CAN YOU DO THIS INTERVIEW Methods for Determining if a Proxy Respondent is Needed

1
CAN YOU DO THIS INTERVIEW?Methods for
Determining if a Proxy Respondent is Needed

• Anne B. Ciemnecki, Kirsten Barrett,
• Karen A. CyBulski, Jason A. Markesich, Matt
  Sloan, and Debra L. Wright

2
Goals

• Contribute to body of best practices for
  interviewing people with disabilities
• Develop a proxy determination tool that is . . .
• Valid
• Reliable
• Objective
• Maximizes self-response
• Maximizes data quality
• Protects human subjects
• Does not create disparities
• Does not create ill will

3
Self Response

• Usually desirable
• Not always possible

4
Why Not?

• Physical or cognitive barriers on the part of the
  respondent
• Resource deficit on the part of the data
  collector

5
Alternatives

• Proxy response is appropriate when . . .
• - Sample member agrees to a proxy
• - Proxy is identified
• - Proxy is willing and able
• Non Response is appropriate when . . .
• - Proxy is not knowledgeable
• - Proxy cannot provide sampled respondents
  perspective

6
The Big Challenge

• Determining proxy need
• Balancing the right of self expression and
  data quality

7
Methods

• Unstructured
• - Gatekeepers or caregivers alert interviewer
• - Sample members alert interviewer
• - Sample members give cues during introduction
  or interview
• Structured
• - Formal assessments of sample members
  ability
• - Programmed checks for inconsistent or
  missing data

8
Structured Assessments Not Designed for this
Purpose

• Mini-Mental State Examination (MMSE)
• Telephone Interview of Cognitive Status (TICS)
• Short Portable Mental Status Questionnaire (SPMSQ)

9
Example of a Specially Designed Structured
Assessment

• Please tell me in your own words, what the survey
  is about?
• When I say your taking part is completely
  voluntary, what does that mean to you?
• When I say that your answers will be kept
  confidential, what does that mean to you?
• Two tries/asked of sample members and proxies

10
Advantages and Disadvantages of Interviewer
Judgment

Well trained Experienced Savvy
May not use the same standards
11
Advantages and Disadvantages of Structured
Assessment
Standardization Protection of human subjects

Begins interview with a test May not be related
to task at hand May produce false
negatives More stringent test for those with
disabilities
12
Structured and Unstructured Methods in an
Interview Context
Continue interview with proxy

• Confirm identity, introduce survey, determine
  accommodations

Conduct assessment
Continue interview without proxy
NO Interviewer judgment if assessment says need
proxy
Interviewer judgment to allow proxy
Interviewer judgment to allow proxy
13
Different Methods Produce Dramatic Differences
Proxy Use by Condition
Percentage
14
Proxy Use by Age of Sampled Person
Percentage
15
Proxy Use by Race and Ethnicity of Sampled Person
Percentage
16
Relationship of Proxy
Percentage
17
Conclusion

• Standardizing methods to identify when a proxy is
  needed is not easy
• Need to balance data quality with the right to
  self-direction and self expression
• Different methods produce different rates and
  patterns of proxy usage
• Depends on survey and population
• We have a lot of work to do