 |
 |
Tuesday, Nov 24
|
|
|
|
The PSID Users Guide
MONOGRAPH PURPOSE AND CONTENT
This monograph describes the origins, design, procedures,
and broad analytical potential of one of the major data bases in
the social sciences--the Panel Study of Income Dynamics (PSID).
After a brief review of the PSID's background, the monograph
discusses the major design parameters, field procedures, and data
preparation activities. These chapters describe the data
collection process and serve as important background for
understanding how PSID data come into being and what they
represent. Issues of the quality of PSID data are addressed next,
reporting evidence about a number of different quality
dimensions. The remainder of the monograph delves into the data
themselves--what topics are covered; what data files are
available; and crucial information regarding analysis issues, key
variables, and choice of data files. The "Data Analysis" chapter
provides details of several analysis examples, so that the reader
can see the assembly of parts needed to create estimates of
earnings regressions, long-run poverty status, changes in women's
income following divorce, and correlations between parents'
income and a child's adult income.
The monograph takes the reader from the drawing board to a
completed product with a minimum of detail. Where most
applicable, it notes other PSID documents, such as the PSID's
documentation books and User Guide, that can provide further
details about particular aspects of the study. In the final
chapter, information is provided about obtaining these documents
and the data files themselves. Throughout the monograph italics
are used to distinguish terms with special meaning in the context
of the PSID, and full capital letters are used to designate PSID
variable names.
OVERVIEW OF THE PSID
The Panel Study of Income Dynamics (PSID) is a longitudinal
survey of a representative sample of U.S. individuals (men,
women, and children) and the families in which they reside. It
has been ongoing since 1968. Data are collected annually, and the
data files contain the full span of information collected over
the course of the study. PSID data can be used for
cross-sectional, longitudinal and intergenerational analyses, and
for studying both individuals and families. The study emphasizes
the dynamic aspects of economic and demographic behavior, but it
contains a wide range of measures, including sociological and
psychological ones. Between 1968 and 1988, the PSID collected
information regarding approximately 37,500 individuals and
spanning as much as 21 years of their lives.
The general design and core content of the study have remained
largely unchanged, and considerable effort has been expended
cleaning the data. These two features greatly enhance the PSID's
potential for longitudinal analysis. Preparation and distribution
of comprehensive documentation and a User Guide also facilitate
use of the PSID data.
The study has been conducted at the Survey Research Center,
University of Michigan since its beginning in 1968, with the
Inter-University Consortium for Political and Social Research
(ICPSR) data archive handling the public distribution of the data
files, documentation, and User Guide. PSID data files have been
disseminated widely throughout the United States and to numerous
foreign countries.
Starting with a national sample of approximately 4,800 U.S.
households in 1968, the PSID has traced individuals from those
households since that time, whether or not they are living in the
same dwelling or with the same people. Adults have been followed
as they have grown older, and children have been observed as they
advance through childhood and into adulthood, forming families of
their own. Each year information is collected about the PSID's
sample members (members of the PSID's 1968 sample families or
their offspring) and their current co-residents (spouses,
cohabitors, children, and others living with them), even if those
co-residents were not part of original-sample families.
Because the original focus of the study was on the dynamics of
poverty, the 1968 sample included a disproportionately large
number of low-income households. The oversampling of families
poor in the late 1960s resulted in a sizable sub-sample of
blacks. Probability-of-selection weights enable analysts to make
estimates from the sample that are representative of the U.S.
population. In the absence of nonresponse bias, the PSID's rules
for tracking individuals and families over time lead to accurate
representation of the nonimmigrant U.S. population both
cross-sectionally each year, and in terms of change, since 1968.
To help correct for omissions in representing post-1968
immigrants, a representative sample of 2,043 Latino (Mexican,
Cuban, and Puerto Rican) households was added in 1990.
The study's tracking rules, along with its Latino
subsample addition in 1990, have meant substantial increases in
the number of individuals in the study as it has progressed
through time. In 1968 the PSID gathered information about
approximately 18,000 individuals; by 1988 this number had grown
to a cumulative total of about 37,500. Similarly, the number of
family units has increased from just under 5,000 at the beginning
of the study to about 7,000 currently, not including Latino
households.
The PSID provides a wide variety of information at the family and
individual level, as well as some information about the locations
in which sample households reside. The central focus of the data
is economic and demographic, with substantial detail on income
sources and amounts, employment, family composition changes, and
residential location. Content of a more sociological or
psychological nature is also included in some waves of the study.
Information gathered in the survey applies to the circumstances
of the family unit as a whole (e.g., type of housing) and to
particular persons in the family unit (e.g., age, earnings). Some
data are collected about all individuals in the family unit, but
the most extensive data are gathered for the family head (who is
male in married-couple families, but female or male otherwise)
and wife. Information about the study's core topics (e.g.,
income, employment, family composition) is gathered annually, and
this is supplemented with data on additional topics (e.g.,
health, wealth, retirement plans, flows of time and money help
among families and their friends, and motivation and efficacy)
gathered intermittently. The amount and variety of data are
substantial; over 300 pages are required to list, by topic and
wave, the variables on the study's main, cross-year data file.
