Online Workshop Announcement: Calculating Power and Sample Size
Need to do some sample size calculations?
Actually running sample size calculations isn’t terribly hard, if you understand the statistical analyses you’re running them for. Software is available that makes it pretty easy. No more power tables (remember those?). The hardest part is often setting them up.
The first part of setting them up is really, truly pinning down your research hypotheses. To do a sample size calculation, you need to know what analysis you’ll be doing. And to know what analysis to do, you have to know very specifically what your hypotheses are.
Then to fill in the numbers, you need some seemingly bizarre pieces of information. The problem is many are estimates (solid estimates at best, and wild guesses at worst), so you’re often making some big assumptions. As in, how correlated will the predictors be in my data? What will the standard deviation be?
Sample size software has gotten good enough that it allows you to estimate some of these pieces in different ways. For example, the differences among 4 means can be entered as the smallest difference, the average difference, or as a variance among all four.
But this also makes it more confusing. Which one should you use?
Once you understand the logic of sample size estimates, how these bizarre pieces fit together, and how to do the same 6 steps for every calculation, it all makes a lot more sense.
On Thursday, June 24th, I’m offering a 3-hour online workshop on Calculating Power and Sample Size. We’ll go over all this stuff–the logic, the info you need, where to get it, how to do the 6 steps, and how to use the specialized power software to get good estimates.
We’ll also go over what these estimates really tell you, and what they don’t. Power analyses are very useful, but they do have limitations. They’re only as good as the numbers they’re based on, and they’re not useful after the data are in.
We’re setting up registration a little differently for this workshop. Registration opens up on June 10, but until then, you can sign up for our Advance Discount Email List to get $20 off the regular registration price. If you do, you’ll get a special invitation and discount code on the 10th.
To get all the nitty-gritty details on the workshop and to get on the dicount list, go to the Calculating Power and Sample Size Workshop registration page.
