Re: Is a 2-way ANOVA appropriate?

djaeger123@xxxxxxxxxxxxx wrote: 
Hi, I'm a 9th grader and I am conducting an experiment that has two
independent variables (Soil texture, and Soil pH).  My dependent
variable is the electrical conductivity of the soil. I was planning to
use a 2-way ANOVA for my statistical analysis and I was wondering if it
is okay to have an independent variable that is discontinuous (Soil
texture), and an independent variable that is continuous (Soil pH) and
still use a 2-way ANOVA.  Is there a more appropriate test I should
use?  Thank you.


Hi!

Understand that there's a whole bunch of data checking and analysis that go into the data before you get to the "two-factor analysis of variance", including whether "two-factor analysis of variance" is consistent with the way the study was actually performed. I mention this because I'm going to skip it and get straight to your question.

I'm not sure why soil texture is discontinuous unless it's qualitative (like "sandy", "rocky") but let's presume it is. It's also unclear whether pH is under your control, but let's presume it is.

ANOVA is reserved for those cases where predictors are qualitative, so, no, treating pH as a continuous variable means it is not a factor in the traditional sense.

So, what is it?  Here's one view http://www.tufts.edu/~gdallal/TERMS.HTM

But, what I think you may really be asking is what software you'll use to actually carry out the analysis. If your ANOVA software allows for covariates and for interactions between factors and covariates, you can use that. Otherwise, you'll probably have to "roll your own" getting a regression program to do it for you. The way things are coded is important, so you should be seeking assistance rather than plugging everything in and hoping for the best. Your analysis will include checking for whether the response is linear in pH. It might not be.

If for some strange reason, you are expected to do everything by hand, you might divide pH into a few categories and run a two-way ANOVA with the now two categorical predictors, a "poor man's approximation" to the more computationally complex analysis. However, it may do a fine job giving you a sense of what's going on in the data.

Good luck. Once you start, you'll undoubtedly have questions, so c'mon back and ask them.
.

Relevant Pages

  • Re: ANOVA QUESTION - Terminology
    ... I know two groups is t-test and more than two is ANOVA. ... But that is usually done in LINEAR MODELS after the introduction ... and Test# could serve as two independent variables, ... a repeated-measures analysis with several ...
    (sci.stat.consult)
  • Re: ANOVA QUESTION - Terminology
    ... I know two groups is t-test and more than two is ANOVA. ... Most examples of ANOVA are 3 groups of race ... and Test# could serve as two independent variables, ... a repeated-measures analysis with several ...
    (sci.stat.consult)
  • Re: ANOVA QUESTION - Terminology
    ... MEANS is called, strangely, the ANOVA. ... -- Reef Fish Bob. ... a repeated-measures analysis with several ...
    (sci.stat.consult)
  • Re: ANOVA QUESTION - Terminology
    ... Most examples of ANOVA are 3 groups of race ... and Test# could serve as two independent variables, ... measures design, probably because the procedure they use in their stats package is found under "GLM->Repeated Measures", or something similar. ... The interaction F-test is equivalent to an independent groups t-test on the change scores. ...
    (sci.stat.consult)
Loading