Hi everyone, I am looking for a bit of a push in the right direction.

I'm a BSc nutrition student at the moment, and although I did a module of stats in Maths A level, I'm totally stuck on this bit of stats for one of my modules at uni.

What I've done is an experiment to see whether vitamins have an effect on how much oxygen your body uses during 5 minutes of exercise, giving people a different vitamin each week.

To analyse the results, I got told to do a one way anova, which I managed no problem, followed by a Tukey test post hoc. There was a statistical difference in my one way anova. Now I managed entered the data ok for the Tukey test, and I got the attached data info out of minitab

What I dont know how to do is tell which week had a difference from these, and why the results show that it was that specific week. I have a feeling it was week 4 though!

Could anyone help me please?

You are correct...the way you tell with Minitab results is to look at the interval. If it doesnt contain zero then theres a statistically significant difference between the means. So 1 is significantly different from 4, 2 is significantly different from 4, 3 and 4, 4 and 5, and 4 and 6.

Hi everybody, I'm new here. I agree with kerri3278 (if you want to see it clearer, you can do analysis by Statgraphics software). But, the point I want to say here is that: you should do comparision between levels of Vitamins (different types of vitamins ----> the real variable of your study), not beetwen weeks (in here, week is not a factor). The name of factor appears in the Turkey test should be Vitamin (vitamin 1, ...n). Please discuss further if you don't agree.

Cheers,