Based on Schaie's (1965) general developmental model, various data-driven and theory-based approaches to the exploration and disentangling of age, cohort, and time effects on human behavior have emerged. This paper presents and discusses an advancement of data-driven interpretations that stresses parsimony when interpreting the results of sequential models. Second, a synthesis of data-driven and theory-based approaches examines the specific predictors of patterns of cross-sectional, longitudinal, and time-lag differences. This approach is exemplified with data from two cross-sectional samples. In 1991 and 1996, representative samples of 13- to 29-year-old Germans were interviewed orally. Parts of these samples were analyzed employing a time-sequential and a cross-sequential strategy (analyzed N=6105). While the data-driven approach allowed for two alternative interpretations, the second approach revealed that parental emotional help for their children declined with age, partly due to the children leaving home. Help provided for parents generally increased with age, however, leaving home had the opposite effect so that overall, only small and inconsistent age increases in help for parents were found.