Report of the conference "Quantitative Methods in Human Sciences (MQSH) 2024"

On Friday, June 7, 2024, the 13th Annual meeting on Quantitative Methods in Human Sciences (MQSH) was held at TELUQ University (Montreal). Five speakers presented their work. Eric Frenette discussed the evolution of the bifactor model, from its proposal by Holzinger following Spearman's works in 1927 to its contemporary version, akin to Frankenstein's monster, having transformed its underlying assumptions and patched parameters to enhance its explanatory potential. Sophie Vanbelle, the invited speaker, presented the main statistical measures of inter-rater reliability and their interpretation to determine reliability and agreement for binary scales. She discussed the importance of reporting different agreement indices, calculating the necessary sample size to obtain reliable measures of inter-rater (and intra-rater) agreement, and the need for guidelines for researchers. Andre Achim proposed an innovative approach to exploratory factor analysis based not on matrix decomposition but on signal cancellation, which allows for extracting all relevant information, the number of factors, and the factor structure, while reducing the number of decisions. Ibtissem Ben Alaya presented the seven-step validation process of a questionnaire on teachers' socio-emotional skills, discussing the challenges of its implementation and the psychometric choices. Louis Laurencelle, in his presentation, counter-argued the various rationales and criteria for altering the classic significance alpha threshold, such as the Bonferroni or Tukey corrections, which undermine the statistical power of research endeavors in human sciences. He examined and argued the absurdity and harmfulness of such approaches. In conclusion, this 13th edition of the MQSH meeting lived up to its banner: competence, critique, and creativity.