5 The direction of effects is family-specific 161 We then explored whether the proportions of families showing reciprocal, parent-driven, adolescent-driven, and null effects were different from what would be expected by chance (i.e., 25% of families showing each type of effect). Most of the proportions (22 of 32; see Table 2) were significantly different (standardized residuals ≥ |1.96|) from expectations. In other words, some directions of effects were more prominent in the sample than other directions of effects. Two findings merit attention. First, more families than expected showed reciprocal effects between adolescent positive affect and parental warmth and autonomy support. Second, daily associations with parental control were often parentdriven: more families than expected showed parent-driven effects between psychological control and positive affect, and between behavioral control and negative affect. Moreover, heterogeneity in directionality was also found within families, such that the direction of effects depended on the specific parenting-affect association. As shown in Table 3, no single family consistently demonstrated only reciprocal effects, parent-driven effects, adolescent-driven effects, or no effects. When examining homogeneity across the models for either adolescent positive or negative affect, a few families (0% to 11.1%) demonstrated a similar direction of effects across associations. Thus, the direction of effects often did not generalize across the eight different parenting-affect associations for a given family. Table 3 Direction of Lagged Parenting-Affect Effects Summarized per Family Direction of effects Associations with PA & NA Associations with PA Associations with NA N (%) N (%) N (%) 1. Completely reciprocal 0 (0.0%) 6 (3.9%) 1 (0.7%) 2. Completely parent-driven 0 (0.0%) 2 (1.3%) 0 (0%) 3. Completely adolescent-driven 0 (0.0%) 1 (0.7%) 1 (0.7%) 4. No lagged effects at all 0 (0.0%) 3 (2.0%) 17 (11.1%) 5. Mix of lagged and no effects 153 (100.0%) 141 (89.8%) 135 (88.2%) Total 153 (100.0%) 153 (100.0%) 153 (100.0%) Note. PA = positive affect. NA = negative affect. Sensitivity Analyses We conducted several sensitivity analyses to investigate the robustness of the findings regarding heterogeneity in the direction of effects (H1). First, we conducted pre-registered
RkJQdWJsaXNoZXIy MTk4NDMw