Full text: Wehner, Sigrid: Exploring trends and patterns of nonresponse

2. EXPLORATION OF THE NONRESPONDENTS IN THE EGLHS 
39 
instead of living with partners which was also asked). It is important to recognise, however, 
that all the listed variables of both studies were collected under exactly the same conditions: 
by the same survey institute and in the same interview situation; by the same wording of the 
questions; the same answer categories and rules for data edition. The fact that some of these 
topics cannot be compared is often a big drawback when a validation of survey data is 
intended by similar studies or by census data. 
Basic Group Comparisons 
The study inherent structure allows for several group comparisons: 
1) NRS persons versus all persons of the original 1991 sample. This is possible for variables 
related to 1991/92. 
2) NRS versus those persons of the initial sample who did not participate in the panel study 
(not-panel-persons). This is only possible for variables from 1991/92. 
3) NRS versus panel persons. This is possible for comparable variables collected in 1996/97 
and of course also for variables from 1991/92. 
Overview of the possible comparisons: 
10 
compare 
comparison time 
EGLHS wave 1 
nonresponse study 
all persons (N=2331) 
August 1991 
(N=201) 
EGLHS wave 1 
only not-panel-persons (N=951) 
August 1991 
EGLHS wave 2 
March 1996 
panel persons (N=1380) 
Working Hypotheses 
Exploratory work does not usually intend to test or to confirm hypotheses, but rather to "see 
into the data" (Ericksson/Nosanchuk, 1992). This is the first unprejudiced phase of data 
analysis. Adopting this approach, the following alternative hypotheses about the possible 
profile of the nonrespondents are intended to be working hypotheses. They shall act as a 
guideline for filtering out meaningful variables so that one can better distinguish between 
interesting and trivial information. 
Number of cases differ slightly from TABLE 1 (status of field report) due to subsequent data edition. 
5 Differing from the interview time, time-varying variables were taken for the earliest possible point in time 
available for all persons to be compared.
	        
Waiting...

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