Categorizing Skewed, Limited Dependent Variables

Bengt Muthén and George Speckart
Additional contact information
Bengt Muthén: University of California, Los Angeles
George Speckart: University of California, Los Angeles

Evaluation Review, 1983, vol. 7, issue 2, 257-269

Abstract: Probit analysis is applied in a situation where analysis of covariance (ANCOVA) would customarily be used. The dichotomous dependent variables arise from dichotomizations of skewed continuous variables recorded as the proportion of time certain activities are observed. The probit approach avoids the biases of ordinary A NCO VA that arise due to skewness (limited variation). To illustrate this, data from 225 experiments and 214 control subjects in a drug treatment program was analyzed. It was found that the probit approach was able to reveal more substantial treatment effects than the ordinary A NCO VA.

Date: 1983
References: Add references at CitEc
Citations:

Downloads: (external link)
https://journals.sagepub.com/doi/10.1177/0193841X8300700207 (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:sae:evarev:v:7:y:1983:i:2:p:257-269

DOI: 10.1177/0193841X8300700207

Access Statistics for this article

More articles in Evaluation Review
Bibliographic data for series maintained by SAGE Publications ().