Part B ECLSK-2023 Preschool Field Test

The Early Childhood Longitudinal Study, Kindergarten Class of 2022-23 (ECLS‑K:2023) includes a preschool sample to collect information on children’s experiences prior to kindergarten. The target population for the preschool field test is children expected to attend kindergarten in fall 2020. The goal of the field test is to test the operational procedures of the preschool round and provide answers to important design issues and assumptions prior to implementing the main study preschool data collection in 2022. The sample for the preschool field test will use an address-based sample (ABS) design that identifies households with preschool children in sampled primary sampling units (PSUs). Sampling households allows virtually complete coverage of children expected to enroll in kindergarten in the following school year that could not be obtained by sampling from any other sampling frame. Section B.1.1 discusses the sample design for the preschool field test in greater detail. Results of the field test will shed light on how to refine the eligibility criteria, the achievable response rates for household screener and preschool parent survey, and how likely a child is to attend the sampled kindergarten school associated with the area from which the child has been sampled. Section B.1.2 discusses the precision requirement and sample sizes for the preschool field test.

B.1.1 Sample Plan for Preschool Field Test

For the preschool field test, a sample of kindergarten schools will be selected from a set of purposefully chosen PSUs. Household addresses will then be geocoded to the catchment areas of those schools so that each address will be linked to one and only one kindergarten school. Addresses will be sampled from these catchment areas and households will be sent an invitation to complete the household screener to screen for households with children who will attend kindergarten in the 2020-21 school year. One eligible child will be sampled from each household with eligible children, and the adult respondent will be invited to complete the parent survey conducted through the web. An option for paper mode will be provided to those who do not respond through the web and respondents who request a paper survey. The parent survey will ask the respondent to identify the school the child is expected to attend in the fall of 2020. The screener and parent survey instruments will be provided in English and Spanish. Finally, households responding to the spring 2020 parent survey will be contacted again for another survey in fall 2020 to ascertain whether the sampled child attends kindergarten, and which kindergarten school the child attends.

B.1.1.1 Multistage Sample in Purposefully Selected PSUs

The plan for the preschool field test design involves multiple stages of sampling. First, 30 PSUs with varying urbanity levels will be purposefully chosen. Second, a sample of 40 public schools will be selected from the 30 PSUs, and then used to define catchment areas. Third, addresses will be geocoded into the catchment areas of the sampled schools to form clusters for address sampling. A total of 28,000 addresses will be sampled from the 40 catchment areas. Finally, in the sampled responding households with eligible children, one child will be sampled per household to be the target of the parent survey.

B.1.1.2 Defining Catchment Areas

Each address in a sampled PSU must be linked to one and only one public school with a kindergarten program.¹ The preschool field test will aid in developing operational methods for “mapping” addresses to the sampled kindergarten schools for the national preschool data collection. The approach includes the collection of data regardless of whether a sampled preschool child is expected to attend a sampled school or not. Information obtained from the mapping done in the preschool field test will be useful in planning for the main study, as the costs for keeping the children in the national sample for kindergarten through fifth grade may be affected by the accuracy of the mapping of the schools sampled for the national collection.

B.1.1.3 Eligibility and Sampling of Children

A child is considered to be an eligible target of a parent survey for the preschool field test if he/she satisfies either of the following criteria:

The expected grade or equivalent reported on the screener for the child for the fall of 2020 is kindergarten of any type regardless of the child’s age; or, if no such children reside in the household,
The child is 3 or 4 years old and is reported on the screener as expected to be in first grade in the fall of 2020.²

Note that based on the criteria above, eligibility for the preschool field test will not be restricted by whether the child is already in kindergarten of any type in spring 2020. For example, even if a child is already in a 2-year kindergarten program in spring, the child is still considered an eligible target of the parent survey as long as he/she is expected to remain in kindergarten in the subsequent fall. A clear distinction between kindergarten and non-kindergarten may be made during sampling for the national study preschool round. The field test findings will help determine whether a more restrictive eligibility definition for the preschool round should be used in the main study.

Screener respondents will be asked to enumerate children age 10 and under living in the household. For the paper screener, the respondent is asked to enumerate up to five children (to keep the length of the paper form brief), starting with the youngest. The American Community Survey (ACS) 5-year 2013-2017 data show that among all the households with children, 99.8 percent have five or less children. For the web screener, there is no limit on the number of children who can be enumerated.

