Part A_NSCH 2020 Production OMB_FINAL

Department of Commerce

United States Census Bureau

Request for OMB Review

Supporting Statement A

National Survey of Children's Health

OMB Control No. 0607-0990

1. Justification

1. Circumstances Making the Collection of Information Necessary

Sponsored primarily by the U.S. Department of Health and Human Services’ (HHS) Health Resources and Services Administration’s Maternal and Child Health Bureau (HRSA MCHB), the National Survey of Children’s Health (NSCH) is designed to produce data on the physical, emotional, and behavioral health of children under 18 years of age in the United States. The NSCH collects information on factors related to the well-being of children, including access to and quality of health care, family interactions, parental health, school and after-school experiences, and neighborhood characteristics. The goal of the 2020 NSCH is to provide HRSA MCHB, their supplemental sponsoring agencies, states, and other data users with the necessary data to support the production of national estimates yearly and state-based estimates with pooled samples on the health and well-being of children, their families, and their communities as well as estimates of the prevalence and impact of children with special health care needs.

The Health Resources and Services Administration’s Maternal and Child Health Bureau (HRSA MCHB) redesigned the NSCH (Blumberg, Foster, Frasier, et al., 2012)¹ and its companion survey, the National Survey of Children with Special Health Care Needs (NS-CSHCN; Bramlett, Blumberg, Ormson, et al., 2014)² into a single, combined survey for the first time in 2016. This updated survey, which incorporates questions from both of the previous surveys, retains the name National Survey of Children’s Health and utilizes an Address-Based Sampling (ABS) frame.

The U.S. Census Bureau conducts the NSCH on the behalf of the HHS under Title 13, United States Code, Section 8(b), which allows the Census Bureau to conduct surveys on behalf of other agencies. Title 42 U.S.C. Section 701 (a)(2) allows HHS to collect information for the purpose of understanding the health and well-being of children in the United States.

Through partner agreements, the United States Department of Health and Human Services’ Centers for Disease Control and Prevention, National Center on Birth Defects and Developmental Disabilities (HHS/CDC/NCBDDD) and the United States Department of Agriculture (USDA) support specific content on the NSCH topical questionnaires. The CDC/NCBDDD supports content on the receipt of training or interventions around the behavioral treatment of attention-deficit disorder and attention-deficit/hyperactivity disorder under the Public Health Service Act, Section 301, 42 U.S.C. § 241. The USDA supports content on food sufficiency under the Healthy, Hunger-Free Kids Act of 2010, Pub. L. 111-296. In particular, 42 U.S.C. 1769d(a) authorizes USDA to conduct research on the causes and consequences of childhood hunger included in 1769d(a)(4)(B), the geographic dispersion of childhood hunger and food insecurity.

Additionally, if approved, the upcoming cycle of the NSCH will feature four individual state oversamples that include the selection of a pre-determined number of sample cases above the current production base sample in those states. The state oversamples will be sponsored by the State of Colorado, the State of Nebraska, the Oregon Center for Children and Youth with Special Health Care Needs, and the State of Wisconsin.

This request to revise the clearance terms under OMB control number 0607-0990 covers the 2020 NSCH, which is the fifth annual production fielding since the redesigned survey was pretested in 2015. The 2020 NSCH is built on the preliminary results of the 2019 NSCH and benefits five years of developmental work, including:

Contact Strategy Evaluation

For reference within this section, a screener can be defined as the first part of the questionnaire that rosters all children under the age of 18 who live at the sampled address (most of the time). A topical can be defined as the detailed data collected about a single child that was selected from the screener roster to be the topic of the second part of the questionnaire (i.e., topical questionnaire).

• In 2015, we learned that topical conversion is higher for web respondents. Therefore, the Web Push group produced more completed topicals when compared to the Mail group.

• In 2016, we moved forward with the Web Push strategy, but selected addresses with a low probability of web response (High Paper) to receive a paper screener earlier. It was determined that the addresses flagged as such were less likely to respond in general (versus only less likely respond by web).

• In 2017, the High Web model from 2016 was modified to better target addresses that would actually respond by paper and only paper (High Paper). We learned that the new approach to mode assignment was more effective at identifying paper-responding households, but there is need for additional evaluation to better identify this population.

• In 2017, we also learned that a pressure-sealed reminder postcard containing web login information is highly effective, increasing returns by 25% or more in the first month of data collection.