The PSID staff merges each new wave of data with prior waves to
provide comprehensive coverage of information collected for
individuals and families over the entire course of the study.
These multi-wave data files become publicly available upon
completion of the merging, numerous data-quality checks, and
generation of variables. This usually occurs 18-24 months
following the completion of interviewing.
ORIGIN OF THE PSID
As part of Lyndon Johnson's War on Poverty, the Office of
Economic Opportunity (OEO) directed the U.S. Bureau of the Census
to conduct a nationwide assessment of the extent to which the War
on Poverty was affecting people's economic well-being. This
Census study, called the Survey of Economic Opportunity (SEO),
completed interviews with about 30,000 households, first in 1966,
and again in 1967.
Interest in continuing this national study of economic well-being
led OEO to approach the Survey Research Center (SRC) at the
University of Michigan about interviewing a subsample of
approximately 2,000 low-income SEO households. Professor James N.
Morgan, who became the new study's director at SRC, argued
successfully for adding a fresh cross-section of households from
the SRC national sampling frame so that the new study would be
representative of the entire population of the United States,
including non-poor as well as poor households. It was also
decided to follow, and keep as part of the sample, members of the
families who moved away from their original households to set up
new households, such as children who came of age during the
study. In this way, the sample could remain representative of the
nation's families and individuals over time.
The study came to be known as the Panel Study of Income Dynamics.
It began interviewing in 1968, successfully completing interviews
with 4,802 households across 40 states--1,872 low-income
households from the SEO plus 2,930 households drawn from the SRC
national sampling frame. The year 1991 marked the study's 24th
annual wave of interviewing, with its family units having
substantially increased in number and having spread to cover all
50 states as well as some other countries.
ADMINISTRATION AND FUNDING
The PSID has been funded principally by a collection of federal
agencies, including the Office of Economic Opportunity; the
Assistant Secretary for Planning and Evaluation of the Department
of Health, Education and Welfare (now Health and Human Services);
the Departments of Labor and Agriculture; the National Science
Foundation; the National Institute of Child Health and Human
Development (NICHD); and the National Institute on Aging (NIA).
The Ford, Sloan, and Rockefeller foundations have provided
important supplementary grants to the PSID. Since 1983, the
National Science Foundation (NSF) has been the principal sponsor
of the study, with substantial continuing support from the Office
of the Assistant Secretary for Planning and Evaluation (ASPE) of
the Department of Health and Human Services. Since 1982, the
study has had an advisory Board of Overseers, created by the
National Science Foundation to foster input from the national
community of scholars, researchers and policy makers.
Throughout its history, the PSID has been conducted at the Survey
Research Center, which is located within the Institute for Social
Research at the University of Michigan. Professor James N.
Morgan, now emeritus, directed the study from 1968-1989.
Beginning in 1982, Professor Greg J. Duncan became co-director,
and subsequently director when Professor Morgan retired in 1989.
ILLUSTRATIVE USES OF THE PSID
Two key features give the PSID its unique analytic power: (i)
individuals are followed over very long time periods and in the
context of their family setting; and (ii) families are tracked
across generations, with interviews often conducted
simultaneously with multiple generations of the same families.
The type of information the study collects, in conjunction with
these unique qualities, builds a number of strengths, including
the following:
- Continuous representation of families and of individuals of all
ages.
- Long annual--and, in some cases, monthly--time series of
employment, income, and demographic information, reported through
annual interviews.
- Extensive intergenerational information, with a long time-series
of adulthood information obtained from each generation directly,
information after individuals have become adults as well as
during their childhoods, and comparable detail for all children
from the same families.
- Coverage of diverse supplemental topics (e.g., health,
wealth, saving, kinship).
- Recent additions of information, accessible to data users
under special circumstances, about neighborhoods (e.g., Census
tract) and about health and mortality (e.g., from Medicare
records and the National Death Index).
- Extensive longitudinal, as well as cross-sectional, checking
of the data and comprehensive documentation of the full data set
since its start in 1968.
These features make the PSID one of the most widely used and
influential data sets in the social science research community.
Some 200 institutions have requested copies of the PSID data. And
over 700 publications using PSID data have appeared in economic,
demographic and sociological journals and books. The data are
also extensively used for dissertations, reports, conference
presentations, and working papers. A comprehensive bibliography
is available from the PSID staff upon request (see final section
of this monograph).
Areas of basic economic research addressed with the data include:
labor supply, consumption, life-cycle earnings, unions,
compensating wage differentials, dynamic aspects of income
distribution and various methodological studies. PSID topics of
interest to researchers in several disciplines--demographers,
sociologists, psychologists, and economists--include poverty and
welfare experiences during adulthood or childhood; motivation and
economic mobility; changes in family structure (e.g., births,
divorce, remarriage); child support; out-of-wedlock births;
teenage childbearing; and the intergenerational transmission of
economic status. This diversity of topics reflects the philosophy
of the PSID to ask limited sets of questions about a wide variety
of topics rather than extensive questions about only a few
topics. The study's multi-faceted information is couched in the
context of substantial detail about income, employment, and
family composition.
| |
Institute for Social Research |
University of Michigan |
Privacy |
Conditions of Use
|