The within-household sampling algorithm will then sample one eligible child for data collection. The algorithm will assign each child in the household who is reported to be expected to attend kindergarten in fall 2020 an equal probability of selection and randomly choose one of these children for the study, irrespective of the children’s ages. If the household has no children who are expected to be attending kindergarten in the fall, but does have one or more 3- or 4-year-old children expected to be in first grade, then the algorithm will randomly sample one these children per household for inclusion in the field test.

In the web data collection, households with children age 10 and under and no children eligible for the preschool parent survey will be asked to provide information about the mapping of their home addresses to their children’s schools. Households will be qualified for these additional questions only if: (1) the highest grade level of the child(ren) is grade 3; and (2) neither of the eligibility criteria for the preschool parent survey (as described above) is satisfied. Among the households that meet these qualifications, the child with the lowest grade level will be sampled for additional questions about where this child in kindergarten, first, second, and/or third grade attended or is attending kindergarten. Other items, such as whether the child has lived at the same address since kindergarten, will also be included so that this information can be used in evaluating the linking of addresses to kindergarten schools for the national study.

B.1.1.4 Fall Follow-Up of Parent Survey

A fall follow-up parent survey is planned for the preschool field test. In the fall of 2020, the preschool parent survey respondents will be contacted for a brief follow-up survey to find out: (1) whether the sampled child is currently in kindergarten; and (2) if so, which school the child is attending if the child is in a public school.³ If the child is not currently in kindergarten, then the parent will be asked: (1) whether the child is currently enrolled in school; and (2) if so, the grade or grade equivalent of the child. Note that the goal of this survey is to evaluate the accuracy of the information provided by the parents in the previous spring. This follow-up survey will not be employed in the national study, during which children from the preschool round will be followed and, if in kindergarten in fall 2022, added to the national sample of kindergartners, taking part in that round’s data collection activities.

B.1.2 Precision Requirements and Sample Sizes

The purpose of the preschool field test is not to generate national estimates, but to address some key study design issues and assumptions, test data collection procedures, and evaluate the parent survey items. Specifically, the preschool field test data will be used to evaluate the following assumptions:

The response rates for the screener and the parent survey. Comparison will be made between the expected rates to the achieved rates in the preschool field test so that these assumptions can be updated for the main study design. In addition, the percentage of respondents by web and by paper mode will be obtained from the preschool field test, as these estimates may have cost implications for the main study. Response rates will also be examined, for example by available frame information and the choice of an English or Spanish instrument.
The ineligibility rate in fall 2020 (due to the sampled child not attending kindergarten) among the children with a completed parent survey in spring 2020. Besides the overall rate, the ineligibility rate will also be calculated separately for subgroups, to allow refinement of the eligibility criteria for the main study.
The proportion of preschool children who will attend the kindergarten schools they are linked to based on the mapping between household locations and school boundaries. One study design assumption is that children in the catchment area for a particular school will attend that school at a high rate. This assumption will be tested with the field test data.
The proportion of children who will, in fall 2020, attend the kindergarten school that the parents expect them to attend in spring 2020. If most parents can accurately predict the schools their children will attend, then this information collected in the spring national preschool data collection can help locate the children for kindergarten sampling activities for the national kindergarten rounds.

After the field test concludes, data collection procedures will be evaluated and changes will be proposed for implementation in the national study. In addition, items in the preschool parent survey will be examined in terms of item nonresponse, timings, and scale psychometrics. For example, item response rates will be evaluated for different modes of data collection and compared to the ECLS-K:2011, and items with a high proportion of “refused” and “don’t know” answers or which resulted in survey breakoffs will be considered for revision for the national collections.

