To: Josh Brammer, Office of Information and Regulatory Affairs (OIRA); Office of Management and Budget (OMB)
From: Nancy Geyelin Margie and Laura Nerenberg, Office of Planning, Research and Evaluation (OPRE); Administration for Children and Families (ACF)
Date: July 31, 2019
Subject: Memo Justification for NonSubstantive Change Request from July 2019 (Information Collection 0970-0402)
The details about these requested changes are specified below and have been incorporated into Attachments 1, 2, 5, 6, 9, 10, and 12, as well as Supporting Statement A and B as appropriate.
Modifications
preLAS
We propose modifying the specifications so that all children will be administered the preLAS, not just those whose families did not complete the interview ahead of the in-home visit or who completed the interview beforehand and indicated that the child is exposed to Spanish. For children whose families completed the interview before the in-home visit and indicated the child is not exposed to Spanish, the preLAS is still valuable to administer because it serves as a as an opportunity for children to become acclimated to the activities before the other assessments.
Deletions
The deletions we propose below do not affect our ability to measure all of the broad constructs in our original design (language skills, working memory, math skills, and sustained attention and inhibitory control).
Woodcock Johnson Oral Comprehension
We propose cutting the Woodcock Johnson IV Oral Comprehension subtest as this task has been taxing for children. We previously had two language assessments.
Attention Sustained task
We propose removing this assessment because it has been difficult for field interviewers to administer.
Backward Digit Span
We propose cutting the backward Digit Span (but retaining the forward Digit Span) because there has been limited variability in the data we have collected from this task for the first set of MIHOPE families. In the first cohort of families, almost half of the children did not make it past the practice items to the test items.
Additions
We propose moving items P11 and P11a, which ask caregivers directly about parental warmth, from the caregiver interview to the end of the direct assessment with the caregiver.
Modifications
We propose modifying some of the introduction text to better describe the nature of the assessment.
Deletions
We propose deleting the forward Digit Span (but retaining the backward Digit Span) because this assessment is taking longer than expected. The backward Digit Span is a more complex measure of working memory than the forward Digit Span, so we will continue to be able to assess caregivers’ working memory.
We propose modifying some of the text to improve the flow of the activity. We also propose adding some text to improve the video quality and the ability of coders to accurately code the interaction (for example, by specifying that we need to see the caregivers’ and children’s faces throughout the activity).
Additions
Introduction/screener
We have been informed by one local Institutional Review Board that their new rules will not allow the interview to be conducted with anyone who is not the child’s biological parent. As a result, we propose adding question SC15_ScrOut to the interview if the respondent’s family enrolled in MIHOPE from this site and the respondent is not the child’s biological mother. We also propose adding language at the end of the interview for these families to thank them for their time.
Confirming or providing address
We edited the caregiver interview so that respondents are asked to confirm or provide their address (via the ConfAddress and NewAddress items) at the beginning of the interview rather than the end in order to obtain accurate address information from those who may not complete the interview.
Maternal health and well-being
We propose adding questions MH1x and MH1xa, which ask if the respondent has had any subsequent pregnancies since the MIHOPE child was born, and if so, the number of times she has been pregnant. These questions are being added because the previous version of the interview only asked about subsequent births, but the previous home visiting studies have found impacts on subsequent pregnancies around the time children are in kindergarten. Additionally, given that mothers can give birth to multiple children at one time, we also propose adding MH2x as a clarifying question to make sure we are capturing information for all children that women gave birth to. This question was used in previous MIHOPE waves.
Modifications
Introduction/screener
We propose making modifications to the MostRes section of the screener because the previous version did not efficiently ensure that the interviewer was speaking to the biological mother if she lives with the child. To correct this, we propose changing the wording and routing of the screener items and adding the MomLiveWith and MostRes2 items to attempt to connect the interviewer with the biological mother.
We also modified the interview consent language in SC2 for families who enrolled in MIHOPE in Washington. For these families, we have added language required by the Washington Institutional Review Board.
Finally, we propose modifying SC4 and SC7 to improve their administration. In the previous version of the interview, interviewers were recording two responses in the same item: whether the name was correct or incorrect and the correct spelling. We have added SC4_open and SC7_open so interviewers can record the correct spelling in these items. This change only affects the manner in which the interviewer records the information; the respondent experience will not change.
Child development and school performance
We propose adding a few probes to CD1 (which asks about childcare settings prior to kindergarten) to ensure that caregivers provide information on all childcare settings and so interviewers can more easily code the correct response option.
Parenting
We propose making some minor modifications to the wording of item P3, which asks about children’s books. Given the responses that we saw for the first set of MIHOPE families, we are concerned that respondents are taking more time than is necessary to answer this question because they are providing exact counts of books. Therefore, we propose adding “your best estimate is fine” to the item text to prevent respondents from taking the time to provide exact counts. We also moved information from the probe to the main item text so that administration is consistent for all respondents.
We also propose adding probes to items P9a, P9b, and P9c, which are part of the CHAOS scale, because the language in the item text (for example, a reference to the home being “a real zoo”) has been confusing for respondents.
