# One-way or Two-way ANOVA?? - Please help with my stats! (Feb/13/2009 )

Hey everyone,

I wasnt sure where to put my question, but I hope someone will be able to help me!

I am studying a system whereby I treat cells with a stimulus and then investigate the effect of another drug on the stimulated cells - I have been using a one-way anova to study the effects, as I need to see if the stimulus is creating a significant difference to control cells and then I want to see if the drug is creating any other significant differences from stimulus only treated cells... (hope that makes sense!)

Does anyone know if its ok to use a one-way anova or should I be using two-way?

Any help would be much much appreciated!

Thanks!

Whereas one-way analysis of variance (ANOVA) tests measure significant effects of

one factor only, two-way analysis of variance (ANOVA) tests (also called two-factor

analysis of variance) measure the effects of two factors simultaneously. For

example, an experiment might be defined by two parameters, such as treatment and

time point. One-way ANOVA tests would be able to assess only the treatment effect

or the time effect. Two-way ANOVA on the other hand would not only be able to

assess both time and treatment in the same test, but also whether there is an

interaction between the parameters.

Hope that helps!

~Labrat

culturelady on Feb 13 2009, 12:06 PM said:

I wasnt sure where to put my question, but I hope someone will be able to help me!

I am studying a system whereby I treat cells with a stimulus and then investigate the effect of another drug on the stimulated cells - I have been using a one-way anova to study the effects, as I need to see if the stimulus is creating a significant difference to control cells and then I want to see if the drug is creating any other significant differences from stimulus only treated cells... (hope that makes sense!)

Does anyone know if its ok to use a one-way anova or should I be using two-way?

Any help would be much much appreciated!

Thanks!

Hopefully you used two controls: cells with stimulus without drug and cells without stimulus with drug.

You should use a two-way anova as you expect (or know) that the two treatments interact. You have to know btw if you have to use model I, II or a mixed model, depending if stimulus and drug are fixed or random effects.

Thanks for the help guys! Now all I need to do is figure out how to actually perform a 2-way anova in the stats package I use - (Graphpad prism). And thankfully I do have the relevant controls