Title: An Indicator of Nonresponse Bias Derived from Callback Analysis
1An Indicator of Nonresponse Bias Derived from
Call-back Analysis
Paul P. Biemer RTI International and UNC
2Outline
- 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
3Estimation for Population Proportions
- Consider a SRS of size n
- Want to estimate some proportion,
- Let denote the observed dichotomous
variable - Let
4Nonresponse Adjusted Estimator
- Estimator of is
- which is unbiased if nonresponse is ignorable
w.r.t. - i.e., if the error in is uncorrelated with
5Bias in the Adjusted Estimator
if nonresponse is
ignorable
6Call-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
7Call-outcomes by LOE for Alcohol
Interviewed positives
Interviewed negatives
8Call-outcomes by LOE for Marijuana
Interviewed negatives
Interviewed positives
9Call-outcomes by LOE for Cocaine
Interviewed negatives
Interviewed positives
10Call-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)
11Call-Back Data for LOE5
12Simple 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
13Simple LOE-5 Model Parameters
11 parameters and 10 degrees of
freedom Over-parameterized requires
constraints These constraints reduces parameters
to 7
14Application 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
15Estimated Response Propensities for Simple
LOE-15 Model
16Prevalence Estimatesfor Simple LOE-15 Model
17Future 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