Hot answers tagged

13 votes

Sample Size impact on Effect size

When your sample size is small, there will be more sampling error attached to your effect size estimates and hence more uncertainty/wider confidence intervals. In that sense, I wouldn't say the effect ...
Christian Geiser's user avatar
10 votes

How does reducing my alpha-threshold to 0.005 affect my required sample size?

As you wrote ".. there is no "simple" function mapping the old size to the new one .." --this is what I expected. However, following your suggestion and then plotting new against ...
PuzzledBiologist's user avatar
10 votes
Accepted

How does reducing my alpha-threshold to 0.005 affect my required sample size?

There is very likely no "simple" function mapping the old sample size to the new one. The way to go about this would be to calculate the effect size detectable at the "old" sample ...
Stephan Kolassa's user avatar
9 votes

Sample Size impact on Effect size

As long as the sample is random, the estimate of the effect size is unbiased even in small samples. However, other things being equal, it will be less precise. It's still your best guess, but the ...
Peter Flom's user avatar
  • 117k
5 votes

Sample Size impact on Effect size

All things being equal, the major difference is the variation in the effect as a function of sample size. A very simple simulation in R shows this as the case. Here I simulate two groups who have the ...
Shawn Hemelstrand's user avatar
4 votes

Sample Size impact on Effect size

The question is fully addressed by meta-analysis (see, e.g., Introduction to Meta‐Analysis by Borenstein et al.), which includes studying of how the effect size vary from one study to another. Indeed, ...
Roger V.'s user avatar
  • 3,863
3 votes

Distinguishing the odd population out

Here is a maximum likelihood approach. For each label $i$, let $N_i$ be the number of observations with that label, let $L_i$ (the linear term) be the sum of those observations, and let $Q_i$ (the ...
Matt F.'s user avatar
  • 4,266
2 votes

Repeated measures ANOVA for case-control study

If I understood correctly, separate t-tests would not be a good idea. If possible, you want to include all or most data into the same model. Repeated-measures ANOVA could be fine, using time as a ...
Sointu's user avatar
  • 1,491
2 votes

Can ANOVAs or Student tests be conducted on any summary statistic at the individual level?

I wouldn't collapse any within-subject repeated measurements of RT into a mean or median, since they're correlated. The within-subject correlation among repeated measurements affects the overall ...
Leif Peterson's user avatar
2 votes

Distinguishing the odd population out

I have no direct solution to the problems below, but I believe that it important to mention, to warn that a straightforward answer is not as simple as it looks like and it should be nuanced that they ...
Sextus Empiricus's user avatar
2 votes
Accepted

Understanding relationship between T-Score and percentile

It looks like the test you are looking at is a psychometric test, which uses a different T-score than what you would get from a T-test. T-scores in psychometric testing is just a scaled z-score. The ...
RocketJoe's user avatar
2 votes
Accepted

Should I normalise my dependent variable by a measured variable that I want to be constant?

Biologically, I would argue that if the ratio on the pad is significantly different from 1, then I can't make valid conclusions from any of my data. That depends on what you are assuming about the ...
EdM's user avatar
  • 90.9k
2 votes

what book should I read to learn inferential statistics?

This question borderlines on an opinion question because it solicits opinions on texts that can be used to learn inferential statistics (which normally here would be off-topic). However, it seems the ...
Shawn Hemelstrand's user avatar
1 vote

How to compare semi-independent groups?

Unfortunately there is no test that will strictly apply. Your data do not satisfy the assumptions/requirements of either independent-groups tests or dependent-groups (paired-scores) tests. There is ...
rolando2's user avatar
  • 12.4k
1 vote

Examining group mean differences in time series using regression and HAC-robust standard errors

A survival analysis approach would be more interesting for these data. You could try a descriptive Kaplan-Meier with a log-rank test to compare the survival of both groups. Then you could try a ...
José Daniel Conejeros's user avatar
1 vote

How to test if the mean of a time series has significantly shifted at a certain point?

This is a well-studied problem. Given the vagaries of time series data, there is no one-size-fits-all technique. If the purpose is to test whether there is a changepoint at a KNOWN location/time, a ...
zhaokg's user avatar
  • 645

Only top scored, non community-wiki answers of a minimum length are eligible