The aim of this study was to develop a model to explain the factors influencing parenting stress in mothers of premature infants, and to verify the appropriateness of the model. The exogenous variables consisted of infant temperament and maternal positive psychological capital (PsyCap). Endogenous variables consisted of parenting efficacy and parenting stress. The participants included 233 mothers of premature infants, with a corrected age under 24 months. Data were collected from March to September 2013, using self-reported questionnaires. The model fit indices suggested that the original hypothesized model adequately fit the data. The final model accounted for 39 % of the variance in parenting stress. Furthermore, we found that parenting stress was directly and indirectly affected by infant temperament, and their mother’s positive PsyCap via parenting efficacy. Premature infant’s negative emotionality was moderately related to higher parenting efficacy and lower parenting stress. Mother’s positive PsyCap directly affected parenting efficacy and parenting, and indirectly affected parenting stress through parenting efficacy. Furthermore, parenting efficacy directly affected parenting stress. These results suggest that nursing strategies based on the parent–child interaction model are needed for promoting positive maternal mental health and parenting efficacy, to reduce parenting stress in mothers of premature infants.