An Indicator of Nonresponse Bias Derived from Callback Analysis

1
An Indicator of Nonresponse Bias Derived from
Call-back Analysis
Paul P. Biemer RTI International and UNC
2
Outline

• Ignorable vs. non-ignorable nonresponse
• Bias in the nonresponse adjusted estimator
• Call-back model for estimating non-ignorable
  nonresponse
• Application for estimating drug use prevalence
• Future research

3
Estimation for Population Proportions

• Consider a SRS of size n
• Want to estimate some proportion,
• Let denote the observed dichotomous
  variable
• Let

4
Nonresponse Adjusted Estimator

• Estimator of is
• which is unbiased if nonresponse is ignorable
  w.r.t.
• i.e., if the error in is uncorrelated with
  

5
Bias in the Adjusted Estimator
if nonresponse is
ignorable
6
Call-back Model Analysis

• Goal is to estimate when nonresponse is
  non-ignorable
• Uses and call-back patterns to predict
  note, are only observed for respondents.
• For example, suppose
• Using data on call outcomes at each call-back for
  users and nonusers, we can estimate response
  propensity as a function of
• Then

7
Call-outcomes by LOE for Alcohol
Interviewed positives
Interviewed negatives
8
Call-outcomes by LOE for Marijuana
Interviewed negatives
Interviewed positives
9
Call-outcomes by LOE for Cocaine
Interviewed negatives
Interviewed positives
10
Call-back Notation

• 1 interview
• 2 non-interview
• 3 noncontact
• Call pattern 31111 gt noncontact followed by
  interview
• Once interviewed, stays interviewed (absorbing
  state)
• Once non-interviewed, stays non-interviewed
  (absorbing state)

11
Call-Back Data for LOE5
12
Simple Call-back Model for NI-NRLOE-5
Log-Likelihood
Likelihood of interview after l calls
Likelihood of no contact after 5 calls
Likelihood of non-interview after l calls

• Obtain parameter estimates by maximum likelihood

13
Simple LOE-5 Model Parameters
11 parameters and 10 degrees of
freedom Over-parameterized requires
constraints These constraints reduces parameters
to 7
14
Application Drug Use Survey

• Compared estimates of alcohol, marijuana and
  cocaine past year use prevalence for
• unadjusted
• current (traditional) adjustment
• call-back model adjustment
• Current adjustment incorporates 13 grouping
  variables and their interactions including a
  number of state specific components
• Call-back model incorporated call-back data (for
  up to 15 call-backs) and the drug use variable of
  interest

15
Estimated Response Propensities for Simple
LOE-15 Model
16
Prevalence Estimatesfor Simple LOE-15 Model
17
Future Work

• Test feasibility of incorporating call-back data
  in the nonresponse adjustment process
• Enter call-backs into the current logistic
  regression model (does not adjust for NI-NR)
• Apply the simple call-back model to the drug use
  data after traditional adjustment to provide
  second adjustment factor for NI-NR
• Use the simple call-back model to assess NI-NR
  bias following traditional adjustment approach