For the preschool field test, the main goal is to have sufficient sample size to estimate the key study sample design parameters. The proposed field test design is to sample a total number of 28,000 addresses from 40 catchment areas associated with public schools with kindergarten programs. This will average 700 sampled addresses in each catchment area. Table B-1 shows the sample sizes and the 95 percent confidence intervals for response rates with 80 percent power for the key steps of data collection. With 3 percent eligibility rate and the expected response rates (which are largely consistent with the National Household Education Study (NHES) 2016 data and NHES:2017 web test experiences), the average number of complete parent surveys per catchment area will be approximately nine. This is estimated to result in approximately 364 completed spring parent surveys and 309 completed fall follow-up surveys across 40 catchment areas.⁴ This is expected to yield a reasonable number of children associated with each sampled kindergarten school for estimating the percentage of preschool children who will attend the kindergarten schools to which they are mapped. This percentage can be calculated overall and by level of urbanicity, so its distribution can be evaluated.

Table B-1. Estimated Sample Sizes and Response Rate Precision Levels for Key Steps of Data Collection*

	Estimated number of fielded cases (denominator for calculating response rate)	Expected response rate	95% confidence interval for
		Expected response rate	response rate with 80% power	number of completes with 80% power
Households invited to complete web screener	26,040⁴	33%	(32.4%, 33.6%)	(8,437, 8,749)
Households that complete web screener and are eligible for/invited to complete parent survey	258⁵	80%	(74.9%, 85.1%)	(193, 220)
Households receiving paper screener	17,447⁶	40%	(39.3%, 40.7%)	(6,857, 7,101)
Households that complete paper screener and are eligible for/invited to complete paper parent survey	209	75%	(68.9%, 81.1%)	(144, 169)
Households that complete screener and parent survey (web or paper) and are invited to complete Fall follow-up survey	364⁷	85%	(81.2%, 88.9%)	(296, 324)

* One of the main goals of the ECLS K:2023 Preschool Field Test is to determine the correct estimated response rates for the national study.

B.2 Procedures for Collection of Information

Overview of the Preschool Field Test Design

The expected field period of the preschool field test is early to late 2020, with the screener and spring parent survey instruments fielded from January through June 2020 and a fall follow-up parent instrument field period of September through October 2020. A sample of kindergarten schools will be selected from a set of purposefully chosen Primary Sampling Units (PSU). Next, each household address will be geocoded to the catchment areas of those schools so that each address will be linked to one and only one elementary school with a kindergarten program. Approximately 28,000 addresses will be sampled from these catchment areas to screen for households with children expected to attend kindergarten in the 2020-21 school year. All sampled addresses will be mailed materials (as shown in Attachment A) with instructions to visit the MyECLS website to complete a web screener survey to determine their eligibility for participation in the study.

If the household is eligible for participation in the preschool parent survey, one child will be sampled from the household and an adult respondent with knowledge of the child’s care and education will be asked to complete the parent survey conducted through the web with a paper mode option. Among other questions, the preschool parent survey will ask the parent to identify the school the sampled child is expected to attend in the fall 2020. Parents who respond in spring 2020 will then be re-contacted in fall 2020 to ascertain whether the sampled child attends kindergarten, and which kindergarten school the child attends. Each preschool field test address will be linked to one kindergarten school so that the responses from the preschool parent follow-up as to where the child actually enrolled for kindergarten in the fall can be evaluated.

All households will receive a five-dollar bill in the initial letter as an incentive to complete the household screener. Eligible respondents who go on to complete the parent survey will receive an additional ten dollars. Those who complete the survey online through the MyECLS website will be given the choice of a ten-dollar digital Amazon code or to receive a ten-dollar bill in the mail. Those respondents who are sent the paper survey will receive a ten-dollar bill prepaid incentive with the paper survey.

Nonresponders to the screener and/or survey will receive email, text, and/or paper reminders to participate. Respondents who break-off during completion of the screener and/or web survey will also receive follow-up reminders. Table B-2 shows the schedule of mailings and reminders, and all of the respondent communication materials are provided in Attachment A.

Preschool Field Test Initial Household Mailing

At the start of the preschool field test data collection, each sampled household will be mailed an ECLS initial letter (Attachment A1) to: (1) briefly introduce the study, and (2) direct respondents to MyECLS website to complete the household screener and, if relevant, parent survey.