• In 2018, we further improved the High Web model and included up to 2 pressure-sealed reminder postcards. Adding the first pressure-sealed postcard in 2017 was associated with a 67% increase in screener returns through the first four weeks of data collection. The second pressure-sealed postcard in 2018 was associated with a 23% increase in screener returns during the next four weeks of data collection. Gains from the second pressure-sealed reminder postcard were replicated in 2019.

• In 2019, we introduced a new contact strategy called the screener card. The screener card was a perforated slip of paper at the bottom of the contact letter that could be detached and returned in a postage-paid envelope. It was designed to streamline response for households without children. We anticipated that this new strategy could allow households with children to respond via web or paper at a reduced cost. Preliminary results show that both households with and without children were more likely to submit a screener when they were assigned to this test treatment group. Further evaluation of this contact strategy is planned for the 2021 NSCH.

Incentive Experiments³

• In 2015, we learned that a $10 screener incentive produced a negligible increase in returns over the $5 screener incentive resulting in the $5 incentive being much more cost effective.

• In 2016, addresses were divided equally between three incentive groups ($0, $2, or $5) for the initial mailing. The $2 incentive increased topical response (among eligible households) by 3.3 percentage points (29.7% to 33.0%); the $5 incentive increased topical response (among eligible households) by 6.6 percentage points (29.7% to 36.4%). While incentives increased costs, they have been the most effective treatment for increasing response and reducing nonresponse bias.

• Also in 2016, the third topical mailing assigned 10% of addresses to the control and 30% each to $2, $5 and $10 treatment groups. Response propensity by treatment group: $0 - 10.9%; $2 - 17.3%; $5 - 22.0%; $10 - 23.9%. The $10 incentives engendered the highest response; however, $2 and $5 incentives were particularly cost effective.

• In 2017, 90% of addresses received a $2 incentive in the initial screener mailing. The $2 incentive increased topical response (among eligible households) by 4.1 percentage points and cost by $1.76 per address. It is generally more cost effective than a third or fourth nonresponse follow-up mailing and reduced nonresponse bias.

• Also in 2017, 90% of addresses received a $2 incentive in topical mailings 7 and 8 (the addresses being mutually exclusive). The $2 incentive increased the odds of response by 51% at approximately half the average cost per topical. We learned that the $2 topical incentive was cost effective.

• In 2018, 90% of addresses received either a $2 (45%) or a $5 (45%) incentive in the initial screener mailing. The $2 incentive increased topical response (among eligible households) by 3.5 percentage points and cost an additional $1.67 per address. The $5 incentive increased topical response (among eligible households) by 6.9 percentage points and cost an additional $4.16 per address. The screener incentive proved effective at obtaining response from groups otherwise less likely to respond, thereby reducing nonresponse bias.

• Also in 2018, 90% of addresses received a $5 incentive in their initial topical mailing. This incentivized group saw an increase in topical response of 12.2 percentage points. It was more cost effective to use the incentive in the initial mailing than to send nonresponding addresses additional follow-up mailings.

• In 2019, the incentive structure remained the same as 2018, with 90% of addresses receiving either a $2 (45%) or a $5 (45%) incentive in the initial screener mailing and 90% receiving a $5 incentive with their initial topical mailing. Further analysis on the populations for which the incentive was most effective will be discussed within the 2019 NSCH Methodology Report that will be publicly available in the fall of 2020.

Packaging and Branding

• In 2015, all nonresponding addresses received a traditional postcard reminder after the first screener mailing, and a third mailing delivered by FedEx. Very little information was printed on the postcard due to privacy restrictions. Since all addresses received the same treatments, we cannot directly evaluate their effectiveness. The FedEx mailing may have increased response, but has been cost prohibitive on the full-scale production NSCH.

• In 2016, the second screener mailing contained a branding experiment. Approximately half of the addresses received materials with Census branding, while the other half received materials with HRSA MCHB branding. We learned that return rates were not significantly different when households received Census versus HRSA MCHB branding (36.4% vs 35.9%, respectively), so we have continued to use Census branding in our future iterations.

• In 2017, an infographic was included with 50% of all initial packages. While 37.3% of all addresses returned a screener, only 36.8% of those addresses that received the infographic returned a screener. From this, we concluded that the infographic was not effective.

• In 2018, a United States Postal Service (USPS) non-signature required certified mail sticker was attached to 50% of the initial mail packages. When delivered, the certified sticker was effective at motivating response, increasing response by 7.5 percentage points. However, too often the package was not delivered and was actually returned by the USPS due to a missing addressee or perceived requirement for a signature. Limitations with this type of delivery have excluded it from future considerations, but additional delivery method treatments will be explored in future rounds.

