Osoby niedostępne dla wywiadów i miary stopnia reprezentatywności prób badawczych CBOS jako wskaźniki stosunku społeczeństwa do sondaży opinii publicznej

Abstract

In 1992 12 consecutive random samples of the Public Opinion Research Center (CBOS) have been analysed together with all the 1888 cases of avoidance and refusals to be interviewed. 14 various categories of refusals or other kinds of avoidance were distinguished as indicative of a negative attitude toward the acceptance of respondent’s role. Percentage points measuring over- and under-representation of sex, age, education level and place-of-residence categories in the samples of respondents were used as indicators of these categories’ willingness/ unwillingness to be interviewed. In all samples over-represented were women, retirement age and least educated persons and inhabitants of the countryside - which fact testifies their more positive attitude toward surveys and acceptance of respondent’s role. Influence of letters announcing the interviewer’s visit upon the degree of realization of planned address samples was ascertained by comparing refusal rates in 3 surveys using such letters with the remaining 9 ones without them: it turned out that in the former consistently lower percentages of refusals were found which shows that a researcher can create in potential respondents a more cooperative attitude.