CFA or EFA

posted
11-Jan-19, 13:05
Avatar for Tudor_Queen
posted about 6 months ago
Is there a simple way to easily tell if a study is using EFA vs CFA? I don't use either of these techniques but need to assess methodological quality on a few papers. So I've read up on what they are, and why they are used. Problem is, papers say things like "factor analysis was run" but don't always explicitly say whether it was confirmatory or exploratory. I know the general background of what each method does - ie. to test a hypothesised structure (CFA) or to come up with a structure (EFA). But this isn't really helping, as it seems as though it could be either (i.e., testing or coming up with a structure for a scale they are interested in)!
Cheers
posted
11-Jan-19, 16:57
edited about 29 seconds later
by pm133
Avatar for pm133
posted about 6 months ago
The quickest way to resolve this is to pick up the phone and call the corresponding author if you can find a number for them.

I don't know anything about these sorts of things but is there a difference in the way both techniques are described in papers where those terms are used? If so, can you correlate with the way the papers on your desk describe their methods?
posted
12-Jan-19, 20:58
edited about 33 seconds later
Avatar for Tudor_Queen
posted about 6 months ago
Thanks pm133 - I am doing that latter thing you suggest but still struggling greatly. I am not familiar with these methods at all. It is a new RA job, and nowhere on the job spec did it say I would need to know/have experience in these methods. And since this is a big part of the role, I am pretty annoyed. I'm trying to learn of course but somehow I don't think I am in the best place to review papers on such techniques when I have never used them myself. Just reading about it and trying to understand doesn't seem to help. I think that I would need some expertise in the methodology to be able to review how others have used it. The job has turned out to be a nightmare! Sorry this has turned into a rant. Nice to correspond with you again after a while!
posted
13-Jan-19, 02:46
edited about 52 seconds later
by pm133
Avatar for pm133
posted about 6 months ago
If you are peer reviewing papers which use these techniques as a foundation, and you don't know the techniques well enough to follow what they've done, then I think you have no option but to pass on this instance. You have to be honest with yourself about whether you think you can realistically give these papers a fair review. either positively or negatively.

If I remember, you are also writing up your thesis whilst doing this job full time so having to struggle with new techniques is presumably making a difficult situation considerably worse for you. I honestly don't know how you do it. That set of circumstances would have finished me off I think.

Nice to talk to you too. Sorry I don't have anything more constructive to offer.
posted
13-Jan-19, 14:25
edited about 4 minutes later
Avatar for Tudor_Queen
posted about 6 months ago
Yeh, I feel pretty much the same way about it... and quite exasperated. It's for a systematic review. After trying to get my head around it for a few weeks, I was pretty straight with the manager about the lack of progress I'd made and why (lack of experience/skills) and that in order for me to make progress I think I need direct guidance from someone. But alas - nothing. I'll see how I get on next week and then raise it again and simply say I need to do something else if direct guidance isn't available. It's not a comfortable situation to be in - being paid and yet not producing tangible results. On the plus side - it's all experience (yeh right!) and it makes finishing my thesis seem like a piece of cake (ie. at least I can understand it)!
posted
13-Jan-19, 16:21
edited about 29 seconds later
by pm133
Avatar for pm133
posted about 6 months ago
I was in a similar situation when I was younger. I was put into a job which I had no understanding of and was given the task of improving what a specific group of the workforce were doing.
I was given a desk in an office on my own and left there with 3 sheets of paper full of fairly abstract ideas about the job. No computer. No colleagues. Just me.
After about a few weeks I walked into my bosses office and told him I wanted a transfer to another job.

They eventually got rid of me but that's a story for another day :-D
posted
14-Jan-19, 09:53
edited about 4 minutes later
Avatar for Tudor_Queen
posted about 6 months ago
Ha! It sounds rather similar to that! I think I would have left (and perhaps they would have got rid of me) if it were not for the fact that a) the other part of the job should be more up my street, and b) they have paid for me to undergo training for it (the other part of the job that is).

It's awkward and annoying though... I can't stand being unproductive and feeling like I was paid for nothing at the end of the day. Going to have another crack at understanding it now. I wish me luck!
posted
16-Jan-19, 19:09
by agrew
Avatar for agrew
posted about 6 months ago
It's not really clear for me from the post what specifically you're struggling with. There are plenty of papers that provide overviews for both methods, just to name one:

Worthington and Whittaker (2009) Scale Development Research A Content Analysis and Recommendations for Best Practices

It's slightly outdated but the core principals are still the same. And because the paper focuses on the previously published papers that employed EFA and/or CFA you should get an idea what generally is reported when the analysis is carried out.

Postgraduate
Forum

Copyright ©2018
All rights reserved

Postgraduate Forum

Masters Degrees

PhD Opportunities

PostgraduateForum is a trading name of FindAUniversity Ltd
FindAUniversity Ltd, 77 Sidney St, Sheffield, S1 4RG, UK. Tel +44 (0) 114 268 4940 Fax: +44 (0) 114 268 5766