search

jasp-stats / jasp-issues

This repository is solely meant for reporting of bugs, feature requests and other issues in JASP.
59 stars 29 forks source link

[Feature Request]: Use correlation matrix as input #1802

Closed glcanivez closed 2 weeks ago

glcanivez commented 2 years ago

Description

Factor Analysis from Correlation Matrix

Purpose

To run factor analysis using a correlation matrix rather than raw data

Use-case

No response

Is your feature request related to a problem?

No, the current EFA/CFA seems to operate similarly to other statistics programs

Describe the solution you would like

It would be very useful to be able to input a correlation matrix and run factor analysis (EFA and CFA) from that matrix when one does not have raw data as input.

Describe alternatives that you have considered

In SPSS I use Syntax to input correlation matrices and in EQS I can input a correlation matrix, means, and standard deviations to reconstruct a covariance matrix for CFA

Additional context

This feature would also be helpful for second-order/higher-order EFA using first-order factor pattern coefficients and the factor correlations. I have also used SPSS syntax for this and third-party freeware (Watkins) to conduct these.

juliuspfadt commented 2 years ago

Hi @glcanivez, thanks for the request. @AlexanderLyNL is there anything on the horizon regarding covariance/correlation matrices as input instead of the data matrix?

mww10 commented 3 weeks ago

Several have requested correlation matrix input for EFA and correlation/standard deviation input for CFA. #2174 is marked as closed. This option is not available in JASP 0.19.1 for my Apple silicon machine. Why was this request closed if it was not implemented?

juliuspfadt commented 3 weeks ago

Looking into https://github.com/jasp-stats/jasp-issues/issues/2174, I see that it was marked as a duplicate which is why it was closed. The issue is still relevant.

mww10 commented 3 weeks ago

Thanks. Correlation matrix and correlation matrix/standard deviations for EFA and CFA, respectively, would greatly increase utility of JASP for me and other colleagues who frequently conduct EFA and CFA in measurement studies.

Marley

On Thu, Oct 10, 2024 at 1:45 PM Julius Pfadt @.***> wrote:

Looking into #2174 https://github.com/jasp-stats/jasp-issues/issues/2174, I see that it was marked as a duplicate which is why it was closed. The issue is still relevant.

— Reply to this email directly, view it on GitHub https://github.com/jasp-stats/jasp-issues/issues/1802#issuecomment-2406007058, or unsubscribe https://github.com/notifications/unsubscribe-auth/BAHKF6ASSBQ4LIUGUNDA7LTZ23RPBAVCNFSM6AAAAABPXKSKH6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDIMBWGAYDOMBVHA . You are receiving this because you commented.Message ID: @.***>

juliuspfadt commented 3 weeks ago

you are right. I should get on this :)

mww10 commented 3 weeks ago

Thank you. I appreciate your support. After this is done I will contact my publisher to recommend a new book on factor analysis using JASP.

Marley

On Thu, Oct 10, 2024 at 11:08 PM Julius Pfadt @.***> wrote:

you are right. I should get on this :)

— Reply to this email directly, view it on GitHub https://github.com/jasp-stats/jasp-issues/issues/1802#issuecomment-2406617846, or unsubscribe https://github.com/notifications/unsubscribe-auth/BAHKF6H2WGPGPERKML5PR6DZ25TPXAVCNFSM6AAAAABPXKSKH6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDIMBWGYYTOOBUGY . You are receiving this because you commented.Message ID: @.***>

glcanivez commented 2 weeks ago

Thank you for your work on this. I look forward to the JASP update where this will be available.

Cheers, Gary

Gary L. Canivez, Ph.D. Professor of Psychology Senior Editor, School Psychology Review Former Associate Editor for Psychological Assessment and Archives of Scientific Psychology Department of Psychology 600 Lincoln Avenue Eastern Illinois University Charleston, IL 61920

217-581-6413 (ph) 217-581-6764 (fax)

email: @.*** web site: www.ux1.eiu.edu/~glcanivez LinkedIn: www.linkedin.com/in/glcanivez Research Gate: https://www.researchgate.net/profile/Gary_Canivez Google Scholar: https://scholar.google.com/citations?user=tWsy9SMAAAAJ&hl=en

“It is easier to fool people than to convince them that they have been fooled.” Mark Twain

On Oct 16, 2024, at 7:44 AM, Julius Pfadt @.***> wrote:

Closed #1802https://github.com/jasp-stats/jasp-issues/issues/1802 as completed via jasp-stats/jaspFactor#239https://github.com/jasp-stats/jaspFactor/pull/239.

— Reply to this email directly, view it on GitHubhttps://github.com/jasp-stats/jasp-issues/issues/1802#event-14677881652, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AVKFSVOITXPMFYH46QR6DHDZ3ZNS7AVCNFSM6AAAAABPXKSKH6VHI2DSMVQWIX3LMV45UABCJFZXG5LFIV3GK3TUJZXXI2LGNFRWC5DJN5XDWMJUGY3TOOBYGE3DKMQ. You are receiving this because you were mentioned.Message ID: @.***>