Publications

Shaheed Azaad (2026) · Behavior Research Methods

Theoretically driven meta-analyses often involve testing whether study characteristics moderate an effect in a pattern that supports one of multiple competing theoretical accounts. To conduct such analyses, researchers must manually extract, i.e., code, these characteristics from sometimes hundreds of studies. Ideally, studies should be coded by two researchers independently to prevent errors and biases from compromising the dataset’s integrity. The laborious nature of this task, however, means that meta-analysts usually settle for double-coding just a portion of the studies included in their study. Some researchers have proposed using large language models (LLMs) as a double-coder, showing that they can reliably extract explicitly stated information (e.g., publication year) from articles. However, meta-analyses in psychology generally require studies to be coded in terms of higher-leve...

John Michael, Shaheed Azaad, Pernille Hemmer, Robrecht van der Wel (2026) · Preprint

A commentary on Zapparoli et al. (2022; Cortex)

Shaheed Azaad, Nick Haslam, Yoshihisa Kashima (2026) · Preprint

Past research has concluded that stereotype threat effects cause female chess players to underperform against male opponents. Here, we investigated whether this effect is large enough be practically meaningful, and whether it varies in line with the stereotype threat account. We analysed moves from 118,053 tournament chess games (N = 29,864 players), to test for a player × opponent gender interaction on performance, whether mixed-gender games were played more aggressively, and whether female players performed better in female-only tournaments. We also tested for moderation by the Gender Inequality Index of a player’s country, a player’s birth year, and the year in which a game was played. Equivalence testing (bounds: β = ± 0.10) found all effects to be unsubstantial. Results suggest that female tournament chess players do not experience stereotype threat effects, and that gender dispari...

Shaheed Azaad (2026) · Preprint

The typical approach to computing standardised mean differences (d) from paired-samples (or within-subjects) designs requires knowing the often-unreported repeated-measures correlation, r_"repeated" . This adjustment enables comparison with ds from independent samples. However, when meta-analysing effects that come exclusively from paired samples, an underutilised option is to compute Cohen’s d_z, which does not require r_"repeated" . Because d_z can be computed from a range of summary and inferential statistics, it enables researchers to obtain standardised effects even when articles report their results in little detail. The present article contains equations and the corresponding R code needed to compute d_z and its variance, and adjust for small-sample bias.