Data Analysis & Cross-National Validity (jibs-data-analysis)

When to trigger

• Cross-country or multilevel data are collected and it is time to estimate and report
• You are unsure your estimator matches nesting (individuals in firms in countries) or a process design
• Reviewers will probe measurement equivalence, CMV, or (dynamic) endogeneity
• A reviewer cites a JIBS "From the Editors" methods editorial against your analysis

Establish cross-national measurement equivalence first

At JIBS, because constructs travel across countries and cultures, measurement equivalence is a first-order review concern, not an afterthought. Before testing hypotheses:

• Reliability & CFA. Report reliability (alpha/composite reliability) and a confirmatory factor analysis with fit (CFI, TLI, RMSEA, SRMR).
• Measurement invariance. Run multi-group CFA and report configural → metric → scalar invariance across countries; if scalar fails, report partial invariance or the alignment method and discuss what cross-country comparisons remain defensible.
• Construct & discriminant validity. AVE per construct; AVE > inter-construct squared correlations (or HTMT). Report the correlation matrix with reliabilities on the diagonal.
• Aggregation (multilevel). Justify any aggregation to the country/firm level with ICC(1), ICC(2), and r_wg(j).
• Qualitative designs. Where the design is case-based, "validity" becomes trustworthiness — a transparent data structure, an audit trail, and representative cross-case quotations.