• In 2019, a test of standard production envelopes against a redesigned envelope that featured color text and “ways to respond” icons was conducted. Preliminary results showed no significant impact in response rates for the redesigned envelope treatment group. Additional research will be conducted in future rounds to evaluate a fully redesigned suite of materials versus a redesigned envelope alone.

There are a number of differences between the 2019 NSCH and the 2020 NSCH for which we are requesting OMB approval. These differences will be discussed in further detail throughout Supporting Statements A & B, but have been summarized here for ease of reference:

• Increased sample size - With additional sponsor funding and realized cost savings from streamlining the survey operations process, we are requesting an increase in sample size. The base NSCH sample plus the proposed state oversamples may reach up to 240,000 addresses for the 2020 NSCH.

• Unconditional incentive distribution percentage - We plan to continue monitoring the effectiveness of the unconditional incentive, but request an increase to the percent of addresses receiving a $5 incentive in the initial screener mailing. For both the 2018 NSCH and 2019 NSCH, the initial screener incentive splits were 45% received $2; 45% received $5; and 10% did not receive an incentive. The proposal for 2020 NSCH is that 30% receive $2; 60% receive $5; and 10% would not receive an incentive with the initial mailing. The incentive assignment to each sampled address would still be random as was done in prior cycles and approved by OMB. Results from the 2018 NSCH indicate that the increased incentive amount proved effective at obtaining a higher response in general and particularly so from underrepresented population groups⁴. Therefore, the goal of a larger $5 incentive treatment group is to further reduced nonresponse bias.

• Redesigned survey contact materials test - Instead of testing a redesigned envelope in the initial mailing only, as was done in 2019, the 2020 NSCH plans to assign a treatment group (30% of addresses randomly selected from the production sample) to receive a redesigned suite of mail packages (screener and topical) throughout the entire data collection period. Two rounds of cognitive testing were approved previously by OMB⁵ and will be used to inform development of the redesigned contact materials and envelopes. The goal of the redesign is to provide the sampled addresses with a cohesive set of items within each survey invitation package. The proposed materials include key facts pertaining to survey data usage, relatable images for the target population, and colors that match the associated paper questionnaires. We plan to evaluate the experimental treatment group after data closeout to determine the effectiveness of the redesigned package strategy and inform future decision-making.

• Revised questionnaire content – The NSCH questionnaires with newly proposed and revised content from the sponsors at HRSA MCHB underwent two rounds of cognitive testing. This testing request was submitted under the generic clearance package and was approved by OMB⁶. Based on the results, a final set of proposed modified content for the 2020 NSCH is outlined in Appendix A.

• State oversample⁷ - In order to inform state-level decision making around various priorities, some stakeholders have shown interest in sponsoring an oversample of addresses within their state as part of the annual NSCH administration. Currently, four states (Colorado, Nebraska, Oregon, and Wisconsin) are moving forward with this option for the first time as part of the 2020 NSCH. Oversamples will provide states with more robust data for analysis and planning at the state level. The oversamples can be classified as either a general state-wide oversample or sub-state oversample. The state-wide oversample increases the total number of sampled addresses within a given state and will be distributed to the geographic areas similarly to the production base sample. State-level estimates of rare populations or outcomes could be evaluated from this larger sample, but sub-state (e.g., county-level) estimates could not. The sub-state oversample is aimed at producing smaller than state-level (e.g., county or county-level grouping) estimates in combination with the NSCH base sample to reach a specific sample size in each targeted group. The requirements to meet each sub-state oversample are primarily determined by county for the 2020 NSCH.

• Envelope Size Test (Screener Only) – Since we do not know the impacts of the envelope size itself on response within the Low Paper (70%) treatment group, we are planning to test the effectiveness of mailing a single letter in a flat envelope (which is currently used for our screener and topical paper questionnaires) versus the standard business size envelope. This test would occur with the first nonresponse follow-up mailing.

• USPS Priority Mail Envelope Test (Topical Only) – Since the certified mail test from the 2018 NSCH was considered successful when delivered, the 2020 NSCH plans to make use of a priority mail envelope that was created in collaboration with the United States Postal Service. This envelope can be delivered to a household without requiring a signature or an individual’s name within the address field. This test would be conducted with 50% of the initial paper topical mailings.

