Intervention(s)
This study asks whether the effectiveness of text-messaged-based parenting support depends on when the support is provided. Specifically, it compares the effects of an early childhood text-messaging program sent during the weekend to those of the same program sent on weekdays. This texting intervention breaks down the complexities of parenting by providing a combination of general information about important literacy skills and parent-child activities (i.e., “FACT” text messages), actionable advice with specific examples of parent-child literacy activities (i.e., “TIP” text messages), and encouragement/reinforcement (i.e., “GROWTH” text messages). We randomly assign parents of pre-kindergarten children into four groups. The first group of parents receives the original program, that is, a “FACT” message on Monday, a “TIP” message on Wednesday, and a “GROWTH” message on Friday, henceforth the Weekday program. The second group received the “FACT”, “TIP”, and “GROWTH” messages on Friday, Saturday, and Sunday, respectively, henceforth Weekend program. The Weekend program differs from the original program (i.e., Weekday program) in that it sends the text messages on different days but also in that these days are consecutive. Therefore, a third group of parents receives the “FACT”, “TIP”, and “GROWTH” message on a Tuesday, Wednesday, and Thursday, respectively, to parse out the spacing effect from the weekend effect. This program henceforth will be called MidWeek program. The fourth group just received a "FACT" message on Mondays. This groups was intended as pseudo control group with a lighter touch treatment since the district did not allow a pure control group. Parents received the text messages over eight months from Mid October 2016 to early June 2017. Eventually, we did not use the single "FACT" group in our analysis because our main goal was to understand how the timing matters for these programs. Therefore, it seemed to be more appropriate to use the Weekday program as comparison group which had shown positive effects in several prior experiments than to use the single "FACT" group for which we had no empirical or theoretically prior on how it would affect students.