|
Size: 1000
Comment:
|
Size: 1161
Comment:
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 8: | Line 8: |
| Content non-responsivity is "responding without regard to item content". It isn't a new problem, but it is a problem without a solution. | Content non-responsivity is "responding without regard to item content". It isn't a new problem, and it is a problem for all surveying methods. But it is a problem without a solution, and it is a larger problem for web surveys. Lack of control over environment leads to more non-responsivity. |
| Line 10: | Line 10: |
| One method of catching such errors is utilizing bogus responses--if a respondent gives a bogus responses, there is little doubt that they are non-responsive. For this to be effective, it should not be possible to straight-line and pass all bogus responses. | One method of catching non-responsivity is utilizing bogus responses--if a respondent gives a bogus responses, there is little doubt that they are non-responsive. For this to be effective, it should not be possible to straight-line and pass all bogus responses. |
Identifying Careless Responses in Survey Data
Identifying Careless Responses in Survey Data (https://doi.org/10.1037/a0028085) was written by Andrew Meade and S. Bartholomew Craig (both NC State University), published in Psychological Methods (APA PsycNET) in 2012.
Content Non-responsivity
Content non-responsivity is "responding without regard to item content". It isn't a new problem, and it is a problem for all surveying methods. But it is a problem without a solution, and it is a larger problem for web surveys. Lack of control over environment leads to more non-responsivity.
One method of catching non-responsivity is utilizing bogus responses--if a respondent gives a bogus responses, there is little doubt that they are non-responsive. For this to be effective, it should not be possible to straight-line and pass all bogus responses.
Consistency indices are also an option, but bake in assumptions about what non-responsive data looks like. It may not be logical for a respondent to give directly opposite responses (i.e., "antonyms") to similar prompts--but is it realistic?