• Ongoing Cognitive Testing and Methodological Projects – Continuous testing of the redesigned NSCH questionnaire and contact materials for cycle 2020 and beyond. Future modification that might impact the instruments and/or burden estimates will be submitted as non-substantive change requests and/or generic clearance requests for OMB review, as applicable. Non-substantive change and generic clearance requests will be submitted to request permission to make subsequent minor modifications to the questionnaire(s) and to continue conducting methodological testing.

2. Purpose and Use of Information Collection

The NSCH is the only survey of its kind that collects information on factors related to the health and well-being of children at the state and national level. This includes access to and quality of health care, family interactions, parental health, school and out-of-school experiences, and neighborhood characteristics. Data from the NSCH are used to measure progress on national performance and outcome measures under the Title V Maternal and Child Health Services Block Grant. This information further informs state-level planning and program development, federal policy and program development, and general scientific research. It is therefore critical that the U.S. Census Bureau conducts this survey on behalf of the HRSA MCHB.

Information quality is an integral part of the pre-dissemination review of the information disseminated by the Census Bureau (fully described in the Census Bureau’s Information Quality Guidelines). Information quality is also integral to the information collections conducted by the Census Bureau and is incorporated into the clearance process required by the Paperwork Reduction Act.

In recent years, the declining willingness of the public to participate in surveys along with changes in household telephone use has resulted in lower response rates for Computer-Assisted Telephone Interviewing (CATI) surveys, the prior mode of data collection for NSCH and NS-CSHCN. Of particular concern is the increasing prevalence of households that have substituted wireless phone service for landline telephone service (Blumberg & Luke, 2015)⁸. The decline in response rates and difficulties in providing a representative sample at reasonable costs continue to be significant parts of planning considerations for the 2020 NSCH. The 2020 NSCH will continue to follow the redesign recommendations and utilize a two-phase multimode (Web or paper) data collection design for a combined NSCH/NS-CSHCN survey. The NSCH consists of two questionnaires: (1) an initial household screener to assess the presence of children in the home and facilitate the selection of a target child within the household (with oversampling of children with special health care needs), and (2) a substantive topical questionnaire that combines selected content from the former NSCH and NS-CSHCN questionnaires, along with updated content.

Increasing response and minimizing nonresponse bias continue to be two high priority focuses of the NSCH. For that reason, the 2020 NSCH is planning for the following design elements and treatment groups (see Table 9A and Supporting Statement B for additional details) that will be assigned to the initial sampled cases:

• Unconditional incentives – Evaluating the relative benefit for reducing survey nonresponse by providing a $2 (30%) or $5 (60%) incentive as a token of appreciation versus a small control group (10%) that receives no incentive. A $5 incentive will also be aimed towards reducing bias and gaining cooperation for those households who answer a paper screener and are mailed their first paper topical questionnaire.

• Internet likelihood – Modeled Web and paper response mode probabilities are assigned to each address and are further broken out into the High Paper (30%) or Low Paper (70%) treatment groups. This sort is done in attempts to target the top 30% of households with the highest paper-only response probability and provide them with a paper questionnaire starting with the initial mailing. The remaining 70% of addresses are offered the Web instrument as the mode of response in the first two mailings before receiving their first paper questionnaire. Results are evaluated and used to improve future iterations of the model.

• Redesign package test – Each sampled address will either receive the traditional NSCH envelope and accompanying materials (70%) or a redesigned envelope and suite of materials (30%) that were informed by two rounds of cognitive testing. Each redesigned package was intended to flow together with each subsequent contact. Research has shown that strategically designed, cohesive materials can positively impact response (Couper, 2008; Kaplowitz et al., 2004)⁹.

• Screener and Topical Envelope Tests – The first nonresponse follow-up screener mailing will evaluate if there is an envelope size preference for those addresses in the “Low Paper” group who receive a web-invite letter only within their package. The initial topical mailings will evaluate if respondents are more likely to open a USPS priority mail package vs. the traditional NSCH envelope. Both tests will have an even split of a 50% test group and 50% control group. Results will help determine future NSCH cycle mailing strategies and best practices.

The sample composition and response characteristics from the 2016 - 2019 NSCH cycles continue to be researched in order to improve the 2020 NSCH and future cycles. The 2016 NSCH enabled the development of refinements in the production use of flags identifying the presence of children in the household, as well as flags indicating the likelihood of responding by Internet. The Internet likelihood flag was revised for 2017 to increase its ability to identify households most likely to respond by a particular mode of data collection, and the flags identifying the presence of children were further refined to more efficiently identify households with children.

