Using item-level analyses to better understand the consequences of partialing procedures: An example using the Dark...

Area of Science:

• Psychological research methodology
• Personality psychology
• Quantitative psychology

Background:

• Partialing procedures are common in psychological research for controlling variables.
• Replicability of results derived from partialing is crucial for scientific validity.
• The Dark Triad (DT) measures are frequently used in personality research.

Purpose of the Study:

• To investigate the replicability of partialing procedures in psychological research.
• To examine the consistency of residual variable content across different samples.
• To assess the impact of sample size on the replicability of partialing-based results.

Main Methods:

• Utilized popular Dark Triad (DT) scales (Machiavellianism, narcissism, psychopathy).
• Compared residual content of DT scales across different sample sizes (Small, Moderate, Large) drawn from a larger MTurk sample.
• Analyzed residual variable content using individual items from the IPIP-NEO-120 for finer-grained analysis.

Main Results:

• Observed low convergence between original and residual DT scales within samples.
• Found that the content of residual Dirty Dozen scales was not similar across different samples.
• Identified comparable results for the short Dark Triad-Machiavellianism scale only in moderate and small samples.

Conclusions:

• Partialing procedures present significant issues, particularly concerning the replicability of residual variables.
• The study highlights the need for further research into the reliability of residual-based findings.
• Researchers should exercise caution when interpreting results derived from partialing methods.