LIST OF TABLES AND FIGURES
Tables
TABLE 1:
Sample Development in the East German Life History Study 20
2:
Reasons for Loss of Target Persons per Cohort - Initial Sampling 1991/92 26
TABLE
3:
TABLE
Reasons for Loss of Target Persons per Cohort - Wave 2 1996/97 28
Reasons for Loss of Target Persons per Cohort - Nonresponse Study 1996/97 30
TABLE 4:
TABLE 5:
Topics and Variables for NRS Data Examination 37
TABLE 6:
Working Hypotheses on Profile of Nonresponse Study 40
TABLE
Inspection of Significance Patterns - Chi-Square and t-Tests 41
7:
TABLE 8:
Logit Models Predicting Being in the Nonresponse Study for Cohort 1930 56
Predicted Probabilities for Nonresponse in Cohort 1930/Initial Sampling 199158
TABLE 9:
TABLE 10:
Predicted Probabilities for Nonresponse in Cohort 1930/ Wave 2 1996 59
TABLE
Logit Models Predicting Being in the Nonresponse Study for Cohort 1960 61
11:
Predicted Probabilities for Nonresponse in Cohort 1960/ Wave 1 199 62
TABLE
12:
TABLE 13:
Predicted Probabilities for Nonresponse in Cohort 1960/ Wave 2 1996 63
TABLE 14:
Logit-Model to Predict Selection for the Biased Samp 70
Linear Regression Model for Household Income 1996/Three Different Calculations. 71
TABLE 15:
Logit-Model to Predict Panel Participation against Nonresponse Study 75
TABLE 16:
TABLE 17:
Heckman Sample Selection Model Using Panel and Nonresponse Study 76
Figures
FIGURE 1:
The Sampling Process and Possible Erro
FIGURE 2:
Trend of Refusal Rates in German Survey
FIGURE 3:
Example of Nonresponse Bias in Linear Regression 13
FIGURE 4:
Sample Development in the East German Life History Study 23
FIGURE 5:
Cohort Composition of Wave 1 and 2 and Nonresponse Study 24
FIGURE 6:
Percentage of Successful Interviews after Low and High Number of Contacts 27
FIGURE
Reasons for Loss of Target Persons before Field Beginning3
7:
FIGURE
Reason for Systematic Loss after Field Beginning - Comparison of Cohorts per Study.. 31
8:
FIGURE 9:
Gender Proportions 32
FIGURE 10:
Gender Proportions per Cohort and Study 33
FIGURE 11:
Cohorts 1930 and 1960: Percentage of Men in EGLHS and Census Data 33
CAPICATI Differences for Percentages of Neutral Loss per Cohort and Study 34
FIGURE 12:
FIGURE 13:
CAPICATI Differences of Refusal Rates per Cohort and Study 35
FIGURE 14:
Percentage of Higher School Degrees in CATI and CAPI Interviews 36
FIGURE 15:
Percentage of Unmarried Persons 199 per Cohort and Gender44
FIGURE 16:
Percentage of Never Married Persons 1991 per Cohort and Gender 45
FIGURE 17:
Educational Degree in School for Cohort 1930 - Nonresponse versus Main Study 46
FIGURE 18:
Level of Vocational Training for Cohort 193047
FIGURE 19:
Average Number of Jobs until 1991 per Cohort and Gender 48
FIGURE 20:
Average Number of Jobs 1989-1996 per Cohort and Gender 49
Percentage of Unemployed Persons 1991 per Cohort and Gender 49
FIGURE 21:
Party Membership in the SED per Cohort and Gende 50
FIGURE 22:
Percentage of Persons Not Members of Any Party 1989 - Cohorts 1930 and 1960 51
FIGURE 23:
FIGURE 24:
Percentage of "Don't know" and "Refused" in the Party Attitude Scales 52
FIGURE 25:
Histogram of Household Income in the Original and in the Biased Sample 69
FIGURE 26:
Position of Estimated Regression Coefficients within Confidence Intervals 73
Bias Correction of Coefficient Estimate for Educational Variable 74
FIGURE 27:
FIGURE 28:
Pattern of Regression Results in Panel Study, Heckman Model, and Panel + Nonresponse......77
2