The initial mailing will contain a letter printed on letterhead with the NCES logo and Department of Education logo at the top, with the ECLS sun logo as a watermark. The letter will provide the MyECLS website address, along with a unique, 8-digit study PIN number that a member of the household should use to gain access to the screener survey on the MyECLS website. It will emphasize that an adult in the household who knows about the household’s children’s care and education should respond to the screener and, if eligible, the parent survey. However, since all households are asked to respond regardless of whether there are children in the household, the letter will encourage everyone to complete the screener. A $5 bill will be included in the initial mailing (prepaid) as an incentive to all households to respond to the screener. Households will be told in the letter that additional study information will be available online at MyECLS. Finally, the letter will contain the study helpdesk’s phone number in case assistance is desired.

Once a respondent from the household logs onto the MyECLS website, he or she will be instructed to enter the PIN number provided in the letter. Assuming a valid PIN is entered, the respondent will be asked the first question: “Are there any children age 10 or younger living in this household?” The PIN will link the response to this question to the address of the household. Respondents who respond that there are no children in the household age 10 and under will then see a message thanking them for their response but indicating that they are not eligible to continue the survey.

Respondents who have children age 10 and under in the household will be welcomed into the additional screener questions and given the option to provide an email address and/or mobile number that can be used to save their data and allow access back into the survey at a later time in the event that the respondent needs to stop and continue later, or the survey times out. The respondent will have the option to choose to proceed through the screener without entering any contact information, although in this case he or she will be informed that, to provide security of the responses, he or she will not have access to the entered data later in cases where the survey is exited without finishing the instrument. After either providing the contact information or declining to do so, the respondent will be taken to the household screener items and then, if eligible, on to the full spring parent survey. Respondents who have entered contact information who exit the screener or parent survey before finishing them will be prompted to provide that contact information at reentry into the survey. A temporary, 6-digit code will be emailed or texted to the respondent (depending on what type of contact information was provided; Attachments A28 and A30), and respondents who enter that temporary code on the MyECLS website will then be able to continue with the screener or survey where they left off. If respondents do not choose to provide an email address or mobile phone number prior to entering the screener, they will be asked to do so at the end of the screener (if eligible to continue to the parent survey) and/or when they click on the logout button in the screener or survey. All of these screens are provided in Attachment C.

Preschool Field Test Spring 2020 Household Screener

Once a respondent has provided an email address and/or mobile number or opted out of providing contact information for the MyECLS website, he or she will continue through the screener, enumerating all the children age 10 and under who live in the household. Names, nicknames, or initials of children will be collected. Respondents will then be asked about one child at a time to collect age, sex, current grade, and grade the child is expected to attend in the fall.

If there is more than one eligible child in the household, the screener will select one eligible child at random, first among those expected to be enrolled in kindergarten (any type) or a kindergarten equivalent in fall 2020, and then (if there are no eligible children in that category) among 3- or 4-years-olds reported to be expected to be in the first grade (or a higher grade) in the fall. Once the child target of the preschool parent survey is selected, the respondent will be immediately asked to continue to complete the parent survey. The transition from the screener to the parent survey will be as seamless as possible for respondents so that it feels like a continuation of the screener and not a separate task.

In order to further evaluate the efficiency of using household sampling to identify future kindergarten attendance, a sample of web screener respondents will be selected for a few additional screener questions. These additional questions will not be asked in the paper version of the screener. Respondents will qualify for these additional questions only if: (a) they do not have a child in the household who is planning to enroll in kindergarten in the fall of 2020, and (b) they have a child in the household who is currently age 10 or younger and currently in kindergarten, first, second, or third grade. Respondents who meet these qualifications will be sampled (to reduce respondent burden), and if selected, will receive a few additional screener questions asking if the child has always lived at the current address, and if not, when did they start living at the current address. If there is more than one child who could be selected to be the subject of the questions, priority will be given to the child in the lowest grade level. After these additional questions are completed, these respondents will be thanked and told they are not eligible for further participation in the study. Note that these additional screener questions are provided in the screener in Attachments B1 and B2.

Follow-Up Procedures for the Household Screener

The proposed model of follow-up contacts is based on the model successfully used in NHES 2011, 2016, and NHES web test in 2017. Multiple follow-ups with nonresponding households are planned to encourage completion of the screener and parent survey. See Table B-2 for a detailed schedule of the follow-up efforts, and Attachment A for the follow-up contact materials.

