Michael Beltz, Department of Philosophy and Religion, University of North Daktoa
Much like Bill Caraher’s introduction to this topic, it feels appropriate to me to start with the most recent outcry about the rise in cheating by today’s youth. In a recent article published in the Grand Forks Herald, Helmut Schmidt describes the high likelihood for individuals under the age of 17 to be more willing to engage in lying. The data that Schmidt relied on for this outrage blends attitudes towards lying and cheating together to reach the conclusion that high school students are more willing to be dishonest than individuals over the age of 50. There is quite a bit of data available about academic dishonesty at all levels. In some ways that data can be very disheartening. In a 2002 book on his research into cheating, Willfried Decoo argues that 70% of high school students in the United States have cheated at least once in their high school career. Furthermore, 56% of middle school students have done so as well. But cheating is not just occurring among individuals who are under the age of 17. In the most definitive work on the subject, Donald McCabe, Kenneth Butterfield, and Linda Trevino, present some sobering statistics about cheating. They start be defining serious cheating as “plagiarism, copying or using crib notes on a test, or turning in work done by another.” Their data reveals that 56% of graduate business school students participated in serious cheating in the previous academic year; 47% of all graduate students, 45% of law school students, and 48% of graduate students in education all report serious cheating in the previous academic year.
While this data may be depression to some people, I am left with a feeling of optimism. I am left asking the question: Why don’t more students cheat? I certainly would like to have students of all academic levels NOT engage in these types of behaviors. I think that academic dishonesty does have negative consequences (for the students, for the instructors, for future employers, and for the institution of higher education). But this does not eliminate the issues of figuring out why some students decide to cheat and others decide not to.
Schmidt’s article points to a sense of entitlement among today’s youth. In his comments to Caraher’s introduction to this subject, Evan Nelson, describes this as a transactional model of higher education. Cynthia Prescott gives the most developed description of this understanding of student cheating in last week’s post. She describes the transactional model as “a transaction in which completing a certain quantity of work should get them a certain grade. They believe that those grades will, in turn, get them a diploma, which will aid them in their ultimate goal of a more desirable or better-paying job… This makes it even more likely that college students will look for the easiest way out on school assignments.” While this model may be intuitively applicable, it seems to be flawed in significant ways. First, this transactional model may be able to describe why students commit serious cheating. The problem is that it does not explain why some students do not cheat. What is it that makes one student cheat and another not cheat? If both of these students were in the same class, had the same major, and would be seeking the same employment, then they would presumably have the same incentives, desires, and goals that would lead to cheating. The best that we could say is that some students buy into the transactional model and others don’t. This answer does not seem satisfactory to me, because it simply moves the question to: why do some students buy into the transactional model and others do not? The second problem with the transactional model is that it does not seem to adequately connect with the data on cheating. Let us consider the aforementioned data on graduate students. The transactional model would indicate that roughly half of all graduate students see a graduate degree as merely an exchange of money for a higher-paying job. I do not doubt that there are graduate students in all programs that feel this way. The problem is that there are only minor differences between the rates of cheating across the different disciplines. The rates of cheating among business graduate students (who are trained to approach problems from an economic standpoint) are only slightly higher than education graduate students (who are inculcated with the intrinsic value of learning). The only way the transactional model can explain the similar results is to state that half of the graduate students in education see education as merely a transaction. If this is true, then we should probably not be worried about cheating in schools. If both the teachers and the students see education as just a transaction, the problem those teachers would have with cheating is that they feel slighted, not that the action was unethical in any way. I also do not feel that the data on the other end of the educational data corresponds with the transactional model. Among middle school students, 56% also self-report serious cheating. This is the same percentage as the MBA graduate students that cheat. For the transactional model to work, we would have to say that the same percentage (of 11 through 13 year olds) see grades as a transaction for a higher-paying job as do the adult graduate students. My third criticism of the transactional model is that it does not seem historically accurate. In the second part of her post, Prescott argues that we see an increasing amount of cheating because it has become easier in the Internet age. Part of Caraher’s speculations is that the increase in cheating might be caused by increased cynicism, changes in the focus of educators, or even that we are in a ‘post-everything nihilism.’ The premise of these speculations is that there have been changes that have led to an increase in the ease and amount of cheating. Unfortunately, the best data on this subject does not support these claims. As reported by McCabe et al., the most comprehensive research on cheating in previous generations, conducted by William Bowers in 1964, found that slightly over 50% of college students committed serious cheating in the years of his study (from 1956 to 1963). In their 1990 study, Emily LaBeff et al. found that 54% of college students reported at least one instance of cheating in the previous academic year. The McCabe et al. (2006) study found that 66% of college students reported at least one instance of cheating in the previous academic year. While there is an increase in the amount of cheating over time, this data seems to be a problem for the transactional model. It would mean that half of college students in the late 1950s already had the sense of entitlement described by Schmidt. This would also mean that the rise of structural and theoretical changes in higher education from the 1950s to the 1990s only accounts for a 4% increase in cheating. It would also mean that the rise of the information age only accounts for a 12% increase in cheating.
