Sampling errors involved in incomplete returns to mail questionnaires.

Research on Mail Surveys

“…Many research articles on mail surveys have examined early versus late response bias. Most of them report a tendency towards earlier response by persons with an interest in the subject of the enquiry (only the first reference reports a within-wave bias): Baur (1947), Clausen and Ford (1947), Donald (1960), Filipello, Berg and Webb (1958), Larson and Catton (1959), Reid (1942), Rollins (1940), Shuttleworth (1941), Stanton (1939). The exception already noted is that of Gray (1957) and, of course, our own Survey E which is closely similar.…”

“…or over (Baur, 1947;Ferriss, 1951;Goldstein and Kroll, 1957;Gray, 1957;Gray and Parr, 1959;Ministry of Health, 1949;C. Scott, 1957;Scott and Jackson, 1960;Sheppard, 1959;Shuttleworth, 1941;Stanton, 1939;Suchman and McCandless, 1940;Tallent and Reiss, 1959;Toops, t It seems that during the ensuing 110 years or so the Church of England became gradually more cooperative: in 1949Lawson reported a response rate of 30 per cent. from English clergymen.…”

A Review of Some Problems Related to the Mail Questionnaire Technique

“…He felt that probably the primary reason for this was lack of interest; those principals interested in radio tended to answer and to return questionnaires on radio more readily. Similar results on an occupational survey were found by Shuttleworth (27). He sent questionnaires to 327 technology and chemistry majors graduated from the City College of New York in 1936.…”

Sociodemographic factors and mail survey response

“…Higher levels of intelligence/achievement are also associated with earlier response (Ellis, Endo, & Armer, 1970;Pavalko & Lutterman, 1973). Income, socioeconomic status, and employment status (employed vs. not employed) were found to relate positively to response rates (Jones, 1979;Ognibene, 1970;Pavalko & Lutterman, 1973;Shuttleworth, 1940;Suchman, 19621, response speed (Dalecki et al, 1988;Finn et al, 1983;Wellman et al, 19801, and item completion (Downs & Kerr, 1986). Landry et al (1988) found a nonsignificant effect of income on response rate, but the sample size was small ( n = 70) and considered only people over 65 years of age.…”