Hi Caroline,
I performed principal components analysis on my baseline data in my
thesis as a way of reducing the number of outcome measures for the main
analyses. I put the factor analysis methods in the Methods chapter,
under a separate heading of 'Variable Reduction', where I described the
rationale for using the PCA rather than other FA methods, the principles
of rotation etc. in general terms. My first Results chapter was the
baseline comparison between the randomised groups; 2nd results chapter
was the factor analysis results, reporting the process of going through
5 solutions to find the best one; then the next 3 reported various
aspects of the longitudinal analysis. As I was using composite scores
derived from the PCA results as my main outcomes, I needed to include
the composites in the baseline comparison between randomised groups
(result ch. 1), but the composites weren't 'invented' until results
chapter 2. I just had to make reference to the 2nd results chapter to
say where the composites had come from.
As results chapter 2 was almost a stand-alone chapter, I considered
putting it in as an Appendix, but decided that would've devalued it as
it was a lot of work and was fairly central to the thesis itself. I also
considered having the FA follow Intro, Methods, Results in a separate
section of its own, but because I needed to describe the baseline data
collection that went into the FA, it didn't flow well. I think that
there is no 'right' answer, except to make sure your thesis flows and is
easy for your examiner to read, and printing out your Table of Contents
is a good way to see if this is the case.
Hope that helps,
Brian
On 26/04/2010 07:40, Caroline Wilson wrote:
> Specifically - where does the 'methods' chapter end and the 'results' chapter begin?
>
> If anyone has any thoughts or recent experience on this I'd be grateful.
>
> I'm aware there's a section called 'data analysis' in the final bit of the methods chapter. In here I was intending to put the pilot study results, the data screening and the rationale for types of statistical methods to use.
>
> In the results chapter I intend to put the 'practical' explanation of each statistical method and the results.
>
> My question is, does my exploratory factor analysis (which is quite long) go in the methods section where it logically sits as it was a means of data screening, or is it a result... in that it gave me some initial results that indicate the theories I was looking at were indeed separate entities?
>
> Or perhaps there is no 'right' answer?
>
> I've found Effective Writing in Psychology by Beins and Beins quite useful but it doesn't go to this level of answer. Thanks in anticipation.
>
> Caroline Wilson
> PhD Research Student
> Institute of Energy and Sustainable Development
> De Montfort University
> UK
>
>
|