Fisher’s LSD and Tukey’s HSD aren’t enemies; they’re personalities. LSD is powerful and liberal… great at finding differences with three simple groups, riskier with many. Tukey balances caution and power… steady, trusted, and happiest with equal sample sizes. Bonferroni is the seatbelt you never unbuckle; Sidak, a sleeker cousin. Games-Howell is your friend when variances misbehave. The rule isn’t “which is correct?” but “which fits this design, risk, and claim?” If a Type I error would mislead practice (think clinical claims), lean conservative. If you’re mapping subtle psychological effects in a tidy three-group design, LSD can be defensible. Pre-register the choice, state assumptions, and report degrees of freedom with p and effect size. Courage and caution aren’t opposites… they’re a matched set.
