This column provides a brief synopsis and reference information for select recent journal publications. Note that this is only a short sampling of publications and is not intended as a comprehensive listing.
Upending Racism in Psychological Science: Strategies to Change How Science Is Conducted, Reported, Reviewed, and Disseminated
The authors of this article describe 25 areas where epistemic oppression exists and bring forth recommendations for stakeholders who conduct, report, review, and disseminate psychological science. Accountability metrics are also suggested to ensure action and accountability related to the recommendations provided.
Buchanan, N. T., Perez, M., Prinstein, M. J., & Thurston, I. B. (2021). Upending Racism in Psychological Science: Strategies to Change How Science Is Conducted, Reported, Reviewed, and Disseminated. American Psychologist, 76, 1097-1112. doi:10.1037/amp0000905
Deep Learning: A Primer for Psychologists
Deep learning has revolutionized predictive modeling but is not commonly applied to psychological data. This article provides an overview of deep learning for researchers who have a working knowledge of linear regression. Three basic deep learning models that generalize linear models are presented: feedforward neural network (FNN), the recurrent neural network (RNN), and the convolutional neural network (CNN). Examples with R code are provided to demonstrate how each model can be applied to answer prediction-focused research questions.
Urban, C. J. (2021). Deep Learning: A Primer for Psychologists. Psychological Methods, 26, 743-773. doi:10.1037/met0000374
The Evidence Interval and the Bayesian Evidence Value: On a Unified Theory for Bayesian Hypothesis Testing and Interval Estimation
The authors bring forward a new theory for Bayesian hypothesis testing and interval estimation. A Bayesian evidence interval is proposed which is inspired by the Pereira–Stern theory of the full Bayesian significance test. The authors assert that this proposed interval solves the problems of standard Bayesian interval estimates and that it unifies Bayesian hypothesis testing and parameter estimation. An R package is available for computations related to this framework.
Kelter, R. (2022). The evidence interval and the Bayesian evidence value: On a unified theory for Bayesian hypothesis testing and interval estimation. British Journal of Mathematical and Statistical Psychology, doi:10.1111/bmsp.12267
Reliability coefficients for multiple group item response theory models
The reliability of scores is population dependent and may vary across groups. In item response theory, this population dependence can be attributed to differential item functioning or to differences in the latent distributions between groups and needs to be accounted for when estimating the reliability of scores for different groups. The authors introduce group-specific and overall reliability coefficients for sum scores and maximum likelihood ability estimates defined by a multiple group item response theory model. Confidence intervals are derived using asymptotic theory and were shown to be unbiased and accurate with moderately large sample sizes through a simulation study.
Andersson, B., Luo, H., & Marcq, K. (2022). Reliability coefficients for multiple group item response theory models. British Journal of Mathematical and Statistical Psychology, doi:10.1111/bmsp.12269
Guidelines for the “Publication Corner” Column
Have you seen a recent publication that you think would be of great interest to Division 5 membership? Please feel free to share suggestions for publications to include in this column. Please send your ideas to The Score Editor, Gill Sitarenios. Please be sure to include full reference details and a short synopsis (2-3 sentences) of the publication. Use “Publication Corner” in the subject line of the email.