Class 6_Stat methods for analyzing non response

Question 1

Lessler and Kalsbeek (1992)

Answer 1

stochastic view
deterministic view
-Measuring non response bias- no good RR
Compared R and NR on aux. variables available for full sample that can be correlated with survey variables- differ could suggest bias
use response propensity as indicator of likelihood to responsd and correlate it with key variables

Question 2

Lessler and Kalsbeek (1992): Stochastic view

Answer 2

Decision to participate based on person specific prob of responding
can be impacted by various design decisions

Question 3

Lessler and Kalsbeek (1992): Deterministic view

Answer 3

population includes two types of people- R and NR

Question 4

Continuing of resistance idea

Answer 4

Do people who choose to participate later produce diff estimates

Question 5

Ways to reduce non response bias

Answer 5

Extrapolation
Nonrespondent sub sampling
Weighting adj

Question 6

Extrapolation

Answer 6

Use observable features of NR to predict what their survey responses would have been

Question 7

Non respondent sub sampling

Answer 7

Subsample of NR selected

alt design strategies to increase response

Question 8

Weighting adj

Answer 8

Use aux variable to form weighting classes(have to be related to key variables Y and response propensity)

Question 9

Bethlehem (2002)

Answer 9

Expands stochastic view
-Bias arises from covariance bw response propensity and key survey variables
Est. NR bias based on diff bw respondents and the pop on aux. variables

Question 10

GREG estimator requirement

Answer 10

requires good auxilliary info for the whole pop

Question 11

O’Muircheartaigh and Campanelli (1999)

Answer 11

example of est. int variance in area prob sample

effects of int on non contact and refusal

Question 12

O’Muircheartaigh and Campanelli (1999) Results

Answer 12

int vary of achieved RR
Urban areas have higher rates of NR
Int good at reducing whole HH R=reducing whole HH non contacts
between int variance larger than between area variance

Question 13

Schouten et al (2011)

Answer 13

role of paradata in studing NR bias
“representatives” indicator
measure degree to which respondent rep full sample

Question 14

Schouten et al (2011): Unconditional Partial R indicator

Answer 14

SD of the response propensities based on single variable Z- determine contribution of Z to the lack of rep

Question 15

Schouten et al (2011): Conditional Partial R indicator

Answer 15

SD of the response propensities based on all other aux variables besides Z -measure contribution of Z to lack of rep. while controlling for other aux. variables