Thank You/Reminder Postcard (Attachment A2)

All households will be sent a thank you/reminder postcard mailed one week after the initial mailing and containing the same PIN the household received in the first mailing. The postcard will be a folded postcard with tear off tabs, which provides respondents with their unique and confidential web survey log-in credentials.

Email and Text Reminders (Attachments A13, A13a, A14, and A14a)

If a respondent breaks off from completing the web screener and has provided an email address and/or mobile phone number, we will send them weekly email and/or text reminders to return to the website to finish the screener questions.

Reminder Letter (Attachment A3)

After two weeks, all households that have not responded to the screener will receive a first class mail reminder letter. The reminder letter will reference the initial mailing and five-dollar bill previously sent.

Reminder and 1^st Paper Screener via First Class Mailing (Attachment A4)

Approximately 4 weeks after the initial mailing, a paper screener will be delivered via first class mail to households that have not begun responding to the screener. Respondents who began but did not complete the web screener, but who have provided an email address and/or phone number, will receive weekly email and/or text reminders to return to the website to finish the screener questions. If they do not complete the screener after two weekly email and/or text reminders, they too will receive a paper screener via first class mail. In all of these cases, a cover letter will be included in the mailing, directing the respondent to complete and return the enclosed paper screener with a postage-paid envelope. Respondents will be told that if they prefer, they may complete the web version of the screener.

Reminder and 2^nd Paper Screener via FedEx Mailing (Attachment A5)

All households that remain nonresponding will receive a final paper screener via FedEx approximately 6 weeks after the start of data collection. In addition, any respondents who have broken off from the web screener and have not responded to either the email and/or text reminders or the paper screener sent via first class mail will be included in the group to receive the final paper screener sent via FedEx.

Table B-2. ECLS-K:2023 preschool field test mailings: Screener and parent survey

Activity	Mode	Instrument	Timeline
Initial first class mailing & $5 prepaid	web	screener/parent survey	week 0
Thank-you/Reminder postcard	web	screener/parent survey	week 1
Weekly email and/or text reminders	web	screener/parent survey	weekly after breakoff
Mail first class reminder letter	web	screener/parent survey	week 2
Mail first class letter, paper screener, postage-paid return envelope	paper	screener	week 4
FedEx letter, paper screener, postage-paid return envelope	paper	screener	week 6

Preschool Field Test Spring 2020 Parent Survey

The main data collection instrument of the preschool field test is the parent survey, provided in Attachments B1 and B2, designed to collect information about sampled children before they enter formal schooling, including their home environments and access to early care and education. Once the screener is complete and a preschool child is sampled, based on the criteria described previously, the respondent will be encouraged to continue onto the parent survey. However, if the respondent does not feel he/she can complete the parent survey about the child because he/she is not knowledgeable about the child, there will be an option on the first screen that allows the respondent to indicate they cannot answer questions about the child. The respondent will then be asked to provide the name and contact information for the adult in the household who can complete the survey. Otherwise, the respondent will be taken directly to the online parent survey.The parent survey is expected to take about 30 minutes. At its end, additional contact information will be asked such as the respondent’s mobile phone number, if not already provided, and the respondent will be asked to give permission for the study to send texts to the respondent. Upon completion of the online survey, respondents can immediately receive an Amazon code valued at ten dollars. If they would prefer not to receive the code, a ten-dollar bill will be mailed to them.

Table B-3 shows the schedule of the follow-up reminders to complete the parent survey for the respondents who complete the screener on the web, and Table B-4 shows the schedule for the respondent who return a paper screener.

Table B-3. Mailings to recruit respondents to complete the parent survey when screener completed via the web

Activity	Mode	Instrument	Timeline
Email and/or text reminders	web	parent survey	weekly after breakoff
Mail first class reminder to complete web parent survey (with 2-page study description)	web	parent survey (screener complete by web)	3 weeks from breakoff
Mail first class letter, paper parent survey, postage-paid return envelope ($10 pre-paid)	paper	parent survey (screener complete by web)	4 weeks from breakoff
Mail first class letter, paper parent survey, postage-paid return envelope (second mailing)	paper	parent survey (screener completed by web)	5 weeks from breakoff
FedEx letter, paper parent survey, postage-paid return envelope	paper	parent survey (screener complete by web)	6 weeks from breakoff