Maternal health and well-being
We propose modifying the alcohol use items (MH5 to MH8) because there was limited variability in the responses provided by the first set of respondents. As a result, we replaced this set of items with two items on alcohol use (MH4x and MH4ax) that have been used in previous MIHOPE waves. We also modified item MH4a to ask more specifically about opioid use.
Child health
We propose modifying the item that asks about the type of health insurance or health care coverage the child has to further reduce the amount of open-ended responses that are provided. The original item used a dichotomy of the respondent or her employer paying for insurance or the government paying for insurance. However, sample members may be financially contributing to government-provided plans, which could contribute to interviewers having difficulty coding responses. We thus propose modifying the item to ask the respondent how they signed up for the insurance – through their employer, through the state or federal government, or through a health care marketplace. We also propose including a separate item (CH3b) that will capture open-ended responses from all sample members who indicate that their child has health insurance or health care coverage. This item will give interviewers a dedicated space to record verbatim responses, which may make them more comfortable coding a response to the new item that asks about type of coverage.
Revisions to item routing
We propose minor changes to the item routing to improve the flow of the interview. The changes include correcting erroneous routing and changing routing determinations that direct who is prompted to respond to a particular question.
Revisions to wording
We propose very minor wording changes to the interview in the following sections to improve flow and administration and reduce repetitive text:
Introduction/screener
Section A: Child development and school performance
Section B: Social support and relationships
Section E: Family economic self-sufficiency
Section F: Maternal health and well-being
Section I: Adverse childhood experiences (ACE)
Section J: Confirming contact information
Deletions
Introduction/screener
We propose deleting items SC18, SC19, SC20, and SC21, which ask about languages spoken by the child and caregiver.
Child development and school performance
We propose cutting questions CD2 to CD6b, which ask the caregiver to provide consent for the teacher survey and the child’s school and teacher information. Caregivers’ consent for the teacher survey has rarely been obtained as a result of these items in the interview because caregivers have not been providing documented consent when prompted to after the interview. Instead, documented consent will be obtained either from the website or during the in-home visit.
Parenting
We propose cutting the general distress items from the Parenting Stress Index (P8). These items were not asked in previous MIHOPE follow-up waves and overlap considerably with other measures on the interview.
We also propose removing items P11, P11a, and P12, which measure parental warmth, from the interview. As noted above, we propose moving P11 and P11a to the in-home visit. P12 has been converted to an observer-rated item, placing no additional burden on respondents.
Additionally, we propose cutting items P7a and P7b, which ask about children’s use of computers or mobile devices. Interviewers noted that these items were confusing for respondents and often had to be repeated a couple of times. Additionally, the items overlapped considerably with the other home literacy and cognitive stimulation items and were not comprehensive enough items to measure total screen time, which has been shown to be influential in child development.
Family economic self-sufficiency
We propose dropping the public assistance items from SS2a to SS2e, which ask about the amount of each benefit received in the past month and the number of months the respondent received each benefit. These items are susceptible to recall error and are not necessary for the cost-benefit analysis.
Confirming contact information
We propose cutting items CI6 to CI6b because we have found that few respondents have specific information about where and when they plan to move.
Additionally, we propose deleting the item MailorEmail, which asks if respondents prefer to receive the gift card in the mail or electronically, because we will only be sending the incentive payment by mail and not electronically. We also propose removing items CI9b and CI10, which ask the respondent to confirm her and the child’s Social Security Number respectively.
Finally, we propose deleting text from Appt1 and removing item App2c, which ask about the times of day that are generally better for the in-home visit as this information can be obtained when scheduling families for the in-home visit. However, we propose retaining some of the text in Appt1 so that respondents have information about the upcoming visit to their homes.
We propose making some minor revisions to the contact materials that will be sent to caregivers. We propose modifying the language to further simplify it, gain respondents’ attention, and encourage them to participate. We also propose adding a few additional materials that are tailored to respondents who may be less likely to participate in data collection.
In particular, we have added two letters that are targeted to families who have firmly requested that we no longer contact them about data collection activities. One letter is for families who have firmly refused in the most recent round in which contact has been established (referred to as “refusal conversion, refusal in past rounds” in Attachment X). This letter will be sent early in the data collection period. A similar letter was developed for families who refuse to participate during the kindergarten follow-up data collection period (referred to as “refusal conversion, current round” in Attachment X). The letters acknowledge their refusal at that point and encourage them to reconsider taking part in the data collection activities. The letters also invite participants to contact the study team to ask any questions or share concerns about their participation.
We have also added a reminder letter that will be sent via priority mail to caregivers who have not yet completed any of the kindergarten data collection activities by approximately the halfway point of the data collection period.
We edited the website text so that it matches the updates to the contact materials.
We propose modifying the wording of some of the materials to improve their flow and readability. We also propose adding additional reminder materials and a letter to send to principals to notify them of the teacher survey, as well as a brief letter to accompany the FAQ document that will be sent to principals.
| File Type | application/vnd.openxmlformats-officedocument.wordprocessingml.document |
| Author | Jose Tineo |
| File Modified | 0000-00-00 |
| File Created | 2021-10-04 |