Internal consistency based on polychoric correlations?

[Ata Ghaderi]

Dear list,

Does anyone know of a SPSS syntax using polychoric correlations for estimating internal consistency. I have a questionnaire in which the items are are respondent to on a 3 point scale, large number of subjects from a random sample (about 1400), and large curtosis or skewness on some items as we are asking some questions that are about odd behaviors related to some sub scales (e.g., stealing, cruelty etc) that are not that common.

Zumbo et el (2007) suggest using internal consistency based on other than Pearson with data like mine. 

Any suggestions that helps me avoiding the currently possible scenario that I can figure out: i.e.  going to LISREL for getting the polychoric correlations and putting them in the formula (Gadermann et al, 2012) like the old days?

Thanks,

Ata Ghaderi

-------------------------------------------------------------------------------

Ata Ghaderi , Ph.D.

Clinical Psychologist, Licensed Psychotherapist

Professor

Department of Clinical Neuroscience, Karolinska Institutet

SE-171 77 Stockholm

Sweden

[hidden email]

[phone]

[phone]

--------------------------------------------------------------------------------

[Jon K Peck]

Polychoric and polyserial correlations are available via the SPSSINC HETCOR extension command available from the SPSS Community website (www.ibm.com/developerworks/spssdevcentral) in the Extension Commands Collection.  Python and R Essentials are prerequisites.


Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
phone: [phone]




From:        Ata Ghaderi <[hidden email]>
To:        [hidden email],
Date:        04/19/2013 06:48 AM
Subject:        [SPSSX-L] Internal consistency based on polychoric correlations?
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

Dear list,

Does anyone know of a SPSS syntax using polychoric correlations for estimating internal consistency. I have a questionnaire in which the items are are respondent to on a 3 point scale, large number of subjects from a random sample (about 1400), and large curtosis or skewness on some items as we are asking some questions that are about odd behaviors related to some sub scales (e.g., stealing, cruelty etc) that are not that common.
Zumbo et el (2007) suggest using internal consistency based on other than Pearson with data like mine.
Any suggestions that helps me avoiding the currently possible scenario that I can figure out: i.e.  going to LISREL for getting the polychoric correlations and putting them in the formula (Gadermann et al, 2012) like the old days?

Thanks,
Ata Ghaderi



-------------------------------------------------------------------------------
Ata Ghaderi , Ph.D.
Clinical Psychologist, Licensed Psychotherapist
Professor
Department of Clinical Neuroscience, Karolinska Institutet
SE-171 77 Stockholm
Sweden
ata.ghaderi@...
[phone]
[phone]
--------------------------------------------------------------------------------

[Ata Ghaderi]

Thanks Jon!

The knowledge and willingness of the list members to help is wonderful!

Best wishes,

Ata

-------------------------------------------------------------------------------

Ata Ghaderi , Ph.D.

Clinical Psychologist, Licensed Psychotherapist

Professor

Department of Clinical Neuroscience, Karolinska Institutet

SE-171 77 Stockholm

Sweden

[hidden email]

[phone]

[phone]

--------------------------------------------------------------------------------

19 apr 2013 kl. 15:35 skrev Jon K Peck <[hidden email]>:

Polychoric and polyserial correlations are available via the SPSSINC HETCOR extension command available from the SPSS Community website (<a href="x-msg://1863/www.ibm.com/developerworks/spssdevcentral">www.ibm.com/developerworks/spssdevcentral) in the Extension Commands Collection.  Python and R Essentials are prerequisites.


Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
phone: [phone]




From:        Ata Ghaderi <[hidden email]>
To:        [hidden email],
Date:        04/19/2013 06:48 AM
Subject:        [SPSSX-L] Internal consistency based on polychoric correlations?
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

Dear list,

Does anyone know of a SPSS syntax using polychoric correlations for estimating internal consistency. I have a questionnaire in which the items are are respondent to on a 3 point scale, large number of subjects from a random sample (about 1400), and large curtosis or skewness on some items as we are asking some questions that are about odd behaviors related to some sub scales (e.g., stealing, cruelty etc) that are not that common.
Zumbo et el (2007) suggest using internal consistency based on other than Pearson with data like mine.
Any suggestions that helps me avoiding the currently possible scenario that I can figure out: i.e.  going to LISREL for getting the polychoric correlations and putting them in the formula (Gadermann et al, 2012) like the old days?

Thanks,
Ata Ghaderi



-------------------------------------------------------------------------------
Ata Ghaderi , Ph.D.
Clinical Psychologist, Licensed Psychotherapist
Professor
Department of Clinical Neuroscience, Karolinska Institutet
SE-171 77 Stockholm
Sweden
[hidden email]
[phone]
[phone]
--------------------------------------------------------------------------------