Supporting Statement B - HS-ECE Landscape Survey

Integration of Head Start and State Early Care and Education Systems

OMB Information Collection Request

0970 - 0356

Supporting Statement

Part B

February 2019

Submitted By:

Office of Planning, Research, and Evaluation

Administration for Children and Families

U.S. Department of Health and Human Services

4^th Floor, Mary E. Switzer Building

330 C Street, SW

Washington, D.C. 20201

Project Officers:

Amanda Coleman, Amy Madigan, and Ivelisse Martinez-Beck

The Administration for Children and Families (ACF) at the U.S. Department of Health and Human Services (HHS) has contracted with Child Trends to complete the Integration of Head Start and State Early Care and Education Systems project. The purpose of this data collection effort is to gather information about perceptions of statewide integration and collaboration between Head Start and state-level systems, barriers to collaboration and integration, and specific questions addressing how respondents’ different offices interact with Head Start or state-level systems. The ACF Office of Planning, Research, and Evaluation will use this information internally to inform the future research agenda.

B1. Respondent Universe and Sampling Methods

Child Trends will conduct a state/territory landscape survey with key state early care and education (ECE) system leaders. Up to five individuals per state/territory will be invited to complete a web-based survey about Head Start’s participation in aspects of the ECE system. Respondents may include: HS Collaboration Directors, Pre-K Director, Subsidy Director, Child Care Licensing Director, QRIS Director, or other state leader with knowledge of HS’s integration into the state/territory ECE system. Respondents will be identified by compiling names and contact information using published ECE agency directories.

Due to their distinct roles with the state ECE system, each respondent will provide unique information regarding state policies and their perceptions of Head Start’s role in the state ECE system. As mentioned in Supporting Statement Part A, these five respondent types represent the five primary systems with which Head Start may interact at the state level. These respondents bring diverse perspectives that will inform how best to measure systems integration in future research studies like FACES and/or potential studies of the implementation of the Preschool Development Birth to Five grants. Our sampling methods propose to invite these five individuals within a subset of 29 states. The subset of states will be selected based on information gathered in the first phase of this study which examined publicly available data on Head Start’s role in state ECE systems.

Sampling Method Step 1

With the goal of describing the variety of ways in which Head Start is collaborating with other state ECE systems, our sampling framework will first identify the extent to which states meet certain indicators of collaboration based on information gathered in the first phase of the study. For each indicator met, a point will be assigned for a possibility of five total points. A higher number of points may suggest that a state has a higher level of collaboration between Head Start and the state ECE system, where as a lower number of points may indicate there is less collaboration. The survey aims to learn more about policies and perceptions of state leaders in states with higher and lower levels of collaboration, so it is important to include states in the sample that vary in the nature of their collaboration with Head Start. Table 1 describes the five sampling indicators.

Table 1: State Survey Sampling Indicators

Sampling Indicators	Points
State provides state supplemental funds for Head Start	Yes = 1 No = 0
Head Start Collaboration Office co-located in agency with three or more ECE programs (e.g. state preK, child care subsidies, child care licensing).	Yes = 1 No = 0
State QRIS provides an alternative rating pathway for Head Start programs	Yes = 1 No = 0
State allows for children to dually enroll in both Head Start and state preK	Yes = 1 No = 0
At least 50% of Head Start programs serve at least one child receiving a child care subsidy	Yes = 1 No = 0

Sampling Methods Step 2

To ensure geographic representation in the survey, we will invite states that represent the various regions of Head Start established within the ACF’s Office of Regional Operations.¹ Within each region, the states with the highest number of points, the median number of points, and lowest number of points will be invited to participate in the survey (up to three states per region). Table 2 describes the regional sampling.

