On the Importance of Learning Statistics for Psychology Students
University of California, Los Angeles
“The quiet statisticians have changed our world; not by discovering new facts or technical developments, but by changing the ways that we reason, experiment, and form opinions” – Ian Hacking
Psychology is a very popular major for undergraduate students. Why do so many students flock to psychology programs? While the answer to this question varies by individual, reasons typically involve interest in the topic, job prospects, the appeal of certain faculty members, or the requirements for completion of the degree (Galotti, 1999). Often, students find interest in psychology because many concepts taught in introductory psychology classes are intuitively understood and directly applicable to their lives as human beings. An additional draw tends to be the avoidance of taking high-level mathematics courses. In many psychology programs, to obtain a bachelor’s degree, the only additional math course required is some form of introduction to social science statistics. However, there is quite a discrepancy between the statistics knowledge required to obtain a bachelor’s degree in psychology and what is necessary to have a career in the field of psychology. It turns out that the importance of understanding and being able to apply and interpret statistics in psychological research cannot be understated.
The public face of psychology is often represented by the therapist and on the surface, this occupation could not seem more removed from mathematics. However, this perception is quite misleading. From the development of new therapy techniques to evaluating the effectiveness of the techniques upon implementation, it is statistical analysis that provides the means by which conclusions can be drawn. While a bachelor’s degree in psychology may allow for a college graduate to obtain entry level jobs in a variety of fields (e.g. human resources, education, customer service, etc.), developing a career within the field of psychology requires a graduate degree. Further, with the exception of graduate degrees aimed at marriage and family therapy licensure, most graduate programs focus on learning to conduct, and then conducting, publishable research.
In 1990, Aiken, West, Sechrest, and Reno surveyed existing psychology programs on a number of issues to gain a sense of what was being covered in methods courses at the graduate and undergraduate levels. Generally, it was found that while analysis of variance (ANOVA) was covered at length, coverage of measurement issues and more advanced statistical methods was lacking. In revisiting this topic, Aiken, West, and Milsap (2008) found that while some improvement to the breadth of methodology courses was noted, the focus of such courses still tended to be on ANOVAs. Although some areas of psychology still rely heavily on true experiments, research in other fields of psychology often require other types of statistics beyond the use of ANOVAs. Rather, advanced techniques, such as structural equation modeling, multilevel modeling, and item response theory, are necessary to address contemporary research questions. Without the ability to apply more advanced statistical techniques, students lack the tools to conduct innovative and relevant research. Psychological research may start with a “great idea,” but this idea must be followed by a solid study design, effective data collection, and appropriate data analysis, combined with the means to analyze the data and interpret any findings. The cost of ignorance lies in the failure to optimize research design, to collect the types of data to best answer research questions, and to avoid improper analysis of data, leading to inappropriate and sometimes incorrect conclusions (Aiken et al., 1990).
As such, taking an introductory statistics course will not be sufficient in providing students with the research skills that they need. Higher level data analysis courses are necessary for success as a researcher. Most psychology programs at major universities offer courses beyond introductory statistics. Though class titles vary, a typical “advanced” statistics course will cover more complex analyses such as factorial ANOVA and multiple regression. Courses such as this provide the foundation for learning more specialized techniques that are not only more interesting, but more powerful for drawing conclusions. Many universities offer semester or quarter-length courses on structural equation modeling, where students can learn about methods like factor analysis, growth curve analysis, and multilevel modeling, which offer techniques for complex data structures and unobserved (latent) variables. Further, some courses may cover not only classical test theory, but also generalizability theory and item response theory, which constitute the future of psychological measurement. Other interesting courses may cover cluster analysis and multidimensional scaling, data mining techniques, or meta-analysis. Taking one or more of these advanced courses is extremely beneficial to undergraduates in psychology. Not only will a student learn to apply advanced techniques in his or her research, but having such courses listed on a student’s transcript gives a considerable edge when applying to graduate programs, allowing an applicant to separate oneself from the “herd.” Further, when reading contemporary research in psychology, understanding of the methods utilized engenders an enhanced ability to evaluate the implications of substantive findings.
For those who truly love the field of psychology and wish to have a career as a psychologist, statistics courses are unavoidable, but also invaluable. Fortunately, in some cases, those who believe they despise math may find themselves drawn in by the allure of techniques like structural equation modeling, which offer more eloquent ways of answering complex questions about systems of variables, rather than simple group differences. Though this might not be the case for all students, taking statistics courses even if they are particularly challenging will build the necessary skills to become stronger researchers and provide better job opportunities in the future.
Aiken, L. S., West, S. G., & Millsap, R. E. (2008). Doctoral training in statistics, measurement, and methodology in psychology. American Psychologist, 62, 32-50.
Aiken, L. S., West, S. G., Sechrest, L., & Reno, R.R. (1990). Graduate training in statistics, methodology, and measurement in psychology. American Psychologist, 45, 721-734.
Galotti, K. M. (1999). Making a “major” real life decision: College students choosing an academic major. Journal of Educational Psychology, 91, 379-387.
Jessica Tessler is a Ph.D. graduate student in Quantitative Psychology at the University of California, Los Angeles. Her research interests involve multilevel modeling, specifically studying the effects of model misspecification with cross-classified data structures. She is also a pre-doctoral fellow in the Advanced Quantitative Methods in Education Research program, which is funded by the Institute of Education Sciences (IES). In addition, she is a part-time faculty member at California State University, Fullerton, teaching a computer applications course in the psychology department.