Hello William Dale Van Lente

In my opinion, there is no problem with this. (Of course, it could be said to some extent that the number of factors is large) However, the results are appropriate, especially since you say " factors seemed to have good construct and content validity based on focused interviews and qualitative responses. "

You can reasonably justify having 12 factors in your analysis, given the details you provided, but it’s essential to ensure the solution is methodologically sound. Here’s how you can address this:

1. Assessing the Number of Factors

Eigenvalue Criterion (Kaiser Criterion): While it is common to retain factors with eigenvalues greater than 1, this criterion can sometimes overestimate or underestimate the number of factors. Without eigenvalues available, alternative methods to justify the factor structure may be helpful.

Alternative Approaches to Determine Factor Count:

Scree Plot: If you can access a scree plot from the analysis, it may help determine the point where the eigenvalues drop sharply (“elbow”), suggesting the optimal number of factors.

Parallel Analysis: This method compares the observed eigenvalues to those from a random dataset. It is one of the more robust methods for deciding the number of factors.

Variance Explained: Typically, retained factors should explain at least 60–70% of the total variance in social science research. You might check this from your original output if available.

2. Sample Size Considerations

With 670 respondents and 92 questions, your sample size is robust. Common rules of thumb include:

N ≥ 300: Generally sufficient for stable factor structures.

Ratio of Cases to Variables: A 5:1 or higher ratio (cases: items) is preferred. Your ratio (670:92) is ~7.3:1, which is more than adequate.

This supports your ability to extract up to 12 factors without compromising reliability.

3. Factor Loadings

Factor loadings ranging from 0.42 to 0.77 are reasonable, indicating moderate to strong relationships between items and factors. Typically, loadings >0.40 are considered meaningful in social sciences, provided cross-loadings are minimal.

4. Construct and Content Validity

Your qualitative validations (e.g., interviews and open responses) add credibility to your factor solution. These insights are critical for contextualizing the factor structure, especially if eigenvalues are missing.

If your factors align logically with theoretical constructs or operational definitions, this strengthens your case for 12 factors.

5. Suggestions for Reanalysis

If possible, reconsider re-running the factor analysis to:

Recover eigenvalues.

Conduct a parallel analysis or scree plot to corroborate the number of factors.

Software like SPSS, R, or Python can handle this reanalysis. If the original dataset isn’t available, documenting your rationale based on past analysis (e.g., factor loadings, qualitative support) is valid.

6. Publication Considerations

When preparing for publication:

Explain the original methodology clearly, including the justification for retaining 12 factors (e.g., variance explained, loadings, content validity).

Address the absence of eigenvalues transparently. For example:

Although eigenvalues were unavailable, the factor structure was supported by factor loadings, qualitative validation, and explained variance consistent with theoretical expectations.

If reviewers question the methodology, you might propose conducting a reanalysis with similar or simulated data to confirm the findings.