The university student as a study subject in psychology

Introduction #

Experiments in modern Psychology and the need for Ethics of Participation #

The science of Psychology has been riddled by countless ethical dilemmas through time. Performing responsible and ethically congruent science has been especially difficult for Psychologists, as the necessary frameworks have been missing for many years, and the anticipation of exploration and knowledge is a motive more than able to push someone “over the edge”.

It is no secret that Psychology throughout the 20th century has become a very famous and respected science through some very ethically dubious scientific experiments and studies. A glaring example, J. B. Watson’s Little Albert experiment was conducted by essentially inducing conditioned fear to an 11-month-old child, found in an orphanage (1920). Yet, Watson is recognized as one of the pioneers of Behaviourism and indeed, his research has been very influential. In the same spirit, Milgram’s experiment on the social phenomenon of obedience (1963), as well as Zimbardo’s Stanford Prison experiment (1973) on the social phenomenon of conformity, are two more examples of astonishingly unethical, yet well-known, experiments.

These experiments provided some very controversial results at their times and it is probable that this combination of ethical controversy with fascinating results made them so famous. Yet, apart from including an abundance of psychological and physical violence, they also have another, more subtle, but very valid reason, they could not be performed today. And this is the lack of participant’s Informed Consent.

Specifically, the qualms highlighted by such experiments (Schlenker & Forsyth, 1977) eventually lead to the American Psychological Association’s Code of Ethics (2017) to include paragraphs on Informed Consent (p. 8.02) and Offering Inducements for Research Participation (p. 8.06). Such paragraphs would totally nullify studies such the ones mentioned. And this is because the subjects of these studies were either forced, coerced or paid to participate. They did not decide using their own motivation, and this is not only a violation of autonomy, but can also produce bad scientific results.

The case of university students as study participants #

Designing a study is one thing, finding the necessary sample size, willing to go through it, is another. The most common place one can look for study subjects in the academic setting is in the universities themselves. And what is better than doing research while also showing future scientists the process of studying and experimentation on the way. Yet, looks like this, while perfectly convenient, does not come without problems. In the remainder of this essay these problems will be explored and analyzed.

Effects on Students #

Recruitment and the possibility of Coercion #

It has been reported that several techniques of coercion are used to “lure” university students to participate in studies. Bowman & Waite (2003) mention the idea of participating in studies, not as a compulsory task, but as an option with alternatives. This concept has also been explored by Walker (2020), but reported as a coercive technique, as the alternative options seem to be a lot more costly in terms of time and effort. Essays and journal summaries are some examples. Additionally, it is stated, that peer-pressure can create “perceived coercion”, if the recruitment takes place by a figure of authority, such as a professor or within a group (Leentjens & Levenson, 2013; Walker, 2020). Additionally, Leak (1981) showed that coercion does not go undetected by the students, effectively making them aware of their lack of autonomy.

Learning through Participation #

 The literature on this subject also argues about the experience of participating in a study, as an inherently educational experience (Bowman & Waite, 2003; VanWormer et al., 2014). This has been backed by studies, such as the one by VanWormer et al. (2014), were students self-reported through a questionnaire, that they gained “hands-on experience” on research. Yet, it is notable that this study ruled out possible coercion of participation to it, through an alternative assignment. Additionally, participation in research is not only considered to educate about the content of the studies, but more importantly, about research itself, as a process (Darling et al., 2007; Zannella et al., 2019).

In the other hand, Walker (2020) is critical of this idea. One of his points is that such results rely on students self-reporting their experience. This idea is doubled by Leentjens & Levenson (2013), coining the term “educational misconception”, which could be briefly described as “students feel they learn because they are being told so”.

Effects on Research #

While student morale is very important and should not be undermined in terms of ethics, the debate of student participation in research goes beyond that. The question that arises next is whether the practice of student participants produces science of good quality.

Sample Generalization of Students #

As Zannella et al. believe that the cultural and socioeconomic background of university students should prevent the studies that use them as subjects from extrapolating the results to the general population (2019). Specifically, they mention the acronym of WEIRD (Western, Educated, Industrialized, Rich, and Democratic) countries, that get constantly over-represented in studies (Zannella et al., 2019).

Additionally, Arnett has shown, in his monumental article titled The neglected 95%, that USA universities, which contribute a huge amount of research and also heavily rely on students as study subjects, create knowledge that lack representability, ending up in “neglecting” the vast majority of the human race, in other words, the recipient of all this knowledge, hence the title (2008).

Self-Selection Bias #

In the case of non-compulsory participation, while the student autonomy is not threatened (excluding the cases of coercion, explained above), the freely participating subjects could provide less-desirable results. Specifically, the students that voluntarily participate seem to concentrate on specific demographics, such as socio-economic status, or gender, as well as characteristics, such as being adventurous, intelligent and others (Sharp et al., 2006).

Self-selection can also create skewed samples, when it comes to participation rewards, in either academic credit or even money. It can be more likely for students that need academic points to participate, over-representing cases with lower grades (Leentjens & Levenson, 2013).

Motivation as a hidden variable #

It is commonplace for universities to award academic credits to students that participate in studies. A study by Sharp et al. (2006) has greatly shown that, indeed, reward does increase participation. Yet, the study goes on and differentiates between types of motivation. This differentiation, borrowed by the Self-Determination Theory (Deci & Ryan, 1980) splits participants into the “intrinsically motivated” and “extrinsically motivated” categories. To create a sample size that can be properly generalized, both categories should be represented, meaning that reward can play a role in the results. While eliminating rewards would create an under-represented “extrinsically motivated” category, using them can also skew the sample, leading students, that are in need of the reward to participate more, as described above (Sharp et al., 2006).

Purposeful contamination of the results #

Finally, compulsory participation can be considered as the last nail in the coffin of student participation in research. Mandatory participation, not to mention several participations during an academic year, has shown that students pressured to attend studies can change the results out of pure spite. Gustav (1962) concluded that, not only students, mandated to be study subjects, felt bored and pressed, but they also had a tendency to contaminate the results with twisted responses.

Additionally, Moyer & Franklin (2011) mention a case where a participant of an EEG study provided the following feedback: The first two tests were very long, repetitive, and boring. The arrow test was the worst of them all because after a while it started to hurt my eyes which led me to stop caring and just clicking the same button over and over, clearly undermining the study.

Conclusion #

To summarize, using university students as samples for all kinds of studies is a convenient and cheap way to get data. Yet, it comes with costs, both to the students’ morale (Nimmer & Handelsman, 1992), and the results scientific themselves (Gustav, 1962; Moyer & Franklin, 2011).

Finally, it is safe to assume that students should be used as study subjects only when the study does explicitly need a subject group of students to test a hypothesis. As utopical as this can be, it seems the only way to consistently produce “good science” for everyone. Or, as Leentjens & Levenson (2013) put it: Whether or not students are suitable subjects for a particular study will depend on the research question.

References #