Statistics
The old adage of “Lies, Damned Lies and Statistics” may be true, but I [like] to believe it has more to do with a lack of knowledge and competence than premeditation. Here is what I have learned so far in my quest to avoid spreading “damned lies”.
How politicians poisoned statistics
In this article, Tim Harford uses the distinction put forth by the Princeton philosopher Harry Frankfurt, between those who lie with statistics, and those who simply do not care what the facts are, but use statistics to support their position. The latter can be...
Type 1 and Type 2 errors… what they are..
Great with visuals to really understand a concept 🙂
Randomized control trial: The gold standard or overrated?
While randomized control trial (RCT) have long been seen as the epitome of evidence based research, having served to modernize the field of medicine, this article questions its use to test social policy. The argument is that where RCT's come into their own when...
How well do you see a correlation?
We have all read correlation matrixes, and have seen: As a rule of thumb, the following guidelines on strength of relationship are often useful (though many experts would somewhat disagree on the choice of boundaries). Value of r Strength of relationship 1.0 to -0.5...
Intro Statistics 9 Dance of the p Values
I use pictures from the ESCI software to give a brief, easy account of the Dance of the p Values. The simulation illustrates how enormously and disastrously variable the p value is, simply because of sampling variability. Never trust a p value! Use estimation, not NHST!
The dance of the p-values
Criticizing the use of p-values has begun to be so common place that it is getting dull repeating it. However, in his class on Moderation and Mediation today, Dominique Muller recommended a YouTube presentation called: “The dance of the P-values”. It puts the...
The rater in surveys as a source of bias.
As a method of collecting data, surveys have been widely criticized for the biases that are included in the data. When the source of biases are the same for both the dependent and the independent variable, (as in, what is used to explain, and what we are seeking to...
The link between “needed to treat”, “side effects” .. and how thinking about the correct statistics can be hard under stress..
This brilliant talk is on how to plan for stressful situations, so as to be best prepared to make the best decisions when the Coriolis in our brains are flowing freely; however, his example about how we evaluate medical risks under stress is brilliant, and worth...
The limitation of the P-value: Nature article: Scientific method: Statistical errors
Easy to read article in Nature covering the severe limitation of the P-value and explains why P-fishing is a dangerous road. The key take-away is that while the P-value can say something about the probability of the a specific null hypothesis in a given data; it does...
Want to catch up on statistics?
Statistics as a field is developing fast; and less you use it regularly, one tends to forget 🙂 Coursera offers a course to help you catch up, or learn for the first time. https://www.coursera.org/course/statistics