Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Ordered response threshold values overlap


From   Christine Kalenborn <[email protected]>
To   [email protected]
Subject   st: Ordered response threshold values overlap
Date   Tue, 09 Jul 2013 15:23:56 +0200

Dear STATA users,

I am looking at the determinants of life satisfaction. As the dependent variable is of ordinal scale (1 "dissatisfied" to 10 "satisfied"), I use an ordered response model.
When I compute the regression results, the cutpoints strongly overlap.
Example:
       /cut1 |  -3.292553   2.309304 -7.818706    1.233599
       /cut2 |  -3.021334    2.30325 -7.535621    1.492953
       /cut3 |  -2.631861   2.296314 -7.132554    1.868832
       /cut4 |  -2.330351   2.291877 -6.822349    2.161646
       /cut5 |  -1.791179   2.291642 -6.282714    2.700356
       /cut6 |  -1.413765   2.285604 -5.893467    3.065937
       /cut7 |  -.9014434   2.286238 -5.382388    3.579501
       /cut8 |  -.1529157   2.264794 -4.591831       4.286
       /cut9 |   .4205266   2.252341 -3.993981    4.835034

I interpreted this as a sign for the missing difference between the threshold values. In addition to that I calculated the test of equality of cut off values:

 ( 1)  [cut1]_cons - [cut2]_cons = 0
 ( 2)  [cut1]_cons - [cut3]_cons = 0
 ( 3)  [cut1]_cons - [cut4]_cons = 0
 ( 4)  [cut1]_cons - [cut5]_cons = 0
 ( 5)  [cut1]_cons - [cut6]_cons = 0
 ( 6)  [cut1]_cons - [cut7]_cons = 0
 ( 7)  [cut1]_cons - [cut8]_cons = 0
 ( 8)  [cut1]_cons - [cut9]_cons = 0

           chi2(  8) = 5817.80
         Prob > chi2 =    0.0000

The null hypothesis is highly rejected which leads to the conclusion that the thresholds are different. Should I rescale the life satisfaction values to avoid overlapping cut points or is this not necessary? I hope anybody can help me.

Thank you! Best wishes


Christine Kalenborn

--
..........................................................
Dipl. Volkswirtin Christine Kalenborn
Chair of Economics, esp. Macroeconomics,
Spatial Economics and Regional Science
Institute of Transport & Economics
Faculty of Traffic Sciences "Friedrich List"
Technische Universität Dresden
D-01062 Dresden
Germany
Würzburger Str. 35
Room 009
Phone: +49-351-463-36806
Fax:     +49-351-463-36819
http://tu-dresden.de/die_tu_dresden/fakultaeten/vkw/iwv/vwl/index_html

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index