Population growth models are integral to ecological studies by providing estimates of population performance across space and time. Several models have been developed that estimate population growth through correlates of demographic traits, as measuring each parameter of the model can be prohibitive in experimental studies. Since differences in female size can accurately reflect changes in fecundity for many taxa, Livdahl and Sugihara developed a population growth index that incorporates size-fecundity relationships as a proxy for fecundity. To investigate the extent to which this model is robust to variation of this proxy, we tested if genetic (source population), temperature and resource treatments affect the size-fecundity relationship in Aedes albopictus (Skuse), the Asian tiger mosquito. We then determined if variation in the size-fecundity relationship alters the population growth estimates, lambda (λ'), when applied to Livdahl and Sugihara's model. We performed 2 laboratory experiments in which we reared cohorts of four different geographic populations of A. albopictus across 5 temperature treatments (18, 21, 25, 18, 31°C) and three resource treatments (low, medium, high larval resources). We determined if the slope of the size-fecundity relationship varied by source population, temperature, or resource; and if variation in this relationship affects lambda (λ') estimates in a competition study between A. albopictus and Culex pipiens (Linnaeus), the northern house mosquito. Temperature treatments significantly affected the size-fecundity relationship, resource level marginally affected the relationship, while source population had no effect. We found positive relationships between size and fecundity when mosquito larvae were reared at high temperatures and low resource levels but the relationship disappeared when mosquitoes were reared at a low temperature or with high levels of resources. The variation in the size-fecundity relationship produced from different temperatures resulted in statistically different lambda (λ') estimates. However, these changes in lambda (λ') did not alter the trends in the population performance across treatments or conclusions of the competition study. This study provides evidence that the population growth model is sensitive to variation in size-fecundity relationships and we recommend biologists apply the most compatible size-fecundity relationship to the models to obtain the most accurate estimates of population performance.