Is psychological science progressing? Explained variance in PsycINFO articles during the period 1956 to 2022.

Area of Science:

• Psychological Science
• Quantitative Psychology

Background:

• Modern psychological science has shown a consistent average explained variance of 42.8% across studies from 1956 to 2022.
• This stagnation in explained variance may be linked to methodological conventions and the semantic properties of psychological variables.

Purpose of the Study:

• To numerically assess the progress of psychological science by analyzing explained variance.
• To investigate the role of methodological conventions and semantic properties in limiting explained variance.

Main Methods:

• Analysis of 1565 articles published between 1956 and 2022 to determine average explained variance.
• Latent semantic analysis (LSA) applied to 50 studies to replicate semantic factor structures of 205 constructs.

Main Results:

• Average explained variance remained constant at 42.8% over the study period.
• Methodological conventions appear to impose mathematical constraints, clustering explained variance around 40%.
• Hypotheses with high or low semantic truth values are unlikely to be studied, favoring those around 40% truth value.

Conclusions:

• The consistent explained variance suggests findings might be predictable from semantic structures alone, potentially without empirical data collection.
• Psychology's shift from individual, causal methods to group-level associations may stem from the complexity and dynamic nature of psychological processes.
• Current research paradigms may inherently limit theoretical progress, necessitating a move beyond the 40% explained variance barrier.