Full text: Brückner, Hannah: Surveys don't lie, people do?

in event history analysis. The extent to which mobility and life course disorder (Rindfuss et al. 1987) 
co-vary with data accuracy of their measures and their predictors bears upon the internal validity of 
studies using these concepts. 
In most research on interviewer effects, the interview is seen as a micro-social system consist- 
ing of two roles-interviewer and respondent-linked by the task of transmitting information (Sudman 
& Bradburn 1974). Advances in survey technology, such as computer-assisted telephone interviewing, 
allow more control over the situation in which interviews are conducted. At the same time, they may 
affect data quality in other ways currently not very well understood. **** 
1.1. Trends and Problems in Data Collection 
The multiple levels on which time enters modern data analyses adds duration, sequence, and 
speed of events to the classical age-cohort-period model and renders cross-sectional approaches 
obsolete (Elder 1985, Mayer & Huinink 1990). Life course research has addressed the need for data 
in a number of ways': by collecting and analyzing narratives (see Bertaux 1981, Bertaux & Kohli 
1984), archival data or official records (e.g. Elder & Caspi 1990; Elder & al. 1991; Voges & Zwick 
1990), census data (e.g. Oppenheimer, 1974) and, last but not least, by designing social surveys. 
Survey design followed one of two strategies: The prospective panel study usually collects 
retrospective data and present status of respondents in a first wave and than continues data collection 
following up respondents for some amount of time." Compared to the panel design, the retrospective 
one-shot study avoids problems of panel mortality and changing measurement conditions (Featherman 
1979) but has to face an exacerbated problem of potential recall error.? Retrospective designs have 
been used by the John Hopkins Social Account Project (Blum et al. 1968) and various other studies 
in the U.S. (e.g. Freedman et al. 1988, Furstenberg et al. 1987). The data used in this paper come 
from the German Life History Study (Mayer & Brückner, E. 1989; Brückner, E. 1993; Brückner, H. 
& Mayer 1995). 
See Elder (1985) for an overview of the history of life course research. 
E.g. The Panel Study of Income Dynamics, Duncan & Morgan (1985); German Socio- 
Economic Panel, Krupp & Hanefeld (1987). 
For an entertaining but distressing account of failed retrospective designs see Bernard 
& al. (1984). A summary of recent research efforts on quality of retrospective data collection can 
be found in Dex (1991).

Note to user

Dear user,

In response to current developments in the web technology used by the Goobi viewer, the software no longer supports your browser.

Please use one of the following browsers to display this page correctly.

Thank you.

powered by Goobi viewer