Table 1.

Compensatory growth rate in relation to dietary and photoperiod treatments in the Winter and Spring experiments. The full linear mixed‐effect model (LME) included season of experiment (Winter or Spring), dietary (restricted or control) and photoperiod (ambient or delayed) treatments as fixed effects, manipulated fish length (at the end of Period 1) as a covariate and tank as a random effect, plus interactions among variables. Non‐significant variables were dropped from the final model. Note that a positive estimate is associated with a faster rate of compensatory growth. The parentheses represent the reference coding of the categorical variable

Final model	Estimate ± SE	F	d.f.	P
Intercept	0·600 ± 0·098
Season (Winter)	0·065 ± 0·014	14·80	1, 98·80	<0·001
Dietary (Restricted)	0·164 ± 0·011	288·54	1, 98·80	<0·001
Manipulated fish length	−0·102 ± 0·029	12·59	1, 98·80	<0·001
Season (Winter) × dietary (Restricted)	0·046 ± 0·016	8·44	1, 98·80	0·004