idiots guide to statistics

[sawman]

For the last couple of days I have been wrestling with statistics terms, which as a stats numpty has proven frustrating to say the least.
Im hoping that someone here can point me in a good direction. I need to determine whether the data from a study is skewed or not, essentially 2 groups were followed, one had an intervention, one did not, whether result a or result b was recorded in both groups, there was no difference between the groups at the end of the study. the study used a non parametric analysis. I am trying to determine whether this was appropriate, it doesnt feel to be right, but I'm having trouble resolving why!

thanks for any help or pointers to a good resource

[chili1]

Im hoping that someone here can point me in a good direction. I need to determine whether the data from a study is skewed or not, essentially 2 groups were followed, one had an intervention, one did not, whether result a or result b was recorded in both groups, there was no difference between the groups at the end of the study. the study used a non parametric analysis. I am trying to determine whether this was appropriate, it doesnt feel to be right, but I'm having trouble resolving why!

Hi,

I work in statistics and I am RSS qualified (usually kills conversation at parties!). So hopefully, can point you in the right direction.
This -> http://www.statsoft.nl/uk/textbook/esc.html should help.
Basically need to look at significance testing, variable distribution, spearman rank coefficient correlation and obviously understand non parametric analysis. Non parametric analysis is less statistically powerful than parametric analysis. You would usually use NPA for smaller samples and PA for larger samples.

If you study two groups, assuming that the variable distribtution is recognised, I can't see why NPA is not appropriate.

I don't fully understand "intervention" - do you mean anomaly or outlier?

Lee.

[sawman]

Thanks lee, I'll go and read the stuff on that link

in terms of intervention, one group is doing something different to the other group - thats what I'm referring to

I 'll be back later once I have read some more

Cheers

S

[NWTony]

Some sort of Anova test?

[sawman]

Lee,

Thanks to your prompt and then what I found out, I have resolved my issues thanks.

I think I grasp the basics of that now- I am glad there are statisticians in the world as it means I dont have to worry about it too much.

cheers

s