Background To examine whether different dimensions of parenting at different ages help small‐for‐gestational‐age (SGA) children ‘catch‐up’ the normal children in cognition and psychomotor. Methods We analyzed data of 800 children born SGA and 3,000 children born appropriate‐for‐gestational‐age (AGA) from the Early Childhood Longitudinal Study‐Birth cohort. The Two Bag Task was used to measure 2‐year or 4‐year parenting dimensions. Children's reading, math, gross motor, and fine motor scores were assessed at 5 years. Multivariable linear regression models were fitted to test the interactions between SGA and 2‐year or 4‐year parenting dimensions on 5‐year cognitive and psychomotor outcomes (dependent variables). Results There were significant interactions between SGA and early parenting on 5‐year reading, math, and fine motor scores. The gap between SGA and AGA children in 5‐year fine motor score was attenuated to null [−0.25 (95% confidence interval, −0.41, −0.09) vs. 0.03 (−0.13, 0.20)] when 2‐year parental sensitivity score increased from 1 standard deviation (SD) below mean (Mean − SD) to 1 SD above mean (Mean + SD). The gap between SGA and AGA children in 5‐year fine motor [−0.28 (−0.44, −0.13) vs. 0.06 (−0.09, 0.22)] and math [−1.32 (−2.27, −0.37) vs. 0.20 (−0.77, 1.17)] scores was also attenuated to null when 4‐year parental emotional support score increased from Mean − SD to Mean + SD. In contrast, the gap between SGA and AGA children in 5‐year reading score increased from 0.49 (−0.90, 1.88) to −1.31 (−2.55, −0.07) when 4‐year parental intrusiveness score increased from Mean − SD to Mean + SD. Similarly, the gap between SGA and AGA children in fine motor score increased with 4‐year parental negative regard from 0.02 (−0.14, 0.18) to −0.23 (−0.38, −0.08). Conclusions Early high‐quality parenting may buffer some adversity in long‐term reading, math, and fine motor skills related to SGA birth, whereas low‐quality parenting can amplify the adversity.