Description
Models how an outcome varies with predictors, estimating effect sizes and enabling prediction of satisfaction, retention or conversion from user and product variables.
Regression analysis models an outcome variable as a function of one or more predictors. Linear regression handles continuous outcomes such as satisfaction scores, logistic regression handles binary outcomes such as conversion or churn, and mixed-effects models handle repeated measures and nested data such as tasks within participants. In UX it is used to estimate which factors are associated with outcomes, control for confounders, and predict outcomes. Rigour comes from specifying models in advance, checking assumptions such as linearity, independence, homoscedasticity and multicollinearity, using appropriate models for data structure, reporting effect sizes with confidence intervals rather than only p-values, validating predictions on held-out data, and recognising that observational regression shows association, not causation.
Attributes
- Discipline
- Statistics & Modelling
- Loss type
- Active
- Stage
- Evaluation, Post-launch
- Data
- Quantitative
- Effort
- Days
Steps
- Define the outcome, predictors and covariates based on the research question.
- Prepare data, handle missing values and code categorical variables.
- Choose the model type for the outcome and data structure (linear, logistic, count or mixed-effects).
- Fit the model and check assumptions with residual plots and diagnostics such as variance inflation factors.
- Interpret coefficients as effect sizes with confidence intervals.
- Validate predictive performance on held-out data where prediction is the goal.
- Report results with clear statements about causal limitations.
Sources
- Statistics for Business and Economics
- Regression and Other Stories (Gelman, Hill & Vehtari)
- Applied Predictive Modeling (Max Kuhn & Kjell Johnson)
- R for Marketing Research and Analytics (Chris Chapman & Elea McDonnell Feit)
- Applied Logistic Regression (David Hosmer, Stanley Lemeshow & Rodney Sturdivant)
- Applied Predictive Modeling (Kuhn & Johnson)
- R for Marketing Research (Chapman & Feit)
- The Inquiry Series, Ch 43 — Regression, Moderation & Mediation for UX Researchers
- Practical Statistics for Data Scientists (Bruce & Bruce)
- Companion Book to Quantifying the User Experience (Lewis & Sauro)
- Benchmarking the User Experience (Sauro)
- Factor and Discriminant Analysis (Sauro & Lewis, MeasuringU training)
- Myatt, G., Making Sense of Data: A Practical Guide to Exploratory Data Analysis and Data Mining (2007)
- Veal, A. J. & Darcy, S., Research Methods in Sport Studies and Sport Management (2014)
Also known as
- Linear regression
- Logistic regression
- Mixed-effects models
- Multiple regression
- Logit model
- Binary logistic regression
Categories
- Analytics & Behaviour
- Statistics & Modelling