All work
Multivariate Analysis · COVID-19 Research
Character Strengths & Mental Health
Used factor analysis, clustering, discriminant validation and multivariate regression to examine character strengths and mental-health outcomes.
Why this matters
Demonstrates depth beyond dashboarding: latent-variable discovery, segment validation and multi-outcome interpretation.
The question
Problem
Understand the latent structure of character strengths and how those factors relate to distress, general mental health and self-efficacy during lockdown.
The work
Approach
- 01
Used scree and parallel analysis to determine the factor solution.
- 02
Applied rotation for interpretable factor labels.
- 03
Validated K-means segments with discriminant analysis.
- 04
Modeled three psychological outcomes using multivariate regression.
Verified evidence
What the analysis surfaced
Latent factors
Variance explained
Validated clusters
Consistent predictor
Transcendence
The decision
Recommended action
- Use latent factors rather than 24 raw strength variables in downstream models.
- Treat cluster labels as analytical segments, not permanent personality types.
Capabilities demonstrated
Factor analysisClusteringValidationMultivariate modeling
Tools
RFactor analysisK-meansDiscriminant analysisMultivariate regression
Limits & responsible use
- Cross-sectional survey relationships do not establish causality.