Combining exploratory and confirmatory factor analysis produces theory-driven results robust across samples

Evidence

In addition, our methodological choice that combines EFA and CFA is more appropriate for this study. First, both EFA and CFA are theory driven and mutually complementary (Kline 2016). Analysis results will be more theoretically understandable and robust than data-driven analysis using EFA as the single methodology. — Ruiqian Li and Paul Froese, The Duality of American Christian…, 94

Categorization

• XREF: connects to: Confirmatory factor analysis on a new sample validates scales better than exploratory analysis alone — Sherman A. Lee, Chelsea A. Reid, Stephen D. Short, Jeffrey A. Gibbons, Ruth Yeh, and Matthew L. Campbell, Fear of Muslims: Psychometric…, 23; Combining EFA and CFA tests both exploratory structure and confirmatory fit of CN dimensions. — Ruiqian Li and Paul Froese, The Duality of American Christian…, 93; Converging multiple research methods strengthens confidence in a finding (+6 more)

Broader Connections

• Combining EFA and CFA tests both exploratory structure and confirmatory fit of CN dimensions. #exemplifies (0.92) — Ruiqian Li and Paul Froese, The Duality of American Christian…, 93
• Confirmatory factor analysis on a new sample validates scales better than exploratory analysis alone #corroborates (0.85) — Sherman A. Lee, Chelsea A. Reid, Stephen D. Short, Jeffrey A. Gibbons, Ruth Yeh, and Matthew L. Campbell, Fear of Muslims, 23
• Cherry-picking scale items can inflate apparent construct validity #qualifies (0.82) — Bob Altemeyer, The Authoritarian Specter, 3949
• Replicated findings across parent and student populations confirm scale validity #exemplifies (0.78) — John T. Jost and Jim Sidanius, Political Psychology, 1036
• Single-sample reliability evidence cannot establish a scale’s trustworthiness #corroborates (0.78) — Sherman A. Lee, Chelsea A. Reid, Stephen D. Short, Jeffrey A. Gibbons, Ruth Yeh, and Matthew L. Campbell, Fear of Muslims, 22
• Robust causal relationships would survive differences in measurement choices #expands (0.76)
• Converging multiple research methods strengthens confidence in a finding #expands (0.75)
• Latent profile count selection blends fit statistics with theory. #corroborates (0.75) — Erin Cooley, Jazmin L. Brown-Iannuzzi, Nava Caluori, Nicholas Elacqua, and William Cipolli, White Americans who perceive themselves…, 37
• Empirically-selected dogmatism scale items fail to replicate across samples, performing at chance level #complicates (0.74) — Bob Altemeyer, The Authoritarian Specter, 4262