[推荐][讨论] Factor Analysis Question

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

I was wondering if could ask some advice on analyzing the results of a factor analysis I performed. I am looking at a bunch of questions from a standard survey I run at my company. There seem to be one or two questions that could fall on to one or two factors. If I supress coefficients < .4 these two questions wouldn't load on any factors. I decreased it to surpress only coefficients < .3 and these two now show up on a number of different factors.

Taking one question for instance, it is originally grouped in with a bunch of other questions that seem to logically fit together. But its coeffiecient on this group is .365 and it loads on another factor at .370. My question is, is this close enough to where I can make a choice to keep it loaded on to its orginally grouped set of questions? Or since the coefficient is higher on the other factor that it needs to be grouped with that factor. I hope this makes sense.

Thanks,

Mike

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享1 收藏1 回帖

关键词：Analysis question Analysi factor alysis 推荐 Analysis 讨论 factor question

I am trying to analyze some survey data by using factor analysis to determine which groups of questions could be combined. Almost all of the questions are on a 1-5 Likert scale. Two of the questions are on a 1-4 scale. When I include these two using the 1-4 scale the way the variables load on to each of the factors makes sense. But when I exclude the two on the 1-4 scale I get 8 factors instead of 6. The additional factors make sense somewhat but they are not as clear.

My question is can you perform a reliable factor analysis with questions using slightly different likert scales? About 20 of the questions are on a 1-5 scale and the other two are on a 1-4 scale.

Thanks,

Mike

Michael:

There are two aspects to your question: one purely statistical and one
practical. It is customary to consider a loading of 0.3 as significant. If
you have a theory according to which variables should group together, and
one variable has a loading of .365 into a factor that groups variables
according to your theory, then you should be OK. The fact that the same
variable loads slightly higher into another factor does not disprove your
theory, but it uncovers the fact that your variable may be associated with
other variables as well. Practical significance should always take
precedence.

Dan

Hi Mike,

In my opinion it is OK to use 1-4 scales, especially if the solution
gives more sense. In FA, you analyse the correlations and not spreads of
the individual scales, and therefore the results are not too much
influenced by it.

Another possibility would be to fix the number of factor as 6. You have
the right to do it - the default Kaiser rule is only a rule of thumb.

Greetings

Unless there is a very strong reason, splitting items should be dropped.

Do you really need that item to be sure there are a reasonable number of
items on the scale?

What happens to the alpha-if item deleted in RELIABILITY if you check
out the whole sale.

When you say "originally grouped set" are you talking about previous
factor analytic results?

Art
Art@DrKendall.org
Social Research Consultants

Hello Jan

Do I have to take your remark that the results are *not too much* influenced
litteraly? If so, is there some kind of limit? Can I, for instance combine
1-4 scales with 1-10 scales ("scool notes" in some countries)?

Regards,
Antoon Smulders

Hello Antoon,

The methodology - e.g. the number of possible answers - always
influences the outcome of a survey in some way. What I mean is that
these influences are usually much less important than other sources of
errors in these correlation analyses (like misrepresentations in
sampling, misunderstandings of respondents etc.). So if the results seem
otherwise OK, you can IMHO accept them.

It is hard to say where lies the border between generosity and
carelessness. In principle, you can correlate every two variables
cardinal enough, and even binary dummies. But be careful and check the
output whether it gives sense.

Greetings

Jan

Dear stats experts:

My name is Cortney Warren and I am a doctoral student in Clinical Psychology at A&M, currently finishing my dissertation. I conducting multivariate analyses and trying to determine if there is a way to get group centroid values on a canonical variable/discriminant function (with 3 DV's) by group (Categorical IV: participant ethnicity (3 levels)) after adjusting for covariates (pre-test scores). In SPSS, the Discrimant procedure doesnt seem to allow for covariates and the manova/GLM procedure doesn't give adjusted centroid values (it will give group centroid values WITHOUT the covariate, but not WITH covariates as far as I can tell).

Does anyone know how to get group centroid values after controlling for a covariate using SPSS? If not, is there another program that will calculate them for me?

Thank you for your time,

Cortney S. Warren

Hi,

I am using two psychological well being tests. One test has 5 questions
(measured from 0 to 5) and other has 12 questions (measured from 0 to 3).

I have two objectives.
1) Compare the two tests to find similarity (so that if they are similar we
can use the test with 5 questions which will be easy to administered.

2) To get the factors (dimentions.) for both the tests.

Analysis:
Correlation of total score for both the tests and for all items.
Generally for categorical data Spearman's corre is used. But can we use
pearson also?

Can I use ROC to get sensitivity & specificity? (Limitation of the study is
that we do not have psychological diagnosis by doctors)
but
if one of the test is validated?
If both tests are not validated?

To get factors Principal Component analysis is better or Conformatory factor
analysis is better? How different are the two results?
Will it be wrong to do Principal component analysis? (exploratory)
Is there facility in SPSS to get Confirmatory factor analysis?
How can one use it?

[此贴子已经被作者于2006-4-30 10:47:11编辑过]

When you mention confirmatory factor analysis (CFA), do you already know
your factors? Because exploratory factor analysis does not require one to
know the number of factors, whereas CFA does.

In essence, you can use SPSS factor analysis procedures for CFA, but specify
the number of factors you need or expect. Then the procedure will either
confirm or refute your factor structure.

Regarding the variable type: when your variables are ordinal (typically the
case for survey responses), you can treat them as continuous and use the
procedure I just mentioned.

Dan