I have treated lps differentiated and undifferentiated Thp1 cells with five different inhibitors. Now I want to know statistical significance of different treatments between lps differentiated and undifferentiated cells and also within the same group. Is it possible to compare with two way ANOVA or should I perform one way ANOVA for a single group (undifferentiated) then for second (differentiated).
Two-way ANOVA seems appropriate here. You'd be able to assess both factors (differentiated/undifferentiated and inhibitors) as well as their interaction - i.e. are the observed differences between inhibitors significantly different between differentiated and undifferentiated groups. Two separate 1-way ANOVAs wouldn't be able to look at differences between differentiated groups.
I agree with Joe. You need a "difference in difference" analysis, what is obtained via the interaction terms in a two-(or multiway) ANOVA.
Make sure your response is approximately normal distributed if you use ANOVA. For biological variables (often concentrations or alikes) you should consider analyzing the logarithms. A direct analysis would be possible using a corresponding generalized linear model based on the gamma distribution with log-link.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Structural Equation Modeling