Propensity scores represent an intriguing alternative method to reduce the impact of confounding variables, especially when there are multiple potential confounding variables. This paper considers a range of models comparing the propensity score approach to the more traditional approaches of adjusting for confounders. I think the conclusions are overly simplistic, but the paper is still worth reading. Continue reading

# Monthly Archives: December 2013

# PMean: Sample size for comparing two diagnostic tests

I had a client come in with a sample size question. It involved the comparison of two diagnostic tests to a gold standard. There a couple of different ways to attack the problem. Continue reading

# Recommended: Study Development (Kansas City area resource)

This is an excellent local resource for those of you in the Kansas City area. They have guidance on coming up with a study idea, finding funding, and identifying study participants. This site offers general information and links to resources with Frontiers (aka the Heartland Institute for Clinical and Translational Research). Continue reading

# Recommended: DataMind website

DataMind is a website offering free tutorials for the R programming language and tools that let you develop your own tutorials based on R Markdown. I have not tested any of this, but it looks interesting. Continue reading

# PMean: Claims lacking specificity are meaningless

I recently encountered a claim about the unlimited applications for a new statistical model. I, for one, tend to view “unlimited applications” as a negative comment rather than a positive comment. Any new model that pretends to be applicable in all areas is probably applicable in no areas. Here’s why. Continue reading

# Recommended: Use = or <- for assignment?

I learned R back when the only way to assign a value was with the <- operator. For example, if you were computing the hypotenuse of a triangle, you’d use c <- sqrt(a^2+b^2). But the language now allows you to use an equals sign instead, which is the choice in many other programming languages. Should you do this? I say “no” emphatically, but this website makes an interesting counter-argument. Continue reading

# PMean: Is Possibility Theory better than Probability Theory?

Someone on a LinkedIn group posted a question about “Possibility Theory.” The question itself had a lot of hype, claiming that “time has expired for Probability Theory.” Still, it is an interesting question and here’s how I responded. Continue reading

# Recommended: “Any other comments?” Open questions on questionnaires – a bane or a bonus to research?

This is a classic reference that is worth re-posting. No one seems to know what to do about those pesky open-ended questions you see on a survey. This article offers practical tips on how to handle this type of data. Continue reading

# PMean: The cost of a bad prediction

Paul Krugman wrote up an interesting application of Bayes Theorem on his blog on the New York Times. I want to adapt his example and expand it a bit. Continue reading

# Recommended: How confidence intervals become confusion intervals.

The desire among researchers for a black and white dichotomy between “significant” and “not significant” results leads to a lot of unnecessary confusion. Continue reading