Acceptable design effect

Home › Forums › Data Analysis and Reporting › Acceptable design effect

This topic contains 1 reply, has 2 voices, and was last updated by  SMART 10 years, 8 months ago.

Viewing 2 posts - 1 through 2 (of 2 total)
  • Author
    Posts
  • August 31, 2015 at 12:54 pm #1759

    Geoffrey Murithi
    Participant

    What is the acceptable range of design effect for W/H when analyzing and interpreting SMART survey results .

    September 2, 2015 at 9:35 pm #1775

    SMART
    Keymaster

    There is not a defined acceptable range of design effect for W/H. In a non-cluster simple/systematic random sample the design effect is always 1. In a cluster sample the design effect of W/H is typically between 1 and 2. When planning a survey if there is no information available for design effect the rule of thumb is to use 1.5. Kindly also refer to the following presentation: 3.SM.04 Sample Size Calculation (slides 17-23) in the Survey Planning Tools/ SMART Capacity Building Toolbox/ Survey Manager Section

  • Author
    Posts
Viewing 2 posts - 1 through 2 (of 2 total)

You must be logged in to reply to this topic.

Recent Replies