Een vertekend beeld van de maatschappij? Een onderzoek naar selectieve uitval in een telefonische enquête ten gevolge van onderdekking en nonrespons

Samenvatting

We studied undercoverage and nonresponse in a telephone questionnaire survey (original N = 7,000) in Groningen, Netherlands. Bivariate and multivariate analyses showed that the undercoverage was strongly related to all sociodemographic variables studied, except sex. Nonresponse was related to age, country of origin, marital status, and household size and composition, but not to sex, unemployment, dependence on social assistance, and education (only known for the unemployed). Both undercoverage and nonresponse contributed to a strong middle-class bias in the final data set: middle-aged and older economically secure persons of Dutch origin living with a partner are overrepresented, while persons of lower social status, the young, people of foreign stock, the unemployed, persons living on welfare, and singles are underrepresented. 4 Tables, 1 Figure. Adapted from the source document.