Survey Disposition
A survey disposition encodes the outcome of an interview.
Contents
Disposition Codes
The four overarching categories for interview outcomes are:
- Complete eligible repondents
- Eligible non-respondents
- Unknown eligible non-respondents
- Ineligible respondents
The codes then have subcategories, i.e. 1.1 for a truly complete respondent and 1.2 for a partial respondent with sufficient information. There may also be codes for completion by a proxy at 1.11 and 1.21. A disposition code can then be truncated at certain levels of precision to simplify categories, up to and including the collapse of all precise categories into the four overarchive categories.
AAPOR has published guidance to standardize the encoding of dispositions.
Web Surveys
The common codes for web surveys are:
Code |
Explanation |
1.1 |
Complete |
1.2 |
Partial with sufficient information (< NN% unanswered) |
2.111 |
Explicit refusal by other individual |
2.112 |
Explicit refusal by addressee |
2.12 |
Partial with insufficient information (> NN% unanswered) |
2.27 |
Too late response |
2.35 |
Technological issues |
3.19 |
Unknown eligibility (i.e. did not login) |
4.11 |
Screenout - Reason #1 |
4.12 |
Screenout - Reason #2 |
4.8 |
Quota Filled |
If respondents were invited as specifically-named individuals, non-respondents may be 2.113 (i.e. Implicit Refusal). Otherwise they are typically 3.19 (i.e. Did not login).
Paper Surveys
Paper surveys are in many ways simpler. Responses are generally higher quality. Whereas web interviews are often shown interactively, paper questionnaire are seen entirely and at once. Questions cannot be conditionally skipped on a paper questionnaire.
On the other hand, some information gathered from nixie codes ought to be incorporated.
Code |
Explanation |
1.1 |
Complete |
1.2 |
Partial with sufficient information (< NN% unanswered) |
2.111 |
Explicit refusal by other individual |
2.112 |
Explicit refusal by addressee |
2.12 |
Partial with insufficient information (> NN% unanswered) |
2.27 |
Too late response |
2.30 |
Deceased |
3.19 |
Unknown eligiblity (i.e. returned blank) |
3.31 |
Mail returned undelivered |
3.32 |
Mail refused by addressee |
4.11 |
Screenout - Reason #1 |
4.12 |
Screenout - Reason #2 |
If respondents were invited as specifically-named individuals, non-respondents may be 2.113 (i.e. Implicit Refusal). Otherwise they are typically 3.19 (i.e. Nothing returned).
Telephone Surveys
Multi-modal Surveys
A multi-modal survey has to incorporate components from each of the above.
A push-to-web survey, for example, typically can start with web survey dispositioning as a template. But nixie information reflected in paper survey dispositioning must be inserted.
Non-Response Bias
Non-response bias is introduced when an individual's likelihood to respond to a survey instrument is correlated to a key measurement.
Note that non-response bias is not the same as low response rates. Bias can be introduced if a subgroup's responsivity increases.