Activity Number:
|
284
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, August 4, 2009 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Section on Survey Research Methods
|
Abstract - #304343 |
Title:
|
Alphabetical Placement in Surveys of Persons at Institutions: A Simulation Study
|
Author(s):
|
Pedro J. Saavedra and Francine Barrington*+
|
Companies:
|
ICF Macro and ICF Macro
|
Address:
|
, , 20705,
|
Keywords:
|
alphabetic sampling ; bias ; establishment ; cluster ; simple random sampling ; simulation
|
Abstract:
|
It is often the case that a survey of persons at institutions (e.g. teachers at schools or staff at offices) requires that one or more persons be selected from each institution by an administrator at the institution. A common way of doing this is to ask that a list of eligible potential respondents be drawn and that a set of random numbers provided to the administrator be used to draw the sample. Experience shows that this task is both difficult and burdensome. An alternate approach when the number to be selected from each institution is small is to provide a list of names (first and last) and ask that the person who follows each alphabetically be selected. This method has several potential biases when the population as a whole or in some institutions differs from the one from which names were drawn. These biases are explored here through simulations.
|