# Thread: What kind of ANOVA is good for this

1. ## What kind of ANOVA is good for this

Hello,

In my research i've collected death % for cell cultures. One variable was the treatment used (0,2,4,10,16 exposure). Another one was the day of recording the result.(day 1,3,5,7).

I was wondering which ANOVA or other tests fit here. I did use a 2 way anova, since they are 2 independent variables, but i'm not sure if that's right.

I was wondering if i can used repeated measurements ANOVA, since i my opinion ( and i have no real idea) i thought i could treat the different cultures, which should be normalized with their high cell count, as the same culture since they all underwent the same procedures.

2. ## Re: What kind of ANOVA is good for this

Originally Posted by Blakey85
I did use a 2 way anova, since they are 2 independent variables, but i'm not sure if that's right.
That's correct.

Originally Posted by Blakey85
I was wondering if i can used repeated measurements ANOVA
That's also correct.

You want a 2 way (a.k.a. factorial) ANOVA, with one between-subjects factor (exposure) and one within-subjects or repeated-measures factor (day of recording). This is called a mixed anova.

Here's a nice explanation from Andy Field: http://www.statisticshell.com/docs/mixed.pdf

3. ## Re: What kind of ANOVA is good for this

I think something may have been lost in translation: In this i have 20 cultures. All of them were treated at day 0 different treatments (0,2,4,10,16 minutes exposure), 4 in each treatment.
The at day 1 i take 4 cultures and fixate them. Counting the dead cell %. Same with days 3,5,7.

Now the aim was to show the acute and chronic effects of the different treatments.
In this case, can i use repeated ANOVA? the cultures are from the same treatment group, but are not the same exact cultures in different times ( although with the thousands of cells in a plate, i would assume any kind of variance gets normalized)

Thank you again.

4. ## Re: What kind of ANOVA is good for this

What do you think was lost in translation?

If you have two variables, it doesn't make sense to ask if you can use repeated ANOVA. That describes a certain kind of one-factor ANOVA. However you have two variables (day of recording, and treatment). Unless I'm still misunderstanding something, I'm still gonna go with mixed ANOVA.

As to your question about there being a few (it sounds like 4) cell cultures per treatment condition, do you have any reason to believe that there WILL be major variability between different cell cultures within the same treatment condition? Because I don't know a lot about cell culture experiments but it sounds like the cell cultures are your cases. With only four cases per condition your experiment won't have a lot of power. So if there is any major variability between the cultures within a single condition, you won't get significant results.

(I'm getting the number four from the fact that you said 20 cell cultures and listed 5 treatment conditions, so correct me if I'm wrong but it sounds like 4 cultures per treatment condition.)

 Tweet

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts