Many capture–recapture studies on adult dragonflies have found male-biased sex ratios. However, few have estimated survivorship of males and females separately from data on frequency of recaptures in the field. Even when daily survival and capture probabilities are estimated separately, controversies can arise on whether sex biases in local survival are to be attributed to mortality or permanent emigration from the study site. The knowledge of male and female movements, assessed on an appropriate scale (i.e., within and outside the breeding site), can help address this issue. In this paper, we performed a 4-yr capture–recapture study of two Libellula fulva Müller populations in northwest Italy. Cormack-Jolly-Seber models were used to get unbiased estimates of demographic parameters (daily survival and capture probabilities, sex ratio, mean life span, and population size). Movement parameters were measured directly by georeferencing encounters. Moderate differences in survival, with males surviving better than females, were found in one population and not in the other, suggesting that these differences are not an inherent characteristic of the species. In the population with lower female survival, females were not more vagile than males, thus indicating their lower survival was caused by actual mortality rather than to emigration. In the population with no survival differences between males and females, marked males outnumbered females, but estimated sex ratios were ≈1:1 or female biased. Therefore, raw field data were misleading because they led to underestimates of the more elusive sex and overestimates of the more detectable one (males). Survival and movement differences detected in the two populations are discussed in the framework of local environmental and demographic factors.