Dear List,
I have posted this query to the SEMNET list, but I'm still stuck and I am hoping that others on this list might have a suggestion. I have a 80-item questionnaire that I want to factor analyse, but I can't figure out how to draw a diagram with so many items in AMOS graphics. It seems impossible as the indicators fall off the edge of the page and I can’t then see them. I’ve tried selecting “resize diagram to fit page”, but this makes it too tiny for me to then drag the SPSS variables into each indicator box. Someone suggested using AMOS basic with text input, but I don't seem to have this option anywhere in the AMOS progam. Surely there must be a way of doing this! Many thanks, Kathryn |
Administrator
|
First of all it is obvious you are attempting an EXPLORATORY factor analysis (EFA) rather than a CONFIRMATORY factor analysis (CFA). AMOS is *NOT* an appropriate tool to use for performing EFA.
The fact that you have 80 items is the give away. Amos Basic is the programming language for Amos and you will have to RTFM and learn how to program to use it. With 80 items you should probably have a few thousand cases. ---
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me. --- "Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis." Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?" |
In reply to this post by Kathryn Gardner
You do not have to drag the variables. You can double click on the indicator boxes and type in the variable names.
On Tue, Apr 24, 2012 at 4:53 AM, Kathryn Gardner [via SPSSX Discussion] <[hidden email]> wrote:
-- Lori Andersen Ph.D. student, Educational Policy, Planning & Leadership College of William & Mary Williamsburg, VA |
In reply to this post by Kathryn Gardner
Please explain the purpose of your factor analysis.
Do you have items that were designed to measure some specific constructs? I.e., do you have a preliminary scoring key? what are the constructs? Are you trying to double check the scoring key from previous work? Do you wish to end up with a scoring key using one/zero weights for the items? Or do you want some other form of factor scores. What do you intend to do with the scores? How many cases do you have? Did you already run a parallel analysis in base SPSS? Art Kendall Social Research Consultants On 4/24/2012 4:41 AM, Kathryn Gardner wrote: ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD
Art Kendall
Social Research Consultants |
In reply to this post by David Marso
I am definitely doing a CFA to test the theoretical structure behind the
questionnaire, and I am very clear on the purpose of AMOS. My email
wasn't worded to reflect this as my query was not about whether I should
be using AMOS or not. There is no need to be rude.
Kathryn > Date: Tue, 24 Apr 2012 03:05:02 -0700 > From: [hidden email] > Subject: Re: AMOS > To: [hidden email] > > First of all it is obvious you are attempting an EXPLORATORY factor analysis > (EFA) rather than a CONFIRMATORY factor analysis (CFA). AMOS is *NOT* an > appropriate tool to use for performing EFA. > The fact that you have 80 items is the give away. Amos Basic is the > programming language for Amos and you will have to RTFM and learn how to > program to use it. With 80 items you should probably have a few thousand > cases. > --- > > Kathryn Gardner wrote > > > > Dear List, > > I have posted this query to the SEMNET list, but I'm still stuck and I am > > hoping that others on this list might have a suggestion. I > > have a 80-item questionnaire that I want to factor analyse, but I can't > > figure out how to draw a diagram with so many items in AMOS graphics. It > > seems > > impossible as the > > indicators fall off the edge of the page and I can’t then see them. I’ve > > tried selecting “resize diagram to fit page”, but this makes it too > > tiny for me to then drag the SPSS variables into > > each indicator box. Someone suggested using AMOS basic with text input, > > but I don't seem to have this option anywhere in the AMOS progam. Surely > > there must be a way of doing this! > > > > Many thanks, > > Kathryn > > > > > -- > View this message in context: http://spssx-discussion.1045642.n5.nabble.com/AMOS-tp5661369p5661542.html > Sent from the SPSSX Discussion mailing list archive at Nabble.com. > > ===================== > To manage your subscription to SPSSX-L, send a message to > [hidden email] (not to SPSSX-L), with no body text except the > command. To leave the list, send the command > SIGNOFF SPSSX-L > For a list of commands to manage subscriptions, send the command > INFO REFCARD |
In reply to this post by Art Kendall
Hi Art,
I am doing a CFA to test the underlying structure of the measure with the items representing 9 correlated factors/constructs based on 9 criteria for a personality disorder. Because these 9 criteria all represent a unitary personality construct, a unidimensional structure has also been hypothesised according to previous studies. I have approx. 1000 cases. There is a scoring key for the questionnaire. Kathryn Date: Tue, 24 Apr 2012 07:17:56 -0400 From: [hidden email] Subject: Re: AMOS To: [hidden email] Please explain the purpose of your factor analysis. Do you have items that were designed to measure some specific constructs? I.e., do you have a preliminary scoring key? what are the constructs? Are you trying to double check the scoring key from previous work? Do you wish to end up with a scoring key using one/zero weights for the items? Or do you want some other form of factor scores. What do you intend to do with the scores? How many cases do you have? Did you already run a parallel analysis in base SPSS? Art Kendall Social Research Consultants On 4/24/2012 4:41 AM, Kathryn Gardner wrote: ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
In reply to this post by Kathryn Gardner
<base href="x-msg://115/">Hi,
Is the concern you're having just with the visual challenges of doing big models in AMOS? I've done larger models (94 item, 11 factor) in graphics, and there doesn't seem to be a way to make it "look pretty" to me. I would just zoom out and use as much space as I needed (ignoring the little rectangular "page"). I read the results off the text output only, and checked the output to make sure all the parameters were correctly specified (rather than attempt to visually disentangle the spaghetti of the diagram looking for errors). The visual doesn't seem that important to me, because it's not as though you could use that in a manuscript. Mike On Apr 24, 2012, at 4:29 PM, Kathryn Gardner wrote:
|
Free forum by Nabble | Edit this page |