|
|
What analysis should I use?!
Pathological 0
Pathological/Reactive 0
Reactive 7
Reactive/Calculative 10
Calculative 10
Calculative/ Proactive 6
Proactive 9
Proactive/Generative 0
Generative 0
this is what I have. Pathological is no presence of
variable....Generative is lots of the variable. the numbers are how
many people thought that that much of the variable is present. I would
like to know if there is a significant difference between the number
of people who rated each variable. I guess... unless you see some more
valuable information that could be gleaned...
I am so lost!
Any help would be much appreciated
Bry
|
|
0
|
|
|
|
Reply
|
brionnyhooper (1)
|
11/30/2009 8:19:33 AM |
|
Please tell us more about you data. What is the subject matter area?
What is a case? What are the variables? Are there subsets of variables?
What do you mean by "variable" in your post?
Is it something like this?
Each case is a person who gave yes or no answers to 5 variables
(Pathological Reactive Calculative Proactive Generative) for
several stimuli. The stimuli were microscope slides and the 5 questions
were asked about different bacteria (stimuli).
In response to a stimulus a person "checked or not" all that apply.
for example with 4 stimuli and the 5 variables for (Pathological
Reactive Calculative Proactive Generative) there would be 20
variables in the data.
Art Kendall
Social Research Consultants
Brionny wrote:
> Pathological 0
> Pathological/Reactive 0
> Reactive 7
> Reactive/Calculative 10
> Calculative 10
> Calculative/ Proactive 6
> Proactive 9
> Proactive/Generative 0
> Generative 0
>
> this is what I have. Pathological is no presence of
> variable....Generative is lots of the variable. the numbers are how
> many people thought that that much of the variable is present. I would
> like to know if there is a significant difference between the number
> of people who rated each variable. I guess... unless you see some more
> valuable information that could be gleaned...
>
> I am so lost!
>
> Any help would be much appreciated
> Bry
>
|
|
0
|
|
|
|
Reply
|
Art
|
11/30/2009 12:35:37 PM
|
|
On Mon, 30 Nov 2009 00:19:33 -0800 (PST), Brionny
<brionnyhooper@gmail.com> wrote:
>Pathological 0
>Pathological/Reactive 0
>Reactive 7
>Reactive/Calculative 10
>Calculative 10
>Calculative/ Proactive 6
>Proactive 9
>Proactive/Generative 0
>Generative 0
>
>this is what I have. Pathological is no presence of
>variable....Generative is lots of the variable. the numbers are how
>many people thought that that much of the variable is present. I would
>like to know if there is a significant difference between the number
>of people who rated each variable. I guess... unless you see some more
>valuable information that could be gleaned...
>
>I am so lost!
>
>Any help would be much appreciated
This is not at all clear, and my guess is different from Art's.
It was helpful to me that Art suggested *some* context -
even if it is the wrong one.
I think that there is an "ordinal" scale represented by
Pathological, Reactive, Calculative, Proactive and Generative;
and there are 4 more possible "scores" available which consist
of half-way between each of the 5 specific descriptors.
That is what you mean by saying that the numbers show
"how much of the variable is present" -- Generative is "a lot"
and Pathological is "none".
In all, there are 42 scores or ratings, and not *variables*.
- I must be totally misconstruing everything, if (as Art seems
to be guessing) the OP actually does mean "variables" in the
sense that these are 9 different categories, each of which
might be independently marked as Yes/No
Now, the only hypothesis that I can figure out seems to
be the rather trivial one, "Is there a uniform distribution
across these levels?" Well, there are 0 scores in the 4
extreme categories, and 42 in the middle 5, so the answer
is no. So what? Should someone bother with a test?
Doesn't the distribution reflect, in some fashion, the manner
of selection of the sample?
I doubt that anyone will be interested the counts being
unequal, unless there is some important information added
from somewhere else.
--
Rich Ulrich
|
|
0
|
|
|
|
Reply
|
Rich
|
11/30/2009 7:36:29 PM
|
|
|
2 Replies
164 Views
(page loaded in 0.05 seconds)
Similiar Articles: dynamic factor analysis - comp.soft-sys.matlabHell to all, Could you please help me with DYNAMIC factor analysis with matlab?(if i have correctly understood the fuction "factoran" in matlab is ... IR-Spectra - comp.soft-sys.matlabI read somewhere the PCA (principal componet analysis) should here a possibility also without the statistical toolbox. Here my Problem: I have actually 60 FT-IT-spectra ... How to analysis high IOWait problem - comp.unix.solarisShould I assume Solaris 11 will have that hard-coded to zero too? A loose cough ... How to analysis high IOWait problem - comp.unix.solaris High load average / nothing ... Re: Help with panel data analysis - comp.soft-sys.sas--- On Wed, 12/9/09, Ryan wrote: > From: Ryan > Subject: Re: Help with panel data analysis > To: SAS-L@LISTSERV.UGA.EDU > Date: Wednesday, Dec... PCA Q statistic (SPE) Hotelling T2 - comp.soft-sys.matlab ...I want to use PCA (Principle component analysis) Q and T statistics test for fault detection control process. How can I use and how can I plot the Q o... Spectrum analysis (FFT) of partial waves / scilab - comp.dsp ...Hi, Could someone recommend the best spectrum analysis software package - or method - for dealing with signals having only partial waves? I've bee... Unit root test using adf - comp.soft-sys.matlabSince the time series has a non-zero mean, I use p=0, i.e., a constant term is ... much about what's under development here, but look for tools for cointegration analysis ... Should I use C++ or Java for Numeric Intensive Calculations - comp ...My primary numerical simulation and analysis work is in C++. I have ported about 5% ... MPI Program help - comp.parallel.mpi Should I use C++ or Java for Numeric Intensive ... Within-factors ANOVA - comp.soft-sys.stat.spssUsing the summary says that your file, going into analysis, can conveniently have just ... entry, assuming that you want all of the data in SPSS, I recommend that you use ... How to add radio buttons, etc. programatically? - comp.text.pdf ...> I only have the simpler PDFlib, should I get the fancier one? > Or should I use some ... Ask in some group for image analysis if they have more knowledge in this field and ... Statistical Tests in SPSS - UCLA Academic Technology ServicesWhat statistical analysis should I use? Statistical analyses using SPSS Introduction. This page shows how to perform a number of statistical tests using SPSS. What statistical analysis should I use? - Choosing the Correct ...What statistical analysis should I use? The following table shows general guidelines for choosing a statistical analysis. We emphasize that these are general ... 7/14/2012 3:02:11 AM
|
|
|
|
|
|
|
|
|