Table 5. Tests of Indirect (Mediation) Effects of Age Differences in Affective and Cardiovascular Responses.
| Dependent Variables | ||||||||
|---|---|---|---|---|---|---|---|---|
| Task Negative Affect | Task Pulse Rate | Posttask Negative Affect | Posttask Positive Affect | |||||
| Potential Mediators | A | B | A | B | A | B | A | B |
| Goals | ||||||||
| Change Partner | −0.59(0.31)† | 0.14(0.04)*** | ----- | ----- | −0.32(0.32) | 0.03(0.01) * |
----- | ----- |
| Task Performance | ----- | ----- | ----- | ----- | 0.81(0.21)*** | − 0.01(0.02) |
0.60(0.22) ** |
0.14(0.08)
† |
| Appraisals | ||||||||
| Task Enjoyment | 0.62(0.19) ** | −0.18(0.07) * | 0.62(0.21)** | −0.05(0.46) | 0.42(0.21)* | − 0.01(0.03) |
0.43(0.20)* | 0.15(0.10) |
| Task Performance | 0.55(0.13)*** | 0.02(0.10) | ----- | ----- | 0.52(0.14)*** | 0.001(0.0 4) |
0.45(0.13)** | 0.23(0.14) |
| Confederate Likeability | 0.74(0.16)*** | −0.13(0.08) | 0.88(0.17) *** |
−
1.37(0.58) * |
0.62(0.16)*** | − 0.03(0.03) |
0.66(0.17)*** | 0.12(0.11) |
| Emotion Regulatory Behaviors |
||||||||
| Move on to new Scenario |
----- | ----- | −0.70(0.30)* | −0.01(0.32) | ----- | ----- | ----- | ----- |
| Reduce Negativity | ----- | ----- | ----- | ----- | −1.68(0.40)*** | 0.01(0.01) | ----- | ----- |
| Direct Path (Age Group to Dependent Variable) |
||||||||
| C | −0.55(0.15)*** | −7.26(1.10)*** | −0.19(0.05)*** | 0.73(0.21)*** | ||||
| C’ | −0.27(0.16)† | −6.03(1.19)*** | −0.13(0.06)* | 0.40(0.22)† | ||||
Note. Each set of columns (A and B) tests whether there is an indirect effect between age group (i.e., predictor variable) and each dependent variable (DV) via the potential mediators. For each DV, Column A represents the coefficient between age group and the mediator, and Column B represents the coefficient of the mediator predicting the DV in the full model with age group included. Standard errors are denoted in parentheses. Dashed lines denote that the potential mediator was not included in the model. Coefficients in boldface denote significant indirect pathways (i.e., mediation) between age group and the dependent variable via the mediator. The rows for the Direct Path represent the direct association between age group and the dependent variable before (C) and after (C’) the mediators have been entered for each model. All models controlled for gender, ethnicity, and health conditions. In models testing mediation of reactivity, the baseline assessment of the key measure (e.g., negative affect, pulse rate) was included as a covariate. In models testing mediation of recovery, both the baseline and task affect were included as covariates. In the test of potential mediators of age differences in pulse rate reactivity, cardiovascular medications were also included as a covariate.
p < .10;
p < .05;
p < .01;
p < .001.