Data Cleaning & Preparation
Missing values, outliers, recoding, merging and reshaping — the unglamorous work that decides whether the analysis is trustworthy.
Data & Statistics
EFA, CFA and full structural models in AMOS, SmartPLS, lavaan or Mplus — with fit indices reported properly.
Structural models are a two-stage job. The measurement model has to hold before the structural paths mean anything, and a great many submitted models skip straight to the paths.
We establish the measurement model first — reliability, convergent and discriminant validity — then estimate the structural model and report the full set of fit indices with the thresholds you are being judged against.
Broadly: covariance-based (AMOS, Mplus, lavaan) for theory testing with larger samples and reflective constructs; PLS-SEM (SmartPLS) for prediction, smaller samples, formative constructs or complex models. We will recommend one for your specific model.
We diagnose why — usually a misspecified measurement model, cross-loading items or a missing path. Modification indices are used cautiously and only where theory supports the change.
Related
Next step
Send the brief, the deadline and anything you already have. You will get a reference number straight away and a considered reply, usually within one working day.