Difference between revisions of "Power Calculations in Stata"

Jump to: navigation, search
Line 13: Line 13:
  
 
== Guidelines ==
 
== Guidelines ==
* organize information on the topic into subsections. for each subsection, include a brief description / overview, with links to articles that provide details
+
 
===Subsection 1===
+
=== Data needed to perform power calculations ===
 +
 
 +
You must have:
 +
* Mean and variance for outcome variable for your population 
 +
** Typically can assume mean and SD are the same for treatment and control groups if randomized
 +
 
 +
* Sample size (assuming you are calculating MDES (δ))
 +
** If individual randomization, number of people/units (n)
 +
** If clustered, number of clusters (k), number of units per cluster (m), intracluster correlation (ICC, ρ) and ideally, variation of cluster size
 +
 
 +
* The following standard conventions
 +
** Significance level (α) = 0.05
 +
** Power = 0.80 (i.e. probability of type II error (β) = 0.20
 +
 
 +
Ideally, you will also have:
 +
* Baseline correlation of outcome with covariates
 +
** Covariates (individual and/or cluster level) reduce the residual variance of the outcome variable, leading to lower required sample sizes
 +
*** Reducing individual level residual variance is akin to increasing # obs per cluster (bigger effect if ICC low)
 +
*** Reducing cluster level residual variance is akin to increasing # of clusters (bigger effect if ICC and m high)
 +
**If you have baseline data, this is easy to obtain
 +
*** Including baseline autocorrelation will improve power (keep only time invariant portion of variance)
 +
 
 +
* Number of follow-up surveys
 +
 
 +
* Autocorrelation of outcome between FUP rounds
 +
 
 +
 
 
===Subsection 2===
 
===Subsection 2===
 
===Subsection 3===
 
===Subsection 3===

Revision as of 18:20, 7 February 2017

NOTE: this article is only a template. Please add content!


add introductory 1-2 sentences here


Read First

  • include here key points you want to make sure all readers understand


Guidelines

Data needed to perform power calculations

You must have:

  • Mean and variance for outcome variable for your population
    • Typically can assume mean and SD are the same for treatment and control groups if randomized
  • Sample size (assuming you are calculating MDES (δ))
    • If individual randomization, number of people/units (n)
    • If clustered, number of clusters (k), number of units per cluster (m), intracluster correlation (ICC, ρ) and ideally, variation of cluster size
  • The following standard conventions
    • Significance level (α) = 0.05
    • Power = 0.80 (i.e. probability of type II error (β) = 0.20

Ideally, you will also have:

  • Baseline correlation of outcome with covariates
    • Covariates (individual and/or cluster level) reduce the residual variance of the outcome variable, leading to lower required sample sizes
      • Reducing individual level residual variance is akin to increasing # obs per cluster (bigger effect if ICC low)
      • Reducing cluster level residual variance is akin to increasing # of clusters (bigger effect if ICC and m high)
    • If you have baseline data, this is easy to obtain
      • Including baseline autocorrelation will improve power (keep only time invariant portion of variance)
  • Number of follow-up surveys
  • Autocorrelation of outcome between FUP rounds


Subsection 2

Subsection 3

Back to Parent

This article is part of the topic Sampling & Power Calculations


Additional Resources

  • list here other articles related to this topic, with a brief description and link