FR3016_20201029_omb_B

Supporting Statement Part B for the
Ongoing Intermittent Survey of Households
(FR 3016; OMB No. 7100-0150)
Summary
For all information collections that involve surveys or require a statistical methodology,
the Board of Governors of the Federal Reserve System (Board) is required to provide a complete
justification and explanation of the use of such a methodology. For collections that employ
surveys without such a methodology, the Board should be prepared to justify its decision not to
use statistical methods in any case where such methods might reduce burden or improve
accuracy of results.
Background
The Ongoing Intermittent Survey of Households was initiated in 1981. Over the past 38
years, the survey data have helped the Board understand consumer credit markets and consumer
behavior. The Board has used the data to meet the current analysis needs of the Board to respond
to mandates from the Congress, to prepare academic research papers, and to provide information
to the public.
The Board has a contract with the University of Michigan’s Survey Research Center
(SRC) to include survey questions on behalf of the Board in an addendum to the SRC’s regular
monthly Survey of Consumer Attitudes and Expectations. The Board drafts and edits the
addendum questions in consultation with the SRC, whose program involves careful questionnaire
development. The SRC’s survey guidelines produce questionnaires that are clear and reliable,
and also mitigate any duplication between the Board’s addendum questions and the SRC’s
regular questions. The SRC conducts the survey by telephone with a sample of 500 households
and asks additional questions of special interest to the Board.
Universe and Respondent Selection
The monthly Survey of Consumers is an ongoing nationally representative survey based
on approximately 500 telephone interviews with adult men and women living in households in
the coterminous United States (48 States plus the District of Columbia). The sample is designed
to maximize the study of change by incorporating a rotating panel sample design in an ongoing
monthly survey program. For each monthly sample, an independent cross-section sample of
households is drawn. The respondents chosen in this drawing are then reinterviewed six months
later. A rotating panel design results, and the total sample for any one survey is normally made
up of 60% new respondents, and 40% being interviewed for the second time.
The method used to select monthly nationally representative samples of persons is
generally referred to as random digit dialing (RDD) of cellular telephone numbers. However, any
single monthly sample consists of two parts, an RDD sample of cell telephone subscribers
selected in that month and a sample of RDD sample cell telephone subscribers who completed

interviews six months previously. The latter is referred to as the re-contact sample, and the
former the RDD sample.
Procedures for Collecting Information
The Board provides the initial draft of the questionnaire items to be included in the
monthly surveys. After review and discussions, SRC develops a pretest questionnaire. The
refinement of questionnaire items is then guided by rigorous and careful pretesting. A draft
questionnaire is then constructed, and testing is conducted under essentially “final state”
conditions--that is, pretest respondents fit all the eligibility criteria of the study, and experienced
interviewers and supervisory personnel are employed. Two pretests are conducted on all new
questionnaire items before each monthly survey. Most questionnaire inserts can be developed
and pre-tested within one month; complex questionnaire designs may need two months or more,
especially if pretest results indicate significant problems.
The SRC designs and builds codes for converting the interview response data to machineusable form, and for error checking. In the actual coding process, as the respondent’s answers to
each question are categorized, the coder also takes into account all marginal comments or
general notes about the interviewing situation which the interviewer may have written, and
answers given to other questions which may aid in interpreting responses that by themselves are
too vague to be coded with precision.
Methods to Maximize Response
Although strenuous efforts are made to interview the hard-to-contact, the SRC maintains
that the efforts made to win over the reluctant are low-pressured and reasonable. This is
consistent with the voluntary nature of the telephone interview, which respondents are informed
of when their consent is obtained. Past experience indicates that there may be negative outcomes
to excessive pressures on truly reluctant individuals. The SRC has found that data obtained from
suspicious and reluctant respondents may be so inaccurate that the reporting error they introduce
outweighs nonresponse bias. All respondents are mailed a brief report on the results of the study.
For the RDD sample, the byproduct of this is to track those respondents that may have moved, as
well as to motivate all respondents to participate in the reinterview.
If possible, persuasion letters are sent to all but the most vehement refusals. Address
information is usually available for the reinterview cases, but not for the RDD sample unless the
respondent volunteers this information when initially contacted. If it seems appropriate, a
different interviewer will be assigned to make the callback. SRC’s experience is that between 25
and 35 percent of refusals are converted to interviews following these procedures. Interviewers
who have high refusal rates are identified and supervisors work with them to improve their
techniques. If refusals appear to be due to interviewer qualities which are not correctable through
retraining, the work is reassigned.

2

Testing of Procedures
Several special procedures are used to guard against the potential falsification of
interviews and the introduction of interviewer bias into the data. SRC conducts a telephone
validation check on a small subset of all completed interviews. In addition, a statistical
evaluation of the detailed administrative reports provides a second effective means of detecting
deviation from specified procedures. It is often easier for an interviewer to make up responses to
certain questions than it is to make up a normal distribution on process measures such as calls
required to obtain an interview, number of refusals, interview length, edit length, and so forth.
SRC requires detailed reporting of these process measures and employs computerized analysis of
these data to identify outliers whose work is then thoroughly verified.

3