Programmer selection has become a critical strategic software engineering issue as organizations become increasingly dependent on IT. Programmer labor shortage, rising expectations for IT, and the large negative impact of a poor choice make selection even more complicated. This study examines the relationship between individual differences (such as personality, domain specific factors, demographics, ability, and programming experience) and programmer performance. We also broaden the concept of programmer performance beyond specification conformance to software complexity.In an experiment, 29 participants attempted to program a problem. Though there was no time limit, only five programmed the solution as specified. Across the two groups, those who satisfied the specification and those who did not, only ability and age differed. There was no difference in personality or domain specific factors. Though the group who completed the specification is too small to test inferences, one point is clear: even among the small group, there was major variation in the complexity of the software produced. The relationship between individual differences and software complexity warrants further study.