Table B-4. Mailings to recruit respondents to complete the parent survey when screener completed via paper survey

Activity	Mode	Instrument	Timeline
Mail first class letter, paper parent survey, postage-paid return envelope, (with 2-page study description) ($10 pre-paid)	paper	parent survey (paper screener complete)	week 5 through 9
Mail first class letter, paper parent survey, postage-paid return envelope (second mailing)	paper	parent survey (paper screener complete)	week 8 through 12
FedEx letter, paper parent survey, postage-paid return envelope	paper	parent survey (paper screener complete)	week 11 through 14

Email/Text Reminders (Attachments A13, A13a, A14, and A14a)

If a respondent chooses to provide an email address or mobile phone number, but does not complete the parent survey, he or she will receive weekly email and/or text reminders encouraging the completion of the survey.

Reminder Letter to Complete Web Parent Survey (Attachment A6)

Respondents who have started the parent survey via web, but who have not responded to at least two email and/or text reminders, will be sent a first class letter reminding them to return to the MyECLS website to finish the parent survey, along with a colorful 2-page description of the study and its goals (also known as the study’s “infographic” – Attachment A34).

Paper Parent Survey (Attachments A7-A12)

Respondents who completed a paper screener will automatically be sent a paper parent survey via first class mail once the completed screener is received at Westat (attachment A7). Additionally, respondents who completed the screener via web and were selected to continue but did not complete the parent survey online after email, text, and/or first class letter reminder(s), will also be mailed the paper survey along with a 2-page description of the study (attachment A10). The first paper survey mailing will contain a letter, the paper parent survey, a postage-paid return envelope, and a ten-dollar bill prepaid as an incentive to complete the survey. Although the respondents who complete the web survey receive their incentive upon completion of the survey, respondents who are asked to complete a paper survey are pre-paid to encourage participation. At the end of the paper parent survey, the instrument will ask for additional contact information such as email, mobile phone number, and permission to send texts.

Respondents who completed the paper screener will receive up to two paper parent surveys via first class mail (attachment A10 and A11), then a third copy by FedEx (attachment A12). Respondents who completed the online screener but did not complete the parent survey via web after email, text, and/or first class letter reminder(s), will also receive up to two paper parent surveys via first class mail (attachment A7 and A8), then a final copy of the paper parent survey via FedEx (attachment A9).

Preschool Field Test Fall 2020 Parent Follow-Up

A brief fall follow-up parent survey is planned for the preschool field test. The preschool parent follow-up will ask: (1) whether the child is currently in kindergarten, and (2) if so, the school the child is attending. If the child is not currently in kindergarten, then the parent will be asked: (1) whether the child is currently enrolled in school, and (2) if so, the grade or grade equivalent of the child. This survey is provided in Attachments B1 and B2, and the respondent communication materials in Attachment A.

Email/Text Message Invitation (Attachments A15 & A16)

In the fall of 2020, an email and/or text will be sent to those who in spring 2020 responded via the web to the parent survey (entirely or partially) and provided an email address or mobile phone number. The email and/or text message will contain an invitation to complete the short follow-up web survey. Table B-5 outlines the fall follow-up survey mailings for these respondents. Upon completion of the online follow-up survey, respondents will have the option to immediately receive a five-dollar Amazon code. If they prefer not to receive the code, a five-dollar bill will be mailed to them.

Thank-You/Reminder Postcard (Attachment A17)

One week later, a thank-you/reminder fold-over postcard will be sent first class mail to all respondents, with a reminder of the promised $5 when the survey is complete.

Email/Text Message Reminders (Attachments A18 & A19)

Respondents who received the electronic invitation to complete the fall follow-up survey will receive an email and/or text message reminder in week two and a final email and/or text message in week four of fall 2020 data collection. If a respondent provided an email address and at any point the email reminders are undeliverable, the respondent will receive the mailings designed for those who did not provide any email address as outlined in Table B-6 below.

First Class Mailing of Paper Survey with Prepaid $5 (Attachment A20)

If by the third week respondents have not completed the follow-up survey, they will receive a cover letter, a paper version of the follow-up survey, and a postage-paid return envelope with a $5 bill via first class mail.

