About | HeinOnline Law Journal Library | HeinOnline Law Journal Library | HeinOnline

105 Monthly Lab. Rev. 40 (1982)
Research Summaries

handle is hein.journals/month105 and id is 238 raw text is: Research
Summaries

Forgotten unemployment:
recall bias in retrospective data
FRANCIS W. HORVATH
It is a well documented fact that the ordering and speci-
fic wording of a survey questionnaire can produce dif-
ferent readings of the same underlying event. Dif-
ferences can also arise when the same general question
is asked of individuals at different times.
One area in which such discrepancies have been
found is in comparisons of unemployment data collect-
ed on a monthly basis with those obtained from a once-
yearly survey. The source of the monthly unemploy-
ment figures is the Current Population Survey (cPs),
conducted by the Census Bureau for the Bureau of La-
bor Statistics. Each month, representatives of approxi-
mately 60,000 households are asked questions about
their labor force activity and that of other household
members during a given reference week. Annual unem-
ployment estimates  are constructed  by  averaging
monthly data for the year.
Annual unemployment data are also obtained, on a
retrospective basis, from the Work Experience Supple-
ment to the Current Population Survey in March of
each year. Persons responding to the supplement ques-
tions are asked to recall events which occurred from
January through December of the previous year. These
respondents are asked to aggregate the year's labor
force activities into summary figures. That is, they are
asked questions such as: How many weeks was (house-
hold member's name) looking for work? and How
many weeks was (household member's name) working?
However, no attempt is made to ascertain the months
in which these events took place.
The supplement data reflect the number of persons
with unemployment at any time during the previous
year; a person experiencing two or more spells of unem-
ployment is counted only once. Recently, this estimate
has been about 18 to 21 -million persons a year. In con-
Francis W. Horvath is an economist in the Division of Labor Force
Studies, Bureau of Labor Statistics.

trast, the annual average of the unemployment data col-
lected monthly-simply the sum of the 12 monthly es-
timates (unadjusted for seasonality), divided by 12-
represents the mean number of persons unemployed
during a typical week of the year. This figure was
about 5 to 7 million during 1975-79. On the surface,
these two figures appear unconnected. However, be-
cause the Work Experience Supplement also includes
questions on the duration of unemployment in the pre-
vious year, it is possible to directly relate the retrospec-
tive supplement information to the annual averages of
the monthly data.
The basic method, developed by Daniel Suits and
Richard Morgenstern,' derives the total number of
weeks of unemployment occurring in the labor force
during the full year from the Work Experience Supple-
ment. Specifically, persons are grouped according to the
supplement information into duration of unemploy-
ment intervals. Multiplying the number of persons in
each group by the midpoint (in weeks) of the duration
interval yields an estimate of the total weeks of unem-
ployment occurring to persons within that particular
cell. The sum over all cells provides an estimate of the
total weeks of unemployment occurring in the labor
force over the year. Dividing this total by 52 gives the
adjusted supplement estimate of a typical week's unem-
ployment, roughly comparable to the results from the
monthly studies.
Exhibit 1 shows this basic relationship, using hypo-
thetical data. Over a given year, there are 155 individu-
als reporting some unemployment. Suppose that most
of these are unemployed in 1 month only, and during
all other months they are either employed or not in the
labor force. However, assume that 25 of the 155 report
unemployment in 2 consecutive months. In the monthly
survey, then, the total reported instances of unemploy-
ment would be 180; annual average unemployment
would be 180 divided by 12, or 15 individuals. With an
accurate annual retrospective survey, the number of per-
sons reporting some unemployment during the entire
year would be 155-180 less the 25 who were unem-
ployed over a 2-month period. Dividing reported weeks
of total unemployment by 52 yields the adjusted ret-
rospective estimate, which is exactly equal to the annual

What Is HeinOnline?

HeinOnline is a subscription-based resource containing nearly 3,000 academic and legal journals from inception; complete coverage of government documents such as U.S. Statutes at Large, U.S. Code, Federal Register, Code of Federal Regulations, U.S. Reports, and much more. Documents are image-based, fully searchable PDFs with the authority of print combined with the accessibility of a user-friendly and powerful database. For more information, request a quote or trial for your organization below.



Short-term subscription options include 24 hours, 48 hours, or 1 week to HeinOnline with pricing starting as low as $29.95

Contact us for annual subscription options:

Already a HeinOnline Subscriber?

profiles profiles most