Word Learning Paper

My erstwhile student, Ian Jones, has a new paper out on word learning in children. With his co-authors, he re-analyzed archival data to answer novel questions that were not addressed by traditional statistical procedures (t-test, ANOVA, regression). If your data are not continuous, why not look for patterns in individual discrete responses rather than analyze means and variances? Here’s the paper: Jones, Kucker, Perry, and Grice (2024). Capturing the Heterogeneity of Word Learners by Analyzing Persons, Behavioral Sciences, 14(8), 708.

Ergodicity Paper

A new paper by Speelman, Parker, Rapley, and McGann document the unfortunate bad habit among modern psychologists of analyzing aggregates and then making unjustified conclusive statements about individuals. As has now been documented in numerous papers, results based on between-person averages do not tell us how many individuals in a study actually behaved in a manner consistent with expectation. In fact, with the small effect size magnitudes of most psychological studies, it is likely that MOST participants in a given study behaved in ways contrary to expectation. Only by conducting person-centered analyses can the number of hypothesis-consistent individuals be determined.

Update Uploaded

I have uploaded an updated version of the software (October 1st, 2023). Three significant changes have been made. First, a new suite of tools has been added for importing and saving to CSV (comma separated values) format. This is a generic format compatible with Excel, R, SPSS, and many other programs. The user can now save and import data definitions along with the data themselves. This suite of methods makes moving data to and from Excel a breeze. Seond, because of these new generically functional methods, the procedures for importing data from SPSS and SAS have been deleted. Third, for the complex Efficient Cause procedures, the user can now save and retrieve the setup for the analyses. Selecting the orderings and choosing the options for these analyses can be time consuming, especially if multiple analyses are required. The new save option should save a lot of time! Finally, a number of minor bugs have been removed from the program.

Please contact me ASAP if you encounter any issues with the new version.

Social Psychology Person-Centered Paper

Psychology Is a Property of Persons, Not Averages or Distributions: Confronting the Group-to-Person Generalizability Problem in Experimental Psychology by Ryan M. McManus, Liane Young, and Joseph Sweetman of Boston College. The authors use OOM-like methods, including the c-value, in their analyses. They show that while a result might be statistically significant, most participants in a study are likely to respond in ways contrary to expectation. The authors also document that majorities of psychologists and laypeople erroneously believe a statistically significant finding means that most participants in a study have behaved in ways consistent with expectation (theoretical prediction). Reanalyzing data from over a dozen published studies, however, the authors demonstrate this belief to be erroneous. Once again, the p-value is shown to create more confusion than clarity. Will psychologists ever let it go or at least relegate it to its proper, limited role in a tiny subset of studies?

Ordinal Pattern Analysis Paper

A new paper titled Ordinal Pattern Analysis: A Tutorial on Assessing the Fit of Hypotheses to Individual Repeated Measures Data by Timothy Beechey has been published in the Journal of Speech, Language, and Hearing Research. The paper is provides an excellent introduction to OPA but is unfortunately behind a paywall. An erratum was published in March of 2023. It is certainly great to see these simple, intuitive methods gaining traction. The Ordinal Pattern Analysis has also made its way into R. Dr. Beechey’s R code can be found online here.

Person-Centered paper on social media viewing

Figure 1 in this interesting article shows clearly how “effects” can vary greatly from person-to-person. Such studies are becoming more common. I only wish the world of academia weren’t so big, as then it would be less likely that I would miss such interesting work.

Beyens, I., Pouwels, J.L., van Driel, I.I. et al. The effect of social media on well-being differs from adolescent to adolescent. Sci Rep 10, 10763 (2020).