AEA365 | A Tip-a-Day by and for Evaluators



WMU Scribing: Kristin Hobson on the Construct Validity of Race

The version below is updated and revised at the request of the presenter.

My name is Kristin Hobson. I am in the Interdisciplinary Ph.D. in Evaluation program at Western Michigan University. Recently, I participated in AEA’s annual conference in San Antonio where I learned from presentations given by colleagues.

I served as a session scribe for one such interactive session that covered the topic of race constructs and its validity. Kelly Robertson and Diane Rogers presented Construct Validity of Race and Its Impact on the Quality of Research and Evaluation.

Lessons Learned:

  • Traditional constructs of race are based on genetics, demography, and/or physical characteristics which are unreliable determinants. According to Campbell & Fiske (1959) to be valid, a construct must have both convergent and discriminate validity. Convergent validity occurs when two measures of the same thing correlate with one another; race lacks convergent validity because the definition is neither consistent nor reliable. Discriminant validity occurs when two measures of different things don’t correlate with one another; race lacks discriminant validity since it usually covaries with socioeconomic and health demographics. For example, when measuring infant death in the US, African American infants are more than twice as likely as White infants to die in the first year of life (Gearon, 2008).
  • Racial stratification is real but race is not, it’s a social construct used to classify people, and is problematic in terms of how it’s interpreted and leads to differences in power and access to resources. Although, the social construct of race should be examined to determine the effects of racial prejudice.
  • Literature indicates that the impact of the traditional constructs of race reinforces racial stereotypes and discrimination and reduces the validity of findings.
  • Some tips for the responsible use of race as a social construct:
    • Gather data on and examine results in terms of structural/systemic context to avoid promoting stereotypes, increase the validity of findings, and identify programmatic changes.
    • Note limitations of using race.
    • Use self-report, allow individuals to indicate multiple races, and provided detailed instructions.
    • Race should not be used as a causal/independent variable or as a proxy for variables such as education and income.
    • Be cautious when using historic race data. Census instruments often change. Race recorded at death is often determined by the mortician, commonly resulting in the classification of an individual’s “race” changing from birth to death.

Rad Resources:

At AEA’s 2010 Annual Conference, session scribes took notes at over 30 sessions and we’ll be sharing their work throughout the winter on aea365. This week’s scribing posts were done by the students in Western Michigan University’s Interdisciplinary PhD program. Do you have questions, concerns, kudos, or content to extend this aea365 contribution? Please add them in the comments section for this post on the aea365 webpage so that we may enrich our community of practice.

· · ·

No comments yet.

Leave a Reply




To top