Table 2: Survey Sample States by Region

Region	Total Number of States (excluding territories)	Number of Survey Sample States
Region 1	6	3
Region 2	2	2
Region 3	6	3
Region 4	8	3
Region 5	6	3
Region 6	5	3
Region 7	4	3
Region 8	6	3
Region 9	4	3
Region 10	4	3
total		29

B2. Procedures for Collection of Information

Respondents for the survey will be identified using public directories of state employees such as the Directory of State Early Childhood Contacts². Once respondents have been identified, they will be invited to participate in a survey using the recruitment email found in Appendix B HS-ECE Landscape Survey_Recruitment Email. The survey (see Appendix A HS-ECE Landscape Survey) will be conducted via a web-based survey platform which allows for easy access for the survey respondent, customizable reminders (see Appendix C HS-ECE Landscape Survey_Follow up Survey Email and Appendix D HS-ECE Landscape Survey_Potential Refusal Response), and restricts access to the survey data only to those on the study team.

To allow us to make comparisons across states, we selected the survey methodology to allow us to collect systematic information in from a purposive sample of 29 states. Similarly, survey methods will also allow us to systematically collect information from within a state to compare perceptions of multiple stakeholders within the same state.

B3. Methods to Maximize Response Rates and Deal with Nonresponse

Expected Response Rates

The total sample of respondents is approximately 145 individuals. Our goal would be to achieve a 90 percent total response rate (131 responses), which is similar to surveys of similar types of respondents.. We also anticipate receiving responses from multiple individuals within a state, which allows us to compare perceptions from various perspectives within Head Start and the state ECE system. A limitation to the proposed design is that it depends on the high response rate within a state to draw comparisons regarding perceptions. To ensure a high response rate, Child Trends will host the survey on a secure online survey platform, which enables us to closely monitor response rates and send email reminders.

Dealing with Nonresponse

Although we hope to gather information from a range of state/territory leaders, we recognize that some individuals may not respond to emails and may not follow through on next steps in the process. We estimate a nonresponse rate of 10% to our initial outreach email. This estimation is based on Child Trends’ previous surveys of similar respondents.³⁴ Given the purpose of the study, we do not need the complete universe of respondents to get a picture of the variability across ECE leaders and states. When developing the internal memo of findings from this formative data collection, we will report the participation rate and consider nonresponse bias when summarizing the information gathered. Methods will include analyzing characteristics of individuals who do not respond (e.g., professional role, geographic location), considering how this affects the types of information shared in surveys. We will note perspectives in our report that may be missing from the group or individuals that did not respond. This data collection will allow the project team, and Office of Planning, Research & Evaluation to understand the role of Head Start in other aspects of the state ECE system in an effort to inform future research and evaluation.

Maximizing Response Rates

To maximize response rates, project staff will do broad outreach via email and will send up to five customized reminders during the survey window. We also limited the number of questions included in the survey to make it as quick as possible to complete, which should maximize response rates. Participants will be able to take the survey using an online link at a time most convenient for them.

We will only collect data from one individual within each of the five administrative areas identified: Head Start, pre-K, child care subsidy, child care licensing, and QRIS. We will attempt to reduce burden by hosting the survey online and allowing administrators to participate at a time most convenient to them. This will help to ensure participation does not interfere with other responsibilities.

B4. Tests of Procedures or Methods to be Undertaken

Head Start and early care and education experts were consulted to maximize survey accuracy and effectiveness. We do not plan to test the survey with participants prior to administration.

B5. Individual(s) Consulted on Statistical Aspects and Individuals Collecting and/or Analyzing Data

The project team for this request is led by Kelly Maxwell, project director. Other team members include Mallory Warner-Richter, Maggie Kane and Audrey Franchett. Head Start and Early Care and Education Integration Project team members from Child Trends will assist in analyzing the data after it is collected. The ACF leads for this project are Amanda Coleman, Amy Madigan, and Ivelisse Martinez-Beck.

1 Our sample will include Regions 1-10, and will exclude Regions 11 and 12 as they serve American Indian and Alaska Native and Migrant Head Start grantees which are not covered under the scope of this project. https://www.acf.hhs.gov/oro/regional-offices.

2 http://ceelo.org/wp-content/uploads/2015/02/ceelo_ece_directory.pdf

3The Build Initiative & Child Trends. (2017). A Catalog and Comparison of Quality Initiatives [Data System]. Retrieved from http://qualitycompendium.org/ on November 8, 2018.

4 Resnick, G., Broadstone, M., Rosenberg, H., & Kim, S. (2015). A national snapshot of state-level collaboration for early care and education. Waltham, MA: Education Development Center.

5