Objectives
The "Education, Training, and Occupation: Life Courses of the 1964 and 1971 Birth Cohorts in West Germany" project extended earlier surveys by tracing the life histories of the 1964 and 1971 cohorts up to the end of the 1990s.

Given the scarcity of training places and difficult labor market situation in the 1980s and 1990s, it was assumed that members of the 1964 cohort, the cohort with the highest birthrate since the founding of the Federal Republic of Germany, would have experienced unprecedented intra-cohort competition in the educational and occupational systems. The 1971 cohort was selected to allow direct comparison with the East German cohort surveyed in the LV-Ost 71 study.

Questionnaire Design
Based on the previous studies, a questionnaire was developed to assess data in the following domains:

- General personal characteristics/attitudes
- Schooling and vocational/professional training
- Spells of employment/non-employment
- Family
- Household/residential

Detailed Overview (in German, pdf format, 161 KB)

One special feature of the "Education, Training, and Occupation: Life Courses of the 1964 and 1971 Birth Cohorts in West Germany" project was that respondents were asked whether they would consent to their life history data being linked up with the social security and employment data held on them by the Institute for Employment Research (IAB).

Questionnaire (in German, pdf format , 551 KB)
List of Variables (in German, pdf format , 706 KB)

Codebook in English

Sample Design
Between June and December 1997, a sample of the German-speaking resident population belonging to the 1964 and 1971 birth cohorts was drawn from the official registers of 97 selected communities of the former West Germany, stratified by federal state, type of region, and cohort. This was the first time that German-speaking non-nationals were included in a sample drawn for the GLHS.

Source Data ( in German, pdf format , 126 KB)

Data Collection
Following a methods test and pretests, the main study ran from May 1998 to January 1999. Data collection was computer-assisted, with a first round of automated data checking taking place during the interviews themselves. Overall, 2,685 respondents were interviewed by telephone and 250 in person by a total of 53 interviewers. In September 1998, a non-response study was initiated to increase sample coverage.

	Birth Cohort		Total
	1964	1971	N	%
Gross sample	3,299	3,220	6,519	100.0
Neutral non-response	968	1,147	2,115	32.4
Adjusted sample	2,331	2,073	4,404	100.0
No contact made to household or target person	64	61	125	2.8
Ill	710	520	1,230	30.5
Refusals	58	42	100	2.3
Realized interviews	1,499	1,450	2,949	67.0
Unusable interviews	25	15	40	32.0
Unable interviews/coverage rate	1,474	1,435	2,909	66.1

With a (net) coverage rate of 66%, the study exceeded the quotas of comparable surveys. Furthermore, most of the gaps in the data could be filled by re-contacting the respondents.

Representativeness
To gain a first indication of the representativeness of the sample, selected, age-specific distributions of the life history study were compared with those of various cohorts in the microcensus data (see Hillmert/Kröhnert 2001). Comparison of these distributions reveals some typical deviations between microcensus respondents and life history study respondents. For example, people with little or no education and training are underrepresented in the life history data, whereas those with intermediate levels of education are somewhat overrepresented. The unemployment rates computed for the respondents in the life history study are also consistently lower.

Method Report by infas (in German, pdf-Format, 69 KB)

Data Editing
Once the data had been collected, they underwent thorough editing, with each individual case being checked for missing or implausible data (see Hillmert, 2002a, for more information on the editing process and implications for analyses of content). The preliminary round of editing was intitated in July 1998. Data editing proper was begun at the Max Planck Institute for Human Development in early January 1999 (see Editionshandbuch, pdf format, 822 KB). A comprehensive set of rules was drawn up for the editing process and continuously optimized. Some 1,000 respondents were re-contacted by telephone and asked to clarify or add to the data they had provided (see Nachrecherchebericht, pdf format, 69 KB). Including re-contacting respondents to clarify any ambiguities and contradictions, entry of corrected data, and data checking, the process lasted until the end of 2002.

Notes on Data Analysis
One of the strengths of this study is the exceptional quality of the data it provides – the product of meticulous preparation, data collection, and data editing. Another is the detailed nature of the responses, which allow longitudinal analyses to be conducted for numerous variables. The key features of relevance to the life course were surveyed separately for each (sub-)episode reported, and the modular structure of the survey made it possible to establish cross-references between parallel activities and different domains of life. Finally, it was also possible to link up the individual-level data with process-produced data (Bender et al. 2001), enabling us to validate the individual-level data, on the one hand, and supplement the data with further information, on the other. This possibility is not available to external users, however.

Documentation (Downloads, in German)