Part B_TRIM_Customer Experience Feedback_Final

Supporting Statement for OMB 1530-NEW

Online Research Survey to Solicit Customer Experience Feedback for Treasury Retail Investment Manager (TRIM)

B. Collections of Information Employing Statistical Methods

1. Universe and Respondent Selection.

Quantitative Study

The quantitative study will include a respondent universe of two groups: current customers and potential customers.

Current customer group: Any individuals who have had a financial transaction on TreasuryDirect since 1/1/2018

Total Population: 340,874

To ensure a confidence interval of 99% and a margin of error of 4%, we are targeting 1,000 respondents using a proportional stratification across product types purchased (bonds, marketable securities, and both product types).

We are assuming a 1% response rate, as these customers are rarely contacted and we do not know the quality of the list.

	Bond-only	Bond & marketable	Marketable-only	Total
Size	234069	34294	72511	340874
% Total Population	69%	10%	21%
Estimated Response Rate	1%	1%	1%
Total individuals contacted for quant	68667	10061	21272	100000
Total respondents	687	101	213	1000

We are using a best effort recruit on this stratified list of customers because, ultimately, we are we are seeking feedback from our most engaged customers (so their willingness to opt in to participate in the study is a necessary qualification).

Potential Customer Group: Americans 22 to 85 with an income above $20,000 per year (essentially, adults not in the bottom quintile of income), who are aware of personal finance concepts and are in our survey partner’s online panel. We estimate this group is approximately 2,337,000 individuals. Since one cannot technically calculate a response rate for an online panel (there is not a standardized method to do so with an opt-in panel), we are instead assuming a 45% completion rate. Further, we are not stratifying the panel. We are seeking 2,000 respondents to give us a 99% confidence interval and a 4% margin of error.

	Potential Customers
Size	2,337,000
Estimated Completion Rate	45%
Total individuals contacted for quant	4444
Total respondents	2000

To ensure we get qualified respondents (e.g., those most likely to participate in the TreasuryDirect program), we are leveraging the following screening (termination) criteria in our survey:

• Those of ages less than 22 years old OR age is greater than 85 years old will be terminated

• Those with a household income less than $20,000 per year will be terminated

• Individuals must make decisions on their household finances (either sole or joint decisionmakers), those who do not make financial decisions will be terminated

• Individuals are familiar with the following terms: return, risk, interest rate, U.S. Treasury, and/or brokerage – those who are unfamiliar will be terminated

• Individuals have used a mobile app or website to manage their finances within the past year – those who have not will be terminated

Qualitative Study

The qualitative study will include a respondent universe of two groups: current customers and potential customers.

Current customer group: Any individuals who have had a financial transaction on TreasuryDirect since 1/1/2018

Total Population: 340,874

We are targeting 12 current customers to have a valid qualitative sample.

We are assuming a 1% response rate

As with the quantitative study, we are stratifying our panel using a proportional stratification across product types purchased (bonds, marketable securities, and both product types).

	Bond-only	Bond & marketable	Marketable-only	Total
Size	234069	34294	72511	340874
% Total Population	69%	10%	21%
Estimated Response Rate	1%	1%	1%
Total individuals contacted for qual	824	121	255	1200
Total respondents	8	1	3	12

We are using a best effort recruit on this stratified list of customers because, ultimately, we are seeking feedback from our most engaged customers (so their willingness to opt in to participate in the study is a necessary qualification).

Potential Customer Group: Americans 22 to 85 with an income above the poverty line, who are aware of personal finance concepts and are in our recruitment partner’s participant database. We estimate this group is approximately 50,000 individuals. Since one cannot technically calculate a response rate for an opt-in panel (there is not a standardized method to do so with an opt-in panel), we are instead assuming a 20% completion rate of the screener. To get a robust qualitative sample, we are recruiting 60 people.

	Potential Customers
Size	50,000
Estimated Completion Rate	25%
Total individuals contacted for quant	240
Total respondents	60

2. Procedures for Collecting Information.

Quantitative Study

We are doing a best effort recruit on both the customer and potential customer groups. No statistical methodology for stratification and sample selection will be used.

Further, since we are targeting a specialized (non-representative sample), we will not leverage a probability-based panel. Therefore, no estimation procedure is needed.

We are seeking the following degree of accuracy within this one time study:

For the current customer group, we are seeking of a 99% confidence interval and a 4% margin of error

For the potential customer group, we are seeking a 99% confidence interval and a 3% margin of error

This is a one-time study and we do not need to employ periodic data collection cycles to reduce burden

Qualitative Study

This study is qualitative in nature and will leverage a best effort recruit. The questions above do not apply to this study.

3. Methods to Maximize Response.

Best Effort Recruit:

For the purposes of our research outlined above and in supporting statement A, we are conducting a best effort recruit to reach our target respondent count for both studies. To encourage participation and quality of responses, we will contact potential participants up to two times and will employ incentives for participation as outlined in supporting statement A. Our collection will not employ probability-based sampling.

The specific uses of the information collected in our studies will be:

• To understand how to better position TreasuryDirect to potential customers

• To gain directional feedback on elements of the TreasuryDirect experience such as messaging, necessary content, tone, security features, and digital product features; which will be utilized by the TRIM development and program teams to improve the features and functionality of TreasuryDirect

4. Testing of Procedures.

There will be no formal test phase of the sampling procedures.

We will pause our fielding/data collection when we have 10% of our sample to ensure the quality of the information we collect is valid. This will include adjusting our survey instrument and data quality controls, if necessary.

5. Contacts for Statistical Aspects and Data Collection.

Provide the name and telephone number of individuals consulted on statistical aspects of the design and the name of the agency unit, contractor(s), grantee(s), or other person(s) who will actually collect and/or analyze the information for the agency.

Agencies and Entities:

• Bureau of the Fiscal Service – Doug Anderson, 304-480-7047, [email protected]

• Treasury Relations Support Office (TRSO), in FRB St. Louis – Alex Baur, 314-323-1103, [email protected]

• FRB Minneapolis – Sharon Hill, 612-204-6930, [email protected]

• Slalom Consulting – Kelley Boumgarden, [email protected]

6