This study's overall goal was to separately examine the relationship between the personal capability beliefs of three cohorts of elementary-level science teachers in Denmark and three in the United States and their teaching environments. Of particular interest in Denmark were the second and third cohorts who were especially trained to teach science and technology at the elementary school. In both countries, we tested the belief that teaching environments are associated with self-efficacy and consequently perhaps with retention in teaching. We used instruments to measure self-efficacy and environmental circumstances as well as a follow-up questionnaire after one or two years of teaching to assess longer-range implications. From data collected three times in Denmark for each cohort during their first year as teachers, we found that by the end of that year, all cohorts had significant correlations between their self-efficacy and the chance of supportive features in their teaching environments (r=.326; p=.002; n=86). Importantly, we found that the higher the context assessments, the greater the positive changes in self-efficacy from the pre-service measures until the end of the first year. Of particular interest, in light of the onset of training for teaching science at the elementary level in the years between the first and the following two cohorts, the self-efficacies of the groups which received the science content and pedagogy training were significantly higher than those of the first group. Early qualitative results from the US data show similar patterns and relationships. This study's suggestion that pre-service training can increase the self-efficacy of science teachers as they enter the profession encourages those who have found that higher self-efficacies are associated with greater success. The follow-up study showed that changes in self-efficacy during the first year of teaching may be relevant to whether teachers continue to teach science. If with further study, these relationships between teaching environment and new teacher self-efficacy are sustained, then not only may new elementary teachers be educated for higher self-efficacies and their teaching environments be enriched, but they may also be followed with these measures to assess their progress during their first years, providing chances for support.