mikebader.net - Entries for the tag research-designhttps://mikebader.net/blog/tags/research-design/The last entries tagged with research-designen-usZinniaThu, 15 Jun 2017 15:10:00 +0000Calculating Simple Power Analyses
https://mikebader.net/blog/entries/2010/10/18/calculating-simple-power-analyses/
<p>I am currently preparing a proposal for submission and one piece of information that the agency suggests is the power required to distinguish effects. This is obviously a perfectly reasonable piece of information to request; however, power calculations fall into that class of things that I know that I should know but I don't. It is one of those topics that every statistics book will tell you is important, but either a) glosses over the topic, or b) provides such a deep background that it is impossible to follow what the authors are talking about. Additionally, power calculations are complicated enormously by the fact that sample designs can become very complicated. </p>
<p>In contrast to this traditional treatment, <a href="http://www.stat.columbia.edu/~gelman/">Andrew Gelman</a> and <a href="http://steinhardt.nyu.edu/faculty_bios/view/Jennifer_Hill">Jennifer Hill's</a> book, <a href="http://www.powells.com/biblio/72-9780521686891-0">Data Analysis Using Regression and Multilevel/Hierarchical Models</a>, provides a very clear description of simple power analyses, which -- thankfully -- is all that I really need for this project. To make sure that I don't forget, I record below how to find the required sample size, <em>n</em>, for varying levels of between-group effect differences, Δ, at 80% power. The formula is relatively easy (see pp. 437-447 for more info): (5.6σ/Δ)<sup>2</sup>. Therefore, if I measure change in units of standard deviations, <code>sd</code>, then I can estimate the sample size <code>n</code> for each unit of change.</p>
<pre><code>drop _all
range sd 0 1 41
gen n = (5.6/sd)^2
</code></pre>
<p>I can then make a graph of the expected sample size required for a standard unit change using the command <code>twoway line n sd</code>; or, alternatively, just print a table of numbers using <code>list</code>.<br />
</p>
michaeldmbader@gmail.com (mikebader)Mon, 18 Oct 2010 18:31:54 +0000https://mikebader.net/blog/entries/2010/10/18/calculating-simple-power-analyses/Methods