Table 6.
Associations Between Delay of Gratification at Age 54 Months and Later Measures of Academic Achievement With Interactions Between Delay of Gratification and Socioeconomic Status
| Variable | Achievement composite |
Behavior composite |
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Grade 1 |
Age 15 |
Grade 1 |
Age 15 |
|||||||||
| (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) | (9) | (10) | (11) | (12) | |
| Delay minutes (continuous) | 0.279*
(0.038) |
0.115*
(0.035) |
0.050 (0.030) |
0.236*
(0.040) |
0.083*
(0.037) |
0.040 (0.034) |
−0.059 (0.042) |
−0.019 (0.043) |
0.012 (0.033) |
−0.062 (0.044) |
−0.023 (0.046) |
0.009 (0.040) |
| High-SES indicator | 0.509*
(0.064) |
0.050 (0.068) |
0.032 (0.059) |
0.747*
(0.067) |
0.270*
(0.071) |
0.266*
(0.066) |
−0.187*
(0.070) |
0.026 (0.084) |
0.031 (0.064) |
−0.286*
(0.074) |
−0.119 (0.088) |
−0.127 (0.077) |
| Interaction | −0.101 (0.067) |
−0.043 (0.058) |
−0.035 (0.050) |
−0.069 (0.069) |
−0.007 (0.061) |
−0.018 (0.057) |
0.010 (0.073) |
−0.038 (0.071) |
−0.058 (0.054) |
0.094 (0.076) |
0.040 (0.075) |
0.017 (0.066) |
| Delay minutes (categorical) | ||||||||||||
| < 0.333 min | ref | ref | ref | ref | ref | ref | ref | ref | ref | ref | ref | ref |
| 0.333–2 min | 0.298*
(0.127) |
0.182 (0.110) |
0.109 (0.096) |
0.353*
(0.131) |
0.202 (0.115) |
0.151 (0.107) |
0.055 (0.140) |
0.060 (0.137) |
0.050 (0.104) |
−0.140 (0.148) |
−0.082 (0.145) |
−0.097 (0.127) |
| 2–7 min | 0.424*
(0.127) |
0.215 (0.110) |
0.053 (0.097) |
0.457*
(0.132) |
0.288*
(0.115) |
0.199 (0.108) |
−0.088 (0.140) |
−0.046 (0.137) |
0.006 (0.105) |
−0.182 (0.146) |
−0.103 (0.143) |
−0.024 (0.126) |
| 7 min | 0.721*
(0.099) |
0.308*
(0.090) |
0.147 (0.079) |
0.646*
(0.105) |
0.222*
(0.097) |
0.121 (0.091) |
−0.121 (0.109) |
−0.025 (0.112) |
0.034 (0.086) |
−0.193 (0.116) |
−0.087 (0.120) |
−0.028 (0.106) |
| High-SES indicator | 0.585*
(0.174) |
0.154 (0.156) |
0.041 (0.136) |
0.951*
(0.178) |
0.428*
(0.163) |
0.417*
(0.151) |
−0.097 (0.187) |
0.163 (0.190) |
0.191 (0.144) |
−0.375 (0.195) |
−0.185 (0.199) |
−0.138 (0.174) |
| Interactions | ||||||||||||
| High SES × < 0.333 min | 0.029 (0.252) |
−0.164 (0.218) |
0.032 (0.190) |
−0.274 (0.259) |
−0.337 (0.226) |
−0.266 (0.210) |
−0.124 (0.275) |
−0.127 (0.269) |
−0.160 (0.205) |
0.075 (0.284) |
0.119 (0.276) |
0.035 (0.243) |
| High SES × 2–7 min | −0.027 (0.240) |
−0.138 (0.206) |
0.010 (0.179) |
−0.241 (0.246) |
−0.293 (0.213) |
−0.258 (0.198) |
−0.188 (0.260) |
−0.185 (0.252) |
−0.199 (0.192) |
−0.136 (0.272) |
−0.090 (0.261) |
−0.156 (0.229) |
| High SES × 7 min | −0.159 (0.192) |
−0.119 (0.165) |
−0.033 (0.144) |
−0.242 (0.197) |
−0.119 (0.173) |
−0.134 (0.161) |
−0.073 (0.207) |
−0.167 (0.201) |
−0.203 (0.153) |
0.186 (0.217) |
0.115 (0.212) |
0.049 (0.186) |
| p value from interaction-term joint F test | .668 | .870 | .968 | .640 | .342 | .507 | .899 | .859 | .610 | .450 | .753 | .720 |
| Control variables included | ||||||||||||
| Child background and HOME | No | Yes | Yes | No | Yes | Yes | No | Yes | Yes | No | Yes | Yes |
| Concurrent 54 month | No | No | Yes | No | No | Yes | No | No | Yes | No | No | Yes |
Note: n = 918. For the continuous and categorical measures of delay minutes, the table gives standardized coefficients (with standard errors in parentheses). For the categorical measure, < 0.333 min was the reference category. Because continuous variables were standardized, coefficients can be interpreted as effect sizes. Estimates shown in the first column of each set (i.e., Columns 1, 4, 7, and 10) contained only the measure of delay of gratification and a given outcome measure. Estimates shown in the second column of each set (i.e., Columns 2, 5, 8, and 11) added child background characteristics, Home Observation for Measurement of the Environment (HOME) scores, and site dummy variables. Estimates shown in the third column of each set (i.e., Columns 3, 6, 9, and 12) added other behavioral and cognitive measures also measured at age 54 months. Post hoc chi-square tests were used to generate p values in order to assess whether respective sets of variables were different from one another (ps below .001 have been rounded to .001). The joint F test evaluated whether the set of interaction terms jointly contributed to the model. In other words, it tested whether the set of interactions were statistically significantly different from zero.
p < .05.