Since there continues to be a significant potential for cost savings for Web data collection over paper data collection, we are implementing the improved Internet likelihood flag to predict households’ response mode preferences again in the 2020 NSCH. We will also be utilizing the flags identifying the presence of children in the household to more efficiently sample households with children.

3. Use of Improved Information Technology and Burden Reduction

The 2020 NSCH will be conducted for HRSA MCHB by the Census Bureau in Web Push + Mail or mixed-mode format. The majority of households (70%) will first have the opportunity to respond online via the Centurion Web instrument. Beginning with the second nonresponse follow-up, the data collection efforts will be augmented via the use of online data collection and paper data collection. A smaller percentage of households (30%) will be placed in the mixed-mode group and will receive an initial mailing with both an invitation to respond online via the Centurion Web instrument as well as an invitation to respond via paper. The Centurion Web instrument allows online reporting while minimizing burden and material costs. In addition, the Centurion Web instrument improves the efficiency and accuracy of the data collection process by providing respondents the opportunity to complete both the screener and topical survey instruments at one time. The paper data collection will rely on three complementary survey systems to efficiently administer this mode of data collection: (1) Amgraf One Form Plus, (2) Docuprint, and (3) integrated Computer-Assisted Data Entry (iCADE).

• Online Reporting. The 2020 NSCH will utilize a Web-based survey with follow-up paper data collection as one of the primary collection strategies. The Web-based survey collection mode allows for features that reduce respondent burden as well as report results more quickly and at considerably less cost. In general, respondents find it less taxing to provide sensitive information about their children in self-administered surveys; however, because of the significant number of filter questions, paper-and-pencil versions of the survey appear quite lengthy. The Web-based survey allows for the programming of skip patterns similar to the original telephone interview version of the survey. Thus, the Web-based format allows for the comfort of self-administration with the ease of seeing and subsequently answering only questions relevant to a particular respondent.

• Forms Design. Questionnaires will be created using Amgraf One Form Plus. Completed hardcopy forms can be processed by iCADE to capture responses through optical mark recognition (OMR), optical character recognition (OCR), and keying from image (KFI). Questionnaires will be printed, trimmed, and stitched through an in-house print on-demand process using a Docuprint system which allows personalization and the ability to tailor items to each specific respondent. The data from the questionnaires will be captured by the iCADE technology/software, which automatically extracts all check box entries (OMR) and preselected numeric answer fields (OCR), then captures, and displays an image of all other entries to an operator for KFI.

• Image Preprocessing. The iCADE software performs a registration process for each individual questionnaire page to match to the appropriate page template. This also allows for corrections due to any skewing during scanning.

• Data Capture. iCADE reads the form image files, checks for the presence of data, processes all check box fields through OMR, processes all preselected numeric answer fields through OCR, then presents an image of all other handwritten fields to an operator for KFI.

• Verification. Extracted KFI data are subject to 100% field validation according to project specifications. If a data value violates validation rules, the data point is flagged for review by verifiers who interactively review the images and the corresponding extracted data, and resolve validation errors.

• Archiving. Images will be scanned and archived to magnetic storage located on a secured server in case they are needed later. This eliminates the need to save paper copies of the completed questionnaires.

4. Efforts to Identify Duplication and Use of Similar Information

The NSCH has been conducted since 2003 under the auspices of the Centers for Disease Control and Prevention’s National Center for Health Statistics on behalf of the HRSA MCHB. In tandem with the NS-CSHCN, the NSCH is considered the most robust data source available at national and state levels on children’s health and well-being. These data are cited broadly in research literature (http://www.cdc.gov/nchs/slaits/slaits_products.htm).

Previously, there was significant duplication between the NSCH and the NS-CSHCN. A key objective in developing the 2016 NSCH instrument was to consolidate the prior version of the NSCH and the NS-CSHCN into one survey, reducing redundancy in the collection of data and the burden on households, which accompanied the administration of two separate surveys. The 2015 NSCH pretest and 2016 NSCH iterations demonstrated the feasibility of conducting the new condensed NSCH using web and mail as new modes of administration.

The 2020 NSCH plans to include a change of content (see Appendix A) to support programs and policies related to children’s health and children with special health care needs. Along with some newly added questions, this list also includes a small set of questions that would be removed from this administration of the survey in order to maintain the overall burden per respondent.

5. Impact on Small Businesses or Other Small Entities

Not applicable.

