Category versus Continuous Survey Responses in Economic Modelling: Monte Carlo and Empirical Evidence
Choice of survey response format involves a tradeoff between information content and measurement error. Estimation precis ion is examined with (1) a Monte Carlo study where ordered logit fits of categorized data with no measurement error are compared with OLS fit s using the same data in continuous form but incorporating substantial measurement error; and (2) an OLS versus ordered logit comparison us ing actual data where 3,800 Army officers stated their service intention s with both continuous and category format questions. In both investigations, continuous data do as well or better than category data. Suggesting that the widespread use of category data may be resulting in unnecessary loss of information. Copyright 1993 by MIT Press.
Volume (Year): 75 (1993)
Issue (Month): 1 (February)
|Contact details of provider:|| Web page: http://mitpress.mit.edu/journals/|
|Order Information:||Web: http://mitpress.mit.edu/journal-home.tcl?issn=00346535|
When requesting a correction, please mention this item's handle: RePEc:tpr:restat:v:75:y:1993:i:1:p:188-93. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Anna Pollock-Nelson)
If references are entirely missing, you can add them using this form.