Display options
Cite
McGinley JS, Curran PJ. Validity Concerns with Multiplying Ordinal Items Defined by Binned Counts: An Application to a Quantity-Frequency Measure of Alcohol Use. Methodology (Gott). 2014;10(3):108-116doi: 10.1027/1614-2241/a000081.
McGinley, J. S., & Curran, P. J. (2014). Validity Concerns with Multiplying Ordinal Items Defined by Binned Counts: An Application to a Quantity-Frequency Measure of Alcohol Use. Methodology : European journal of research methods for the behavioral & social sciences, 10(3), 108-116. https://doi.org/10.1027/1614-2241/a000081
McGinley, James S, and Curran, Patrick J. "Validity Concerns with Multiplying Ordinal Items Defined by Binned Counts: An Application to a Quantity-Frequency Measure of Alcohol Use." Methodology : European journal of research methods for the behavioral & social sciences vol. 10,3 (2014): 108-116. doi: https://doi.org/10.1027/1614-2241/a000081
McGinley JS, Curran PJ. Validity Concerns with Multiplying Ordinal Items Defined by Binned Counts: An Application to a Quantity-Frequency Measure of Alcohol Use. Methodology (Gott). 2014;10(3):108-116. doi: 10.1027/1614-2241/a000081. PMID: 25383075; PMCID: PMC4217492.
Copy Download .nbib Format: NLM
Share it on

Methodology (Gott). 2014;10(3):108-116. doi: 10.1027/1614-2241/a000081.

Validity Concerns with Multiplying Ordinal Items Defined by Binned Counts: An Application to a Quantity-Frequency Measure of Alcohol Use.

Methodology : European journal of research methods for the behavioral & social sciences

James S McGinley, Patrick J Curran

Affiliations

  1. University of North Carolina at Chapel Hill.

PMID: 25383075 PMCID: PMC4217492 DOI: 10.1027/1614-2241/a000081

Abstract

Social and behavioral scientists often measure constructs that are truly discrete counts by collapsing (or binning) the counts into a smaller number of ordinal responses. While prior quantitative research has identified a series of concerns with similar binning procedures, there has been a lack of study on the consequences of multiplying these ordinal items to create a desired index. This measurement strategy is incorporated in many research applications, but it is particularly salient in the study of substance use where the product of ordinal quantity (number of drinks) and frequency (number of days) items is used to create an index of total consumption. In the current study, we demonstrate both analytically and empirically that this multiplicative procedure can introduce serious threats to construct validity. These threats, in turn, directly impact the ability to accurately measure alcohol consumption.

Keywords: Psychometrics; alcohol measurement; coarse categorization; construct validity; quantity-frequency

References

  1. Addiction. 2003 Dec;98 Suppl 2:1-12 - PubMed
  2. Alcohol Res Health. 2003;27(1):18-29 - PubMed
  3. J Subst Abuse. 2000;12(1-2):1-21 - PubMed
  4. Addiction. 2008 Jul;103(7):1082-99 - PubMed
  5. J Gambl Stud. 1996 Jun;12(2):161-78 - PubMed
  6. Int J Addict. 1986 Sep-Oct;21(9-10):1059-79 - PubMed
  7. Int J Epidemiol. 1999 Apr;28(2):219-24 - PubMed
  8. Psychol Methods. 2002 Mar;7(1):19-40 - PubMed
  9. Addiction. 1994 Apr;89(4):407-12 - PubMed
  10. J Stud Alcohol. 1997 Nov;58(6):622-4 - PubMed

Publication Types

Grant support