2. EXPLORATION OF THE NONRESPONDENTS IN THE EGLHS 
32 
required to get interviews (lowest refusal rates, least number of contacts). I will name this 
cohort 1960: "the mobile generation". 
2.2.2 Gender Proportions 
Looking at the loss of people due to invalid addresses, the no-contact rates and the necessary 
number of contacts, wave 1 has a better sample participation for women. On the other hand, 
women tend to be more reluctant than men. The following results are reported for wave 1 
(see: infas (1995), overview 11)): 
women 
 
7,9% 
4.1% 
total neutral loss in wave 1 
4.1% 
category: "person unknown" 
2,0% 
no contact in wave 1 
3,9% 
5,3% 
refusal rate in wave 1 
38,7% 
32,2% 
The analysis of the number of contacts also revealed earlier success in interviewing women 
(see above the text following FIGURE 6). 
The next figure presents the proportion of men and women in the realised samples. 
FIGURE 9: 
Gender Proportions 
Nonresponse Study versus wave 1 and 2: Proportion of Men and 
Women 
6 34 
nonresponse: 
5E 
82 27. 
wave 2: 
47,3%— 
 
50, 97 
wave 1: 
49,17 
100% 
50% 
0% 
smen Qwomen 
We see that the men start to disappear in the panel study and even more so in the NRS file. In 
general the picture is the same when inspecting the gender proportion per cohort (see next 
figure). If we concentrate on both ends of the cohorts, there is a contrary result once again. 
Concerning the survey participation, the youngest people show virtually no gender differences 
with 46,5% men in the beginning, 43,8% in wave 2 and 45,5% in the NRS study. The oldest 
group has an exceptional position: wave 2 contains more men than the initial sample (52,2% in 
wave 2; 49,0% in wave 1) whereas the NRS file has lost men.