of students in how they looked upon and rated cheating.

Also new technique, like the World Wide Web, is used by students in order to download papers, essays, etc produced by other students but presented to the examiner as own work. One illustration of this is a report from a Swedish university, where several students were found out using not accepted means for getting course credits among which the downloading of ready-made course works from the web was mentioned (Lunds Universitet Meddelar, 1998). Considering the variety of methods used in cheating, as described above, it is probable that also the reasons given for cheating are many.

Reasons for cheating and not doing it

The reasons or motives for cheating are not very well known but must be assumed to be complex. In a North American study of school students cheating by Anderman, Griesinger, and Westerfield (1998) it is claimed that the schools’ obsession with performance measures spurs cheating. It is suggested that classrooms that emphasise high grades and test scores may drive the students to cheat (Newsline, 1998).

Similar conclusions are reported from several investigations, where the students’ workload is found to be an important explaining factor (Lipson & MacGavern, 1993). Davis et al (1992) point out that pressures for good grades in higher education, student stress, ineffective deterrents, teacher attitudes, and an increasing lack of academic integrity are important determinants of cheating. Baird (1980) previously reported similar findings. In that study 35 percent of the students stated that they had too little time for studying for the exam and 26 percent of the students said their working load made it necessary to cheat. In a study by Singhal (1982) as much as 68 percent of the students regarded the wish to get good results as the reason for cheating. Newstead, Franklyn-Stokes, and Armstead (1995) found that 21 percent of the cheaters say it was lack of time to study that made them cheat and 20 percent explicitly stated that their cheating was a consequence of their wish to get better grades. A third frequently occurring reason for cheating was “everybody else does it” (16%), which effectively reflects students’ attitudes towards cheating. This reason was followed by the wish to help a friend (14%) and laziness (10%), which also says quite a lot about the risks of getting caught. It is obviously easier to help a friend cheat than to e.g. help the friend learn to an exam. Also Maramark and Maline (1993), when looking for causes for cheating, found that stress, competition for jobs, scholarships and admission to post-graduate programs were important determinants.

On a general level the causes or explanations identified can be organised in two classes of factors, external, and individual/personal. In Table 1 below the two super-ordinate factors and some elements/reasons mentioned in research done by Baird (1980), Davis et al (1992) and Hetherington & Feldman (1964) are presented.

Table 1. Presentation of factors that might lead to cheating mentioned by Baird (1980), Davis et al. (1992) and Hetherington & Feldman (1964).

Researcher / External factor / Personal factors
Baird / Seating order
Importance of the test
Level of test-difficulty
Unfair test
Scheduling
Supervision / Laziness
Awareness of the
performance of fellow
students
Low grades
Previously experienced
failure
A certain expectation of
success
Davis et al. / Overcrowded, great classes
Multiple-choice questions
Economic benefit / Wish to help a friend
Aversion to teacher
Hetherington
& Feldman / Difficult test
Lacking supervision
Badly organised
course / To gain social
acceptance/liking

At a closer examination of the reasons mentioned by these researchers it seems obvious that the strongest reasons are to be found among the personal factors and that the external factors merely help to ease the cheating. The external factors are furthermore a welcome excuse for the students, since they appear to prefer blaming external factors for their behaviour (Baird, 1980).

Anderman et al (1998) identified two general types of study approaches, which on a general level seem to be similar to the deep and surface approaches to studying and learning. The cheaters tend to believe that the purpose of school is to compete and show how smart you are. Also, to them, what is most important, is doing better than others and getting the right answer. They also worried about school and made use of self-handicapping behaviours, blaming others and making excuses for not performing well at school, more often than their counterparts. Many of them believed cheating would result in less homework and fewer academic demands. The non-cheating group of students, in comparison, expressed interest in their learning of science concepts and tried various problem-solving methods and sought to connect ideas.

In several previous studies it is suggested that the effect of an explicit and unanimously accepted honour code will lower the frequency of cheating behaviour (McCabe & Bowers, 1994). But honour code may have an effect in two opposite directions. A very common reason for some types of cheating is the wish to help a friend (Franklyn-Stokes and Newstead, 1995; McCabe & Trevino, 1996). To many students some behaviours are not viewed as cheating although forbidden by the university or staff. For instance, letting a fellow student borrow or copy an individual course work or a written assignment or even have a look at the answer in a test may be regarded as honest and correct behaviour. Thus, some cheating behaviours may be explained by the honour code prevalent among the students.