Statistical analysis service
Regression Analysis and Generalised Linear Models
Plan, diagnose and interpret regression models with attention to coding, assumptions, effect estimates and uncertainty.
Request statistical guidanceRegression begins with model structure
Linear, logistic and generalised regression models answer different questions. The correct model depends on the outcome distribution, link function, sampling structure and relationship between predictors and outcome.
Review areas
- coding and reference categories
- linearity where required
- multicollinearity and influential observations
- residual or model diagnostics
- interaction terms and confounding
- effect estimates with confidence intervals
Interpretation
Regression coefficients should be interpreted in the units and context of the model. Statistical significance alone is not enough; magnitude, uncertainty and model fit also matter.
Request guidance for this analysis
Describe the study context so the method, assumptions and scope can be assessed before any commitment is made.
