Implementation Intentions Increase Parent–Teacher Communication Among Latinos

Abstract

This research tested an implementation intentions intervention to increase parent–teacher communication among Latino parents of young children. Parents (n = 57) were randomly assigned to form implementation intentions or simply goal intentions to communicate with their child's teacher. They completed measures of communication and goal intentions immediately prior to the manipulation and after the manipulation for 6 consecutive weeks. Implementation intentions increased parent–teacher communication among parents with higher initial (premanipulation) goal intentions but not among those with lower initial goal intentions. The findings support existing work on the conditions for implementation intentions to work, and address an important aspect of Latino children's educational success.

ACKNOWLEDGMENTS

The study was supported by grants R21MH077680 from the National Institute of Mental Health. This study would not have been possible without the collaboration of parents and children who participated in various aspects of the research, Angela Moreland Begle who helped with many of the stages in developing CANNE, and Jean E. Dumas who initiated CANNE and developed its English counterpart (PACES). Their help and encouragement are gratefully acknowledged.

Notes

¹This sample included only eight Head Start parents, who did not differ significantly from other parents in their premanipulation communication (M _non-HS  = 1.73, SD = 1.78, M _HS  = 2.00, SD = 1.77), t(55) = .39, ns, or postmanipulation communication (M _non-HS  = 2.26, SD = 2.08, M _HS  = 2.93, SD = 2.12), t(55) = .84, ns. Although these means are consistent with the idea that Head Start parents are more involved at the outset, and possibly benefit less from the intervention, the small number of Head Start parents in the current sample did not allow a sound empirical test of this idea.

Note: At Time 0, participants completed the measures of parent–teacher communication, intention to communicate, and parent acculturation level. They were randomly assigned to conditions and received either the experimental manipulation or control at Time 0. One to 2 weeks later, they started postmanipulation measures that composed the dependent variable.

²Random assignment to conditions occurred before parents were assigned to specific group sessions for the parenting program. Participants’ group affiliation was not a factor in random assignment, but it was included as a covariate in analyses. Groups did not differ in the distribution of control versus experimental participants, χ²(5, 57) = 2.67, ns.

³Week 7 of the parenting program explored ways in which parents can help their children do well in school, including developing a positive relationship with teachers. During that week, all parents were asked to develop a plan to communicate with their child's teacher, in essence comprising an implementation intentions intervention for all parents. When treatment and control conditions were compared for Weeks 7 and 8, they did not differ (M _intervention  = 3.16, M _control  = 2.93), F(1, 29) = 0.61, p = .441. Descriptively, both conditions increased communication with teachers relative to Weeks 1 through 6 (M _control  = 2.08, M _treatment  = 2.66), but especially control participants who until then had not benefitted from implementation intention strategies.

Note: Child gender was coded 1 for male and 2 for female; percentage reported is percentage male. Head Start (HS) status is percentage enrolled in HS. INT = intent to communicate; COM = communication; ACC = acculturation; AGE = child age; Gender = child gender.

†p < .10. *p < .05. **p < .01.

⁴There was a marginal main effect of group membership in the full model predicting parent-teacher communication, F(5, 47) = 2.36, p = .054, such that some groups varied in total communication reported. Group membership, however, did not moderate the effects of any variables (i.e., the effectiveness of the manipulation did not differ between groups).

⁵We repeated all analyses including as a covariate the number of times/sessions a participant completed. The results did not vary with this covariate, which was dropped from further analysis.