Analysing Survey Data
Created @January 22, 2023 3:26 PM
Class RW: Survey
Type Reading
Stage? Drafting
Reviewed
Column
Materials
Reading 1 Chapter 10 (Fowler)
Reading 2
Once data have been collected and a data file created, the next step is to analyse the
data to make statistical estimates and reach conclusions.
Adjusting for Sample Nonresponse and Sample Frame
Deficiencies
Nonresponse and an Imperfect Sample Frame are not uncommon in surveys.
Of concern, of course, is the extent to which those not responding or who never had
a chance to be selected are different from those who do respond with respect to
variables the survey attempts to estimate.
If there is information available about nonrespondents’ values, then corrections in
the survey estimates can be made directly. However, that is seldom the case, as
there would be little need for conducting a special-purpose survey if the information
was already available from other sources.
A more common situation is that some information is available about non-
respondents, but not about the key variables being estimated.
Analysing Survey Data 1
Created @January 22, 2023 3:26 PM
Class RW: Survey
Type Reading
Stage? Drafting
Reviewed
Column
Materials
Reading 1 Chapter 10 (Fowler)
Reading 2
Once data have been collected and a data file created, the next step is to analyse the
data to make statistical estimates and reach conclusions.
Adjusting for Sample Nonresponse and Sample Frame
Deficiencies
Nonresponse and an Imperfect Sample Frame are not uncommon in surveys.
Of concern, of course, is the extent to which those not responding or who never had
a chance to be selected are different from those who do respond with respect to
variables the survey attempts to estimate.
If there is information available about nonrespondents’ values, then corrections in
the survey estimates can be made directly. However, that is seldom the case, as
there would be little need for conducting a special-purpose survey if the information
was already available from other sources.
A more common situation is that some information is available about non-
respondents, but not about the key variables being estimated.
Analysing Survey Data 1