FedEx Cover Letter and Fall Follow-up Paper Survey (Attachment A21)

If by the fourth week respondents have not completed the follow-up survey, they will receive a cover letter, a paper version of the follow-up survey, and a postage-paid return envelope via FedEx.

Final Email and/or Text Reminder (Attachments A22 & A22a)

The final email and/or text reminder for the fall follow-up survey for respondents who provided an email address and/or text number and have not yet completed the follow-up survey will be sent in the fourth week.

Table B-5. Fall 2020 follow-up survey mailings to respondents who completed the spring parent survey on the web and provided a valid email address/mobile phone number

Activity	Mode	Timeline
If email and/or mobile number is available
1. Initial email and/or text message request for web survey with $5 promised incentive	web	week 0
2. Thank-you/reminder fold-over postcard for web survey	web	week 1
3. 2^nd email and/or text request for web survey	web	week 2
4. First class mail letter, paper survey, postage-paid return envelope, and prepaid $5	paper	week 3
5. FedEx mail with cover letter, paper survey, and postage-paid return envelope	paper/web	week 4
6. Final email and/or text reminder	paper/web	week 4

Paper Follow-up Survey Invitation

Any respondents who did not provide an email address or mobile phone number, or who completed the parent survey on paper, will receive a hard copy of the fall 2020 follow-up survey. As in the spring 2020, various follow-up methods will be used to remind respondents to complete the survey. Table B-6 outlines the fall follow-up survey mailings for respondents receiving paper surveys. Respondents who complete the paper fall survey will be pre-paid in the form of a five-dollar bill, included in the initial mailing of the paper follow-up survey.

Initial Mailing for Follow-Up Survey (Attachment A23)

Any respondents who completed the spring parent survey on paper or who did not provide a valid email address/mobile phone number will receive a cover invitation letter, a paper follow-up survey, a postage-paid return envelope, and a $5 bill via first class mail.

Reminder Mailing for Follow-Up Survey (Attachment A24)

A week later, those who have not completed the follow-up survey will receive a cover letter with a second paper version of the follow-up survey and a postage-paid return envelope via first class mail. Since these respondents have already received the $5 prepaid with the initial first class mailing, these packages will not include any additional incentive.

FedEx Fall Follow-up Paper Survey Mailing (Attachment A25)

If by the third week respondents have not completed the follow-up survey, they will receive a cover letter and a third paper version of the follow-up survey via FedEx, with a postage-paid return envelope.

Table B-6. Fall 2020 follow-up survey mailings to respondents who completed the parent survey via paper or did not provide a valid email address and/or mobile phone number

Activity	Mode	Timeline
If email/text is unavailable or bounces back and/or paper screener was completed
1. First class mail letter, paper survey, postage-paid return envelope, and prepaid $5	paper	week 0
2. First class mail 2^nd reminder letter, paper survey, postage-paid return envelope	paper	week 1
3. FedEx mail with cover letter, paper survey, and postage-paid return envelope	paper	week 3

B.3 Methods to Secure Cooperation, Maximize Response Rates, and Deal with Nonresponse

This section describes methods for securing cooperation and maximizing completion rates for the preschool field test in the spring and fall data collections. A major challenge in any survey today is obtaining high response rates. The main problem associated with nonresponse is the potential for nonresponse bias in the estimates produced using data collected from those people who do respond. Bias can occur when the people who respond are systematically different from the people who do not. The approach we will use to reduce the potential for bias is to design the data collection procedures and methods so as to reduce nonresponse (e.g., employing both web and paper versions of the survey and timely survey reminders). While the statistical approaches are important in controlling biases and costs, the recruitment and data collection procedures and materials are at the heart of a successful survey.

Instrument length. Overall, instrument design occurs with the knowledge that completion rates typically are higher when the interview is shorter in length. The average length of time to complete items in the ECLS-K:2011 was carefully considered during the design phase to determine if questions should be cut entirely or simplified to keep the interview to the desired time. Usability testing of the screener and parent survey is planned for fall 2019 to obtain estimates of the time it takes to complete the survey so that the length of the instruments can be adjusted if necessary.

Consideration of questions. As much as possible, the survey is streamlined so the questions included are straightforward and easy for respondents to answer. In addition, skip patterns based on answers to questions that appear earlier in the survey are built into the survey so that not all questions need to be asked of all parents. The planned instrument usability testing may also inform the type and format of questions that are included in the preschool field test; in which case a change request will be submitted to OMB in early winter 2019-20.

