It may seem complicated for the respondents to evaluate the situations by comparing list of about
thirteen numbers each. One has to keep in mind, however, that before expressing their opinion on the
situations, people answered several other questions, which involved one specific part of the list of
numbers, e.g. their opinion on the contribution rate or on the rate of return for the old generation.
Furthermore, at the end of the questionnaire, the respondent has been provided with a summary of the
situations and one’s own grades assigned to these situations. The respondent could change the grades
if he wanted to. Therefore, we think that there is no reason to believe that the respondents were not
able to make a serious comparison.
4 Empirical results
4.1 Data and specification
In total, the sample contains 1103 respondents. Per generation, the numbers are as follows: 278 old
respondents, 362 middle-aged and 463 young respondents. Due to several reasons4, we are left with
988 respondents. The overall response rate including technical non-response is 70%5. The overall
response rate corrected for this technical non-response amounts to 83%, which is rather high. Table
4.1 shows the number of respondents involved in each situation.
4Some respondents did not receive all questions, some of them did not answer all questions and others
obviously gave incorrect answers etc.
5Technical non-response refers to non-response due to technical problems or due to holidays.