6. Consequences of Collecting the Information Less Frequently

The 2020 NSCH is the fifth year of production in an annual effort to collect and produce data on the physical and emotional health of children under 18 years of age living in the United States. The NSCH collects information on factors related to the well-being of children, including access to and quality of health care, family interactions, parental health, school and out-of-school experiences, and neighborhood characteristics. NSCH data are used to measure progress on national performance and outcome measures under the Title V Maternal and Child Health Services Block Grant Program in HHS. Without the annual collection of this data, the HRSA MCHB would not be able to produce these timely national performance and outcome measures.

7. Special Circumstances Relating to the Guidelines of 5 CFR 1320.5

This data collection will be consistent with the general information collection guidelines of

5 CFR 1320.5. No special circumstances apply.

8. Comments in Response to the Federal Register Notice/Outside Consultation

The 60-day Federal Register Notice was published in the Federal Register on November 19, 2019 (84 FR, No. 223; p. 63839-63842). No substantive comments were received.

9. Explanation of any Payment/Gift to Respondents

Incentives were treated as a design element for each of the prior cycles of the NSCH that have been administered by Census. The evaluation of results from prior cycles continues to show that there is a statistically significant difference in the response rates among respondents who received an incentive compared to those who did not receive an incentive. In addition, there continues to be an increase in response rates among households mailed a $5 incentive compared to those mailed a $2 incentive with their initial survey invitation. Screener completion rates (from eligible households) for the 2018 NSCH were 43.8% for respondents with no incentives, compared with 48.3% for those with a $2 incentive and 51.6% for those with a $5 incentive. Topical completion rates (from eligible households) for the 2018 NSCH were 33.6% for respondents with no incentives, compared with 37.1% for those with a $2 incentive and 40.5% for those with a $5 incentive. The cost of incentives is offset by the reduction in follow-up effort and the cost required to collect the data.

For the 2020 NSCH production sample, a slightly modified incentive structure is planned: 30% of addresses receive a $2 cash incentive, 60% receive a $5 incentive, and the remaining 10% serve as a control group that do not receive a cash incentive. These incentive amounts will be evenly distributed between the Low Paper/High Paper and redesigned package treatment groups. Survey methods research strongly support the use of unconditional incentives to reduce nonresponse bias in self-administered survey data collection¹⁰. The 2020 NSCH project plan allows for continued monitoring of the effectiveness of cash incentives in the initial mailing. The sample distribution is presented in Table 9A (the experiments listed in the table along with the others mentioned in Supporting Statement A will be discussed in further detail in Supporting Statement B).

Table 9A. Production Treatment Groups by Incentive Amount and Internet Likelihood

Incentive Treatment Group	Initial Cases	Redesign Package Test Group	Maximum Cases for Mailing Comparison	High Paper- Treatment	Low Paper- Treatment
Control	24,000	Redesign Group	7,200	2,160	5,040
		Traditional Group	16,800	5,040	11,760
$2	72,000	Redesign Group	21,600	6,480	15,120
		Traditional Group	50,400	15,120	35,280
$5	144,000	Redesign Group	43,200	12,960	30,240
		Traditional Group	100,800	30,240	70,560

Table 9A NOTE: The redesign package test, high paper, and low paper treatment groups are described in more detail in Supporting Statement B.

Incentives are commonly used in other HHS-sponsored surveys including the National Health Interview Survey (NHIS), the National Survey of Family Growth (NSFG), the National Health and Nutrition Examination Survey (NHANES), the National Survey on Drug Use and Health (NSDUH), and the Health Center Patient Survey (HCPS). Recent experimentation within a general population mixed-mode (Web and Mail) survey found that the use of a prepaid incentive more than doubled the response rate within that population from 25% to 56% (Messer & Dillman, 2011)¹¹.

10. Assurance of Confidentiality Provided to Respondents

The following confidentiality statement will be presented to respondents within both the Centurion Web instrument and paper questionnaires:

The U.S. Census Bureau is required by law to protect your information and is not permitted to publicly release your responses in a way that could identify you or your household. The U.S. Census Bureau is conducting the National Survey of Children's Health on the behalf of the U.S. Department of Health and Human Services (HHS) under Title 13, United States Code, Section 8(b), which allows the Census Bureau to conduct surveys on behalf of other agencies. Title 42 U.S.C. Section 701(a)(2) allows HHS to collect information for the purpose of understanding the health and well-being of children in the United States. Federal law protects your privacy and keeps your answer confidential under 13 U.S.C. Section 9. Per the Federal Cybersecurity Enhancement Act of 2015, your data are protected from cybersecurity risks through screening of the systems that transmit your data.

