Table 3.
Decision Points for Constructing a Clinical Vignette
Consideration | Options | Advantages | Disadvantages |
---|---|---|---|
Question type | Open-ended | Avoids social desirability bias or bias from cueing | Time intensive and expensive to score7 |
Closed-ended | Easier and less expensive to score | Biased responses possible4,7,9 | |
Question format | Dichotomous | Easy to administer and interpret | Biased responses possible19 |
Multiple choice | Easy to administer and interpret | Biased responses possible | |
Likert scale | Familiar scale | Respondents may not interpret the scale in the same way20 | |
Fill in the blank with numeric response | Avoids bias from providing response options19 | More likely than other question formats to have errors occur in data entry of survey responses19 | |
Mode of administration | Hard copy | Easier than in-person interviews if respondents are geographically dispersed For vignettes on sensitive topics, less likely than interviews to elicit biased responses Likely least expensive of all modes of administration |
Greater respondent burden than interview or computer administration if vignette is open-ended |
Telephone/in-person interview | Telephone interviews are easier than in-person interviews if respondents are geographically dispersed May be designed to allow adaptation of vignettes to physicians’ responses |
In-person interviews are challenging and costly if respondents are geographically dispersed Possibility of biased responses, particularly for vignettes related to sensitive topics Likely most expensive of all modes of administration |
|
Computer or tablet | Easier than in-person interviews if respondents are geographically dispersed May be designed to allow adaptation of vignettes to physicians’ responses For vignettes on sensitive topics, less likely than interviews to elicit biased responses |
Not all respondents may be comfortable using computers Likely highly expensive |