Ordinary least squares for a continuous outcome: the workhorse family. Use these when your response is a measured quantity (a score, a time, a concentration) and your predictors are any mix of continuous variables, binary groups, and factors.

Examples

• Simple linear regression: one continuous predictor wage ~ years_education — one continuous predictor's effect on a continuous outcome, nothing held constant.
• Two-predictor multiple regression plant_biomass ~ rainfall + soil_nitrogen — two continuous predictors, each adjusted for the other.
• Three continuous predictors, side by side cholesterol ~ age + bmi + exercise_hours — three additive continuous slopes, mutually adjusted.
• Continuous-by-continuous moderation well_being ~ income * social_support — does income's slope change across levels of social support?
• Three-way continuous interaction growth_rate ~ temperature * moisture * soil_ph — a full three-way interaction among continuous predictors.
• Interaction-only term (no moderator main effect) yield ~ nitrogen + nitrogen:water — water bends the nitrogen slope but has no main effect of its own.
• Two groups as regression (independent t-test) pain_score ~ treatment — compare two group means as a single-binary-predictor regression.
• ANCOVA: group effect adjusting for a baseline blood_pressure ~ treatment + baseline_bp — group effect net of a continuous baseline (parallel slopes).
• ANCOVA homogeneity-of-slopes test blood_pressure ~ treatment * baseline_bp — does the treatment effect depend on the baseline covariate?
• Adjusted two-group comparison (parallel slopes) monthly_income ~ union_member + experience_years — a binary group gap holding a continuous control fixed.
• Binary-by-continuous moderation wage ~ gender * experience_years — does the experience slope differ between two groups?
• Three-level categorical predictor abundance ~ habitat — a single 3-level factor, read as its two dummy contrasts vs the reference.
• One focal predictor adjusted for covariates hourly_wage ~ years_education + age + experience_years + tenure — one focal predictor net of correlated controls.
• Factor interacting with a continuous predictor biomass ~ habitat * rainfall — does the rainfall slope differ across the levels of a factor?
• Two interacting categorical predictors (2x2 factorial) job_satisfaction ~ gender * sector — two two-level factors and their interaction (factorial as regression).
• Continuous moderation with a covariate recovery_days ~ dose * age + baseline_severity — a continuous interaction adjusted for an additive covariate.
• Ordinal predictor as a linear trend tumor_shrinkage ~ dose_level — an ordered dose (0–3) read as one slope.