Access to records maintained in the system is restricted to Census Bureau employees and certain individuals authorized by Title 13, U.S. Code (designated as Special Sworn Status individuals). These individuals are subject to the same confidentiality requirements as regular Census Bureau employees identified above and as permitted under the Privacy Act of 1974 (5 U.S.C. Section 552a) and SORN COMMERCE/CENSUS-3, Demographic Survey Collection (Census Bureau Sampling Frame).

11. Justification for Sensitive Questions

Sensitive questions are generally not included on the NSCH. However, it is possible that respondents may find some questions related to their children’s health or disease status to be sensitive in nature. Respondents are made aware of the voluntary nature of this survey in the cover letter that accompanies the invitation to complete the questionnaire and on the material distributed with the paper questionnaire. Individuals are free to refrain from answering any question that they do not feel comfortable responding to. The U.S. Department of Health and Human Services requires that race and ethnicity be asked on all HHS data collection instruments and questions on both race and Hispanic origin appear on the NSCH. There is, however, no requirement that respondents answer these questions.

12. Estimates of Annualized Hour and Cost Burden

Estimates of annualized hour burden and annualized cost to respondents are listed in Tables 12A and 12B, respectively. The total number of estimated respondents is expected to be approximately 94,370. Of these, approximately 54,774 will complete only the screener (comprising households without children and households with children that do not complete the topical interview), and the remaining 39,596 will also complete the topical interview. The total number of annual burden hours for the return rates mentioned is 29,642. The estimated total annual respondent cost is $835,311.56.

12A. Estimated Annualized Burden Hours

Type of Respondent	Questionnaire Name	Expected Number of Respondents12	Number of Responses per Respondent	Average Burden per Response (in hours)	Total Burden Hours
Adult Parent or Caregiver	Screener Only	54,774	1	.083	4,565
Adult Parent or Caregiver	Screener and 0-5 Topical Instrument	13,199	1	.633	8,359
Adult Parent or Caregiver	Screener and 6-11 Topical Instrument	13,199	1	.633	8,359
Adult Parent or Caregiver	Screener and 12-17 Topical Instrument	13,199	1	.633	8,359
NSCH Burden Total		94,370			29,642

Table 12A NOTES: 1) Details may not sum to totals due to rounding.

12B. Estimated Annualized Burden Costs

Type of Respondent	Total Burden Hours	Hourly Wage Rate	Total Respondent Costs (rounded to nearest dollar)
NSCH Production
Adult Parent or Caregiver (Screener Only)	4,565	$28.18	$128,642
Adult Parent or Caregiver (Screener and 0-5 Topical Instrument)	8,359	$28.18	$235,557
Adult Parent or Caregiver (Screener and 6-11 Topical Instrument)	8,359	$28.18	$235,557
Adult Parent or Caregiver (Screener and 12-17 Topical Instrument)	8,359	$28.18	$235,557
Total	29,642		$835,313

Table 12B NOTES: 1) Details may not sum to totals due to rounding.

13. Estimates of Other Total Annual Cost Burden to Respondents

There are no direct costs to respondents other than their time to participate in the study.

14. Annualized Cost to the Federal Government

Costs for this survey are estimated at $5,645,000. This includes all direct and indirect costs of the design, data collection, analysis, and reporting phases of the survey, as well as delivery of the data sets to HRSA MCHB.

15. Explanation for Program Changes or Adjustments

This is a revision request of a currently approved collection. The burden impact increased between the 2019 and 2020 survey cycles because of an increase in overall sample size and additional sponsored funding. The sample size was approximately 184,000 addresses in 2019. For the 2020 NSCH cycle, the production survey will be mailed to approximately 240,000 addresses. The increased number of sampled addresses are in response to additional funding via the main HRSA MCHB agreement along with the funding received from the four state oversample projects. Each cycle there continues to be a streamlining of NSCH processes with the creation of a mailing strategy that has proven effective in increasing early response and reducing nonresponse follow-up and bias.

Total estimated burden per respondent for the production survey remains the same as was stated within the 2019 OMB request, but the total burden hours for the survey administration are higher due to the increased sample size and a small projected increase in response rate.

Future modification that might impact the instruments and/or burden estimates will be submitted as non-substantive change requests and/or generic clearance requests for OMB review, as applicable. Non-substantive change and generic clearance requests will be submitted to request permission to make subsequent minor modifications to the questionnaire(s) and to continue conducting methodological testing.