There is no doubt that some proponents of the transactional model would be able to look at this data and see it as supportive. I, however, would like to propose an alternative model that may explain these challenges in a simpler way. We might consider this a habit model of cheating. Instead of seeing cheating as a function of a conscious exchange of work for a higher-paying job, we should consider cheating as a set of reinforced behaviors that are learned. Instead of being left with the question of why some students buy into the transactional model and others do not, we end up with the question of why some students develop the habit of cheating and others develop a habit of not cheating. This second question seems explainable in a way that the first one is not. Because each student is trained in different classrooms and environments throughout their career, they will each have developed different habits. This would mean that instead of trying to figure out individual student’s psychologies (as the transactional model would need to do), we should instead look to examine what educational practices promote the habit of cheating and which ones promote the habit of not cheating. If we see cheating as a habit, we also benefit from certain realizations that apply to all habits. For example, every successful activity reinforces the habit being practiced. Thus, every time a student cheats, it becomes easier to cheat the next time. Every time a student does not cheat, the easier it is to not cheat in the future (assuming that both of these result in success).
There are two specific situations where I feel that the habit of cheating is being reinforced at the university level. I believe that there are many more places the habit of cheating and not cheating are reinforced, but the following two areas are the most blatant.
1) Introductory science laboratory settings- Most universities require an introductory course in science with a laboratory component. Students are often graded, in the labs, based on upon how close their results are to the expected lab results. These experiments can be time consuming. Equally important, laboratory space can be limited on many campuses, making it difficult for students to take extra time during the labs or start the experiments from scratch on a different day. These factors combine to create a large amount of incentives for students to get experiments correct the first time they try them. Students are also told how to determine the results that they should expect from the experiment. The problem occurs when a student makes a mistake conducting the experiment. In these cases, the students have the incentives to fabricate the data, the tools to create expected results, and they are typically only tested on the theoretical components of the experiments on their exams. If a student decides to not cheat on these labs, the only way they will be successful (and reinforce the habit of not cheating) is if they never make a mistake in any of their experiments. This is a very high threshold for developing this habit. On the other hand, cheating to make sure that your data is close to the expected data leads to success whether the student makes errors or not. Students are often told that the most important aspect of the laboratory lessons is that they learn the theories being demonstrated. This is reinforced by the theoretical questions on exams and a low tolerance for error in the grading of the lab work. This form of fabrication is not tolerated by professional researchers, but it is subtly encouraged when those future researchers are first being trained in the long-term habits of scientific research.
2) When we discipline students that have been caught cheating – One of the questions that we must face with student cheating: what is the most appropriate way to punish the transgressor? The transactional model does not give us a straightforward way to determine what is appropriate. By shifting to the habit model, we can see why some forms of reprisal are more appropriate than other forms. When we discover that a student has cheated, we need to make sure that the actions taken by the educator promote the right habits. Too often educators take steps that teach a student that they do not want to get caught cheating, instead of that they do not want to cheat in the first place. The transactional model seems to promote the importance of not getting caught cheating. The tacit exchange involved in the transactional model is often reinforced in the classroom setting. When an instructor catches someone cheating, they will often penalize the student by giving them a zero on the assignment (or less often, failing them for the entire course). The rationale for this is that the student violated the social compact involved in the transactional model. If the student does sufficient work they will get a certain grade. If they do not do sufficient work (by cheating for example), they receive no credit for that work. In this regard getting caught cheating would have the same punishment as any action that violates the transactional compact. If a student falls asleep during class, they have violated the requirement for sufficiently meeting the transactional expectations, so they receive no points for participation that day. While the punishment may seem more detrimental in the case of cheating, that is only a product of the assignment being worth more points. The importance of this comparison is that this form of punishment does not necessarily promote the habit of not cheating. From the student’s perspective, if she falls asleep in class she will suffer the same type of punishment as if she is caught cheating. This student does not recognize that the harm of cheating is larger than just getting caught. Thus, when we connect cheating to a grade, we are promoting the habit of not getting caught cheating, as opposed to the habit of not cheating. This means that we need to radically rethink about how institutions (and instructors) deal with cheating. Students need to be taught that the harm occurs when a student cheats, even if the student is never caught. Students need to be taught the larger ramifications of cheating and not taught that it is just their grade that will suffer if they are caught. Some students have learned this, which explains why so many students decide not to cheat. However, institutions of higher education need to be more systematic about developing these positive habits. When a student cheats, it is not their grade that is harmed; it is the institution as a whole and the entire student population. As such, to decouple grades from cheating, the punishment for cheating should be dealt with at an institutional level.
