n****c 发帖数: 78 | 1 I have a data with only 8 patients, 16 knees (left and right). The score was
measured under 2 conditions (Intact and Deficient) for each knee of each
patient. Therefore, the total sample size is 32. I want to compare the score
between two different conditions. Due to the small sample size,I think non-
parametric test should be used. But I don't know which non-parametric test
shoud I use? Thanks
ID Knee Condition Score
1 R Intact 5
1 L Intact 7
1 R Deficient 8
1 |
f***a 发帖数: 329 | 2 So I think you are looking for a statistical test employed with two
independent samples. Depends what you want to test, there are several tests:
for testing if they have same mean, use t Test;
for testing if they have same median, use Mann-Whitney U Test;
for testing if they have same distribution, use Kolmogorov-Smirnov Test;
for testing equal variability, use Siegel-Tukey Test or Moses Test;
Hope this helps, cheers! |
f***a 发帖数: 329 | 3 Or just fit a LM or GLM to see what the coefficient of condition looks like? |
l*******l 发帖数: 204 | 4 However I would not think all obs in the data are independent. I am not sure
if there is non-parameter method for the correlated data. I would like to
know as well |
n****c 发帖数: 78 | 5 The scored between left and right knee are correlated within the same
subject, so two sample t-test is no appropriate. But my sample is so small 8
patients, 32 obs. Parametric method is ok? Thanks |
q********i 发帖数: 795 | 6 It is a clustered data, parametric mixed model should work, try
proc mixed;
class id knee condition;
model score = knee condition;
random id;
run; |
n****c 发帖数: 78 | |