16. Plans for Tabulation, Publication, and Project Time Schedule

The following is a project time schedule for the 2020 NSCH:

2020 NSCH Project Time Schedule and Deliverables
Mail Date	Description of Mailing
June 2020	Initial mailout of all treatment group survey invites
July 2020	Pressure-sealed postcard reminder (containing Web login information)
	Low paper first follow-up mailing (Web invite only)
	High paper first follow-up mailing (Web invite & paper questionnaire)
	Low paper pressure-sealed postcard reminder (containing Web login information)
August 2020	High paper pressure-sealed postcard reminder (containing Web login information)
	Low paper second follow-up mailing (Web invite & paper questionnaire)
September 2020	High paper second follow-up mailing (Web invite & paper questionnaire)
	Low paper third follow-up mailing (Web invite & paper questionnaire)
October 2020	High paper third follow-up mailing (Web invite & paper questionnaire)
August 2020 – December 2020	Paper topical questionnaire mailings (only applicable to households who responded by mail with an eligible paper screener)
January 2021	Survey closeout – data collection ends
Summer 2021	Delivery of fully documented public use data sets (topical and screener level files) and any other preliminary data files requested by HRSA MCHB
Fall 2021	Delivery of codebook, user’s manual, and methodology report

The NSCH will generate datasets, statistics, and reports. Below are the deliverables that the Census Bureau intends to provide HRSA MCHB:

Datasets, Statistics, and Reports

• A fully documented public use data set including two types of files:

  • Screener level files – These files will contain all of the child data collected on the screener instruments along with any other variables (derived, flag, admin, etc.) requested by HRSA MCHB.

  • Topical level files – These files will contain all of the child data collected on the topical instruments and any other variables (derived, flag, admin, etc.) requested by HRSA MCHB.

• Codebooks with weighted and unweighted frequencies of all variables for each of the files mentioned above

• A user’s manual and methodology report created by Census staff

17. Reason(s) Display of OMB Expiration Date is Inappropriate

Not applicable. No exception requested.

18. Exceptions to Certification for Paperwork Reduction Act Submissions

Not applicable. No exception requested.

1 Blumberg, S.J.; Foster, E.B.; Frasier, A.M. et al. (2012). Design and operation of the National Survey of Children’s Health, 2007. National Center for Health Statistics. Vital Health Stat 1(55), 1-159. http://www.cdc.gov/nchs/data/series/sr_01/sr01_055.pdf

2 Bramlett, M.D.; Blumberg, S.J.; Ormson, A.E. et al. (2014). Design and operation of the National Survey of Children with Special Health Care Needs, 2009–2010. National Center for Health Statistics. Vital Health Stat 1(57), 1-282. http://www.cdc.gov/nchs/data/series/sr_01/sr01_057.pdf

3 Incentive experiment results can be found within the Methodology Report for each survey cycle: https://census.gov/programs-surveys/nsch/data.html

4 Incentive experiment results can be found within the Methodology Report for each survey cycle: https://census.gov/programs-surveys/nsch/data.html

5⁵ Generic Clearance Information Collection Request: https://www.reginfo.gov/public/do/PRAViewIC?ref_nbr=201606-0607-003&icID=236843

6 Generic Clearance Information Collection Request: https://www.reginfo.gov/public/do/PRAViewIC?ref_nbr=201606-0607-003&icID=237067

7 State Oversampling in the National Survey of Children’s Health: Feasibility, Cost, and Alternative Approaches https://census.gov/content/dam/Census/programs-surveys/nsch/NSCH_State_Oversample_Summary_Document.pdf

8 Blumberg, S.J. & Luke, J.V. (2015). Wireless substitution: Early release of estimates from the National Health Interview Survey, January–June 2015. National Center for Health Statistics. Available from: http://www.cdc.gov/nchs/nhis.htm.

9 Couper, M.P. (2008). Designing effective web surveys. Cambridge University Press, New York.

Kaplowitz, M. D., Hadlock, T. D., & Levine, R. (2004). A comparison of web and mail survey response rates. Public opinion quarterly, 68(1), 94-101.

10Alexander, G.L. et al. (2008). Effect of Incentives and Mailing Features on Online Health Program Enrollment. American Journal of Preventive Medicine, 34(5), 382-388.

11 Messer, B.L. & Dillman, D.A. (2011). Surveying the general public over the internet using address-based sampling and mail contact procedures. Public Opinion Quarterly, 75(3):429 -57.

12 The expected number of respondents is an estimate of the expected number of completed screener and topical questionnaires, discussed in section B.1.3. This is different from the number of respondents that were mailed a screener or topical questionnaire.

12