Non-English instruments. To achieve high response rates, it is important that study procedures work to include non-English speaking parents to the greatest extent possible. As described in the data collection procedures, web and mailed versions of all household screeners, parent surveys, and respondent communication materials (including the reminder messages) will be available in English and in Spanish. The (a) initial letter, (b) reminder letter, (c) reminder and 1^st paper screener cover letter, (d) reminder and 2^nd paper screener FedEx cover letter, and (e) letter for completed web screener (Attachments A1, A3, A4, A5, and A6 respectively) will be in English language on one side and in Spanish language on the reverse side. If a person in a household accesses the website and toggles the website/screener/survey to Spanish, this information will be saved as their chosen language in a database, and any automatically generated reminder emails, text messages, and paper mailings, including hardcopies of the surveys, will be sent to them in both Spanish and English. For households for which their chosen language is not known because they have not begun to complete the online screener, hardcopy versions of screeners and surveys will be mailed in English and Spanish to households in areas with higher percentages of Spanish speakers. Other households will receive emails, text messages, and paper mailings in English language only unless they request a Spanish version (all hardcopy letters and survey instruments provide a phone number at which to request Spanish version of the mailings and surveys).

B.4 Tests of Methods and Procedures

This request is to conduct the ECLS-K:2023 preschool field test, the results of which will inform the sampling and operational procedures employed for the national preschool data collection, planned to take place in 2022. A request to conduct usability testing of the preschool field test survey instruments will be submitted to OMB in August 2019 (under OMB# 1850-0803). Results of the usability testing will be used to revise the preschool field test instruments. Summary of the usability testing results and the revised data collection instruments will be submitted to OMB as a change request in early winter 2019-20. The preschool field test findings will be summarized in the request to conduct the 2022 national preschool round.

B.5 Individuals Responsible for Study Design and Performance

The following individuals are responsible for the study design and the collection and analysis of the data for the ECLS-K:2023.

Jill McCarroll, NCES	(202) 304-2920	Liz Bissett, Westat	(301) 294-4414
Gail Mulligan, NCES	(202) 245-8413	Nancy Vaden-Kiernan, Westat	(512) 351-8270
Chris Chapman, NCES	(202) 245-7103	Michael Brick, Westat	(301) 294-2004

1 Schools that do not have official kindergarten programs but serve kindergarten-age children in an ungraded setting will be included. In this document, whenever the term “school with kindergarten programs” is used, it is meant to include programs that serve kindergarten-age children as well.

2 These children will be included in the preschool field test because the parent-reported predicted fall grade level is likely an error, and the child may instead enter kindergarten in the fall.

3 Only public school names are collected because private schools are not used in the mapping process to create the address sampling frame for the preschool round of data collection.

4 These numbers were estimated using the lower, 7%, end of the range for post office nondeliverable (PND) rate. The estimated numbers of completes for the spring parent survey and fall follow-up survey is 341 and 290, respectively, if the higher, 13%, end of the range for PND rate is used. The numbers 364 and 309 are used to estimate respondent burdens in Supporting Statement Part A of this submission, as they provide the upper limit of the expected number of respondents.

5 Only about 3% of U.S. households are estimated to have a child that meets the eligibility requirements for this study. A total of 467 respondents are estimated to be eligible to complete the parent survey (258 as a result of being screened from the web and 209 as a result of being screened on paper). Of these up to 78% (approximately 364 individuals) are expected to complete the parent survey.

6 Of the 26,040 households that will be invited to complete the web screener, 67% (approximately 17,447 households) are expected not to do so within the allotted time and thus to be sent the paper version of the screener.

7 Respondents who complete the parent survey (either on web or on paper) are eligible to complete the fall follow-up parent survey. It is estimated that up to 78% of the 467 screened, eligible respondents (approximately 364 households) will complete the parent survey and thus be eligible to complete the fall follow-up parent survey.

File Type	application/vnd.openxmlformats-officedocument.wordprocessingml.document
File Modified	0000-00-00
File Created	2021-01-15

Part B ECLSK-2023 Preschool Field Test