Introduction
The following factors have been studied as potentially affecting survey response rates.
As to each factor there haye been various research findings as to whether there is a
demonstrated effect between that factor and a higher response rate.
Additional Contacts Before or After Survey
Advance letters, postcards, letters, follow-ups with copy of additional questionnaire and telephone calls are examples. Each contact increased the predicted response rate by 12% (Herberlein, p. 451). Preliminary letter no effect on rates for Parsons and Medford although it did serve to accelerate the daily rate of return. Another study showed an 8.1% increase in response rate (by Myers and Haug) but it was done at a cost of 22%. ( in excess of value according to researchers)
Follow-up techniques
Levine and Gordon achieved 100% response after 3 follow-ups-an advance letter was also sent. Single follow-up letters were effective in achieving 50% in a large mailing to a group ofWWII veterans. Kephart and Bressler found a preliminary letter to be very ineffective with follow-up very effective and the combination to be no more effective than the follow-up used alone.
Questionnaires
Length of Questionnaire
Common sense might suggest that shorter questionnaire would result in higher response rates because less time is demanded from respondent. The evidence from studies gives little support for that. Clausen and Ford did not find this to be true when they added one or two pages to a questionnaire already three to six pages in length. Length may signal impotance to the respondent. possibly even enough to overcome the costs associated with it. (Herberlein, p.459)
Salience
Questionnaires are more likely to be returned if they were judged to be salient to th respondent. (42% response rate versus a 77% return rate). Market joumal research
showed only 40% response while studies in scientific journals showed a 65% response and public health surveys showed an 81% response .(Heberlein, p. 451)
Sponsorship of Survey
Government sponsored research got higher responses independent of contacts and
salience. Many writers on mail surveys recommend use of official support-there is litle experimental evidence on effect of sponsorship. In one Study University sponsorship had large advantage over commercial sponsorship.
Population Being Surveyed
*A positive correlation between education and questionnaire response has been reported
*Students, employees and military personnel are more likely to return questionnaires
Incentives
One article summarizing research says that in all but 2 of 30 experiments reported,
incentives raised the response rate. How much money? One study showed that the
conditions in which at least $5 was sent led to a significantly higher likelihood of diaries being returned than no cash incentive. There was no significant difference in likelihood of return among the $6 conditions. In one study respondents who had previously agreed to complete and return the questionnaires had higher response rates than non-contacted and refusing households given $10.
*Immediate rewards were more effective for middle-class respondents anda promised reward was more effective for lower-class respondents.
*The resuts suggest that monetary or non-monetary incentives mailed with the survey instrument should provide improved return rates worth the investment of time and effortinvolved in their implementation. …The use of prepaid cash rewards for completing surveys had the most significant impact on increasing response rates among the observations in this meta-analysis.” (Church, 74-75) (non-monetary incentives included such things as entry in a lotery, donations to charity, coffee, books, pens, key rings, golf balls, tie clips, stamps and a turkey.”
*More studies need to be done to test the independent variables because there is missing data in the studies done thus far and other variables could have interacted with the incentive type to vary the result.
How Much Does The Incentive Need to Be?
*No consensus on what constitutes an optimal incentive amount after eight decades of research.
Study where incentives varied between 0 and $10:
*For households that had a positive outcome when they where contacted in the RDD sampling stage, the first dollar sent performed essentially as well as the second, third and fourth dollars sent. It was not until the $5 condition that response and cooperation were significanty higher than that $l and afler the $6; the $7- $10 conditions did not significantly enhance response rate or cooperation.
*For households never contacted during RDD stage (and presumably had neutral
Apenence with researcher and ones that refused to participate (a negative experience) Each incremental dollar essentially has a slightly larger impact on response and cooperation. (Trussell, 363)
Sources
Kanuk, Leslie and Conrad Berenson. 1975. “Mail surveys and response rates: a literature review.” Journal of Marketing Research l12: 440-453.
Herberlein, Thomas and Robert Baumgartner. 1978 Factors affecting response rates to mailed questionnaires: a quantitative analysis of the published literature.” American Sociological Review 43: 447- 462.
Trussell, Norm and Paul J. Lavrakas. 2004. The influence of incremental increases in token cash incentives on mail survey response: is there an optimal amount?” PUBLIC OPINION QUARTERLY 68 349-367.
Church, Allan H. 1993 Estimating the effect of incentives on mail survey response rates: a meta analysis.” PUBLIC OPINION QUARTERLY 57: 62-79.
Fox, Richard J. et al., “Mail survey response rates: a mneta-analysis of selected techniques for inducing responses.” PUBLIC OPINION QUARTERLY 52: 467-491.