#5 - Child Observation Study - Wave 2

Direct Child Assessments

Wave 2 Data Collection

Supporting Healthy Marriage Demonstration Evaluation

Direct Child Assessments

Wave 2 Data Collection

Self-regulation Tasks

1. Walk-a-line task: In this task, children are asked to first walk down a line that is tapped to the floor (baseline walk). They are then asked to walk the line again, but to walk it as slowly as they can (slow walk 1). Upon completion of that work, they are asked to walk the line one last time, but to walk it as slowly as they can. The measure of interest here is the difference in the length of time that it takes to walk the line between the baseline and the slow walks (usually averaged across slow walk 1 and slow walk 2). The more that a child can slow down, the more behaviorally regulated he or she is thought to be. In SHM, the expected sample size for this task is 2,034 across program and control groups pooled across sites. The targeted age range for this task is children who are 2 to 3 years, 5 months old at the 30-month follow-up. This task is being used in BSF as well.

2. Item selection/attention shifting task: In this task, we are measuring children’s executive functioning, which is their ability to filter though extraneous stimuli or information and focus on the task at hand. The wisdom here is that children who are better able to filter through the extraneous stimuli and stay on task are more self-regulated. In the task, children are shown two pictures which are generally the same (e.g., two cats), but differ on one key dimension (e.g., one cat is red, the other is blue), and then are asked to identify which of these two pictures a third picture (e.g., blue flower) is most like. The children who identify the blue cat would get this item correct, because they have appropriately sorted the pictures based on the dimension of color. The children are shown a series of items that increase in difficulty. For this task, we are also able to modify the content of the items to show to children based on the child’s age at the time; more difficult items can be used for older children, and easier items can be used with younger children. This allows us to broaden the age range of children who can be assessed with the task. Thus, the targeted age range for this task is children who are 3 years, 6 months old to 8 years, 5 months old at the 30-month follow-up. The expected sample size for the task is 2,080 children.

3. Head-to-(Knees-to-Shoulders-to)Toes Task: This task is a little like Simon Says. The rules of the game are that children are supposed to touch their head when the interviewer says “touch your toes,” and touch their toes when the interviewer says “touch your head.” Children who are better able to keep the rules of the game in their head and suppress their natural instinct to follow the interviewers’ instructions verbatim are thought to have higher levels of executive functioning, and in turn higher levels of self-regulation. The task can be adapted for children in the sample by adding knees into the mix. The targeted age range for this task is children who are 3 years, 6 months old to 8 years, 5 months old at the 30-month follow-up. The expected sample size for the task is 2,080 children.

Interviewer assessment of children’s social, emotional and behavioral regulation:

We also propose having interviewers provide their assessments of children’s behaviors during the tasks using an adapted form of the Leiter-R scale. In these items, interviewers are asked questions about whether the child stays on task, is disruptive, compliant, etc. The measure consists of 13 items that are completed immediately after the interviewer leaves the family’s home. This measure has been used in other work conducted by Cybele Raver and Karen Biermann in her REDI-HS evaluation.

Language development

Lastly, we propose obtaining direct child assessments of children’s language abilities at the 30-month follow-up. There are a number of reasons to expect that children’s language abilities will be affected by the SHM evaluation. To the extent that the intervention keeps parents together, children in the program group may be more likely to be in households where there are greater amounts of ambient language than experienced by their counterparts in the control group. As a result, children’s language skills may develop faster in such households. Second, warmth, responsive parenting is associated with better language abilities in young children. To the extent that the SHM intervention affects parenting styles though changes in the quality of the relationship between parents, we may also expect the intervention to indirectly affect children’s language abilities.

PPVT: We propose using the PPVT, which is a measure of children’s receptive language abilities to assess children’s language skills. In the PPVT, children are told a word, and then asked to identify the word from a series of pictures. We propose using this measure with children who are between the ages of 2 and 4 years old at the 30-month follow-up. This measure is also being used in BSF.