🚀 go-pugleaf

        Call for Abstracts for a Conference on Attrition in
                       Longitudinal Surveys

     The National Science Foundation, Bureau of Labor Statistics,
Bureau of the Census and National Center for Educational Statistics
are jointly funding a conference on attrition in longitudinal
surveys.

     All longitudinal surveys, including the Panel Study of Income
Dynamics (PSID), the various cohorts of the National Longitudinal
Surveys (NLS), the Survey of Income and Program Participation
(SIPP), the National Longitudinal Study of the Class of 1972, High
School and Beyond, and the National Education Longitudinal Study of
1988 lose portions of their samples to nonresponse. And while it is
easy to calculate response rates and describe the initial
characteristics of subsequent nonrespondents, there has been very
little work on how attrition might affect parameter estimates of
behavioral models.

     In order to encourage researchers to extend their modelling to
include an examination of the effects of attrition, the PSID Board
of Overseers is organizing a conference on the topic. A subset of
papers chosen for the conference will be considered for publication
in the Journal of Human Resources.

     Analyses of attrition in the above-mentioned surveys are
encouraged. Comparative analyses of two or more of these data sets
are especially encouraged, as are studies that bear on the design of
future longitudinal surveys.

     For the PSID, the highest priority area is attrition in the
context of intergenerational models. For the NLS, the first priority
is on longitudinal labor market issues and the second is on
demographic factors as they relate to labor market behavior. Only
proposed analysis of active NLS cohorts (Youth, Young Women and
Mature Women) will be considered. The highest priority for the
Mature Women cohort is retirement behavior. SIPP priority areas
include longitudinal studies relating to income, program
participation and family living arrangements. Priority areas for the
educational longitudinal surveys include studies relating to
education transitions, program participation, family and community
involvement, education-to-work transitions, and the economic returns
of education.

     The calendar for the selection process and completed papers is
as follows:

     * Abstracts are due 15 April 1992. A committee appointed by the
PSID Board of Overseers will select authors by May 15. We anticipate
that about 20 proposed papers will be selected. Abstracts should be
three to four pages long and reflect careful thought about aims,
data, methods and models.


     * A 12-month analysis period is provided for completion of the
papers. We expect that authors will devote considerable effort to
their papers in the summer months of 1992 so that preliminary
versions of papers might be presented at seminars and professional
meetings during late 1992 and early 1993.

     * Completed papers will be presented at a conference in the
spring-summer of 1993 and circulated to discussants and other
authors at least one month prior to the conference. Travel and
living expenses will be paid for by the grant. The editorial board
of the Journal of Human Resources will decide which of the
conference papers they will send out for review for possible
publication in the Journal.

     Abstracts should be sent by Aped 15, 1992 to: Mary Wreford,
Room 3200 ISR, University of Michigan, PO Box 1248, Ann Arbor,
Michigan, 48106-1248. You may s e n d electronic mail to
WREFORD@UM.CC.UMICH.EDU or USERSSWA@UMICHUM.BITNET.



-------------------------------------------------------------------------
Jim Cassell                                    CASSELL@UNCVM1.OIT.UNC.EDU
Institute for Research in Social Science       CASSELL@UNCVM1.BITNET
University of North Carolina-Chapel Hill       Phone:  919-962-0782
Chapel Hill, NC  27599-3355                    Fax:    919-962-4777
-------------------------------------------------------------------------