Paper Screener Experiment for NSFG, Implemented in Q3&Q4 of 2022

NSFG OMB Attachment D1 OMB No. 0920-0314

Paper Screener Experiment for NSFG,

Implemented in Quarters 3 and 4 of 2022

The paper screener experiment was implemented in Quarters 3&4 of 2022. Sample addresses were mailed a short paper screener in the third mailing for the NSFG screener, in addition to including the web login information again.

The sole objective of the paper screener was to help increase the screening and overall response rates. The paper screener adds cost and complexity, including necessary procedures and time to mail the screener forms, scan returned forms, process the data and apply respondent selection algorithms, and add information back into the survey systems.

Receipt of completed paper screeners does not imply increased participation. Those who complete the paper screeners may have completed the screener using another mode (web or in-person) if not given the paper screener option. Similarly, someone else in the household may complete the screener in another mode before a completed paper screener is received and processed. Therefore, the paper screener experiment should be evaluated as a net effect on the response rate, rather than limited to the number of received completed forms.

The interim AAPOR RR4¹ unweighted response rates for the screener, conditional male and female, and overall male and female response rates for the conditions without and with a paper screener on the third screener mailing are shown in Table 1. Even if there is a positive impact of the paper screener on the screener response rate, any benefit on the overall response rate may be lost if those who complete the screener on paper are less likely to complete the main survey on the web or in-person. However, there is no beneficial impact of the paper screener on the screener response rate (49.5% with a paper screener vs. 51.5% without a paper screener) or on the conditional main survey response rates.

Table 1. Unweighted Response Rates by Paper Screener Condition, Quarters 3&4 of 2022.

	Paper Screener Condition
	No Paper Screener n=10,069	Paper Screener n=9,921
Screener	51.5%	49.5%
Male Survey	62.9%	62.6%
Female Survey	59.8%	56.7%
Overall Male Response Rate	32.4%	31.0%
Overall Female Response Rate	30.8%	28.1%

Since the overall response rates are the product of the screener response rates and the conditional male and female main survey response rates, there is no increase in the overall response rates from inclusion of the paper screener. The overall response rates are one percentage point lower for males (31.0-32.4) and three percentage points lower for females (28.1-30.8). The paper screener is not recommended for inclusion in the NSFG study design, as tested.

1 The American Association for Public Opinion Research. 2016. Standard Definitions: Final Dispositions of Case Codes and Outcome Rates for Surveys. 9th edition. AAPOR.

Page 1 of 1