Two Measures for Assisting Sample Size Discussions

• Activity Number: 582
• Type: Contributed
• Date/Time: Wednesday, August 1, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Beat Hulliger*+
• Companies: University of Northwestern Switzerland
• Keywords: Sampling ; Precision ; Relevance ; Size resolution ; Difference resolution

Abstract

Social surveys often must estimate the sizes or the proportions of many small groups and small differences between group sizes. The discussion of the needed precision of the estimators and the corresponding sample size is difficult since many different objectives must be considered and often lay persons are involved. The size resolution and the difference resolution of a sample are two measures which are derived from approximations to the probability of not observing a group or a difference in the sample. The size resolution is an operationalization of the smallest group which can be estimated from a sample. The difference resolution describes a minimal difference between two group sizes which can be estimated from a sample. Since these resolution measures embody elements of statistical hypothesis tests without the need of a complete test specification they orient the users to a reasonable sample size while remaining simple enough to assist the discussion with various stakeholders. The European Social Survey serves as an example of the application of the resolution measures.