Standard Article

Validity and Reliability

Part 2. Marketing Research

  1. Frederick Winter

Published Online: 15 DEC 2010

DOI: 10.1002/9781444316568.wiem02056

Wiley International Encyclopedia of Marketing

Wiley International Encyclopedia of Marketing

How to Cite

Winter, F. 2010. Validity and Reliability. Wiley International Encyclopedia of Marketing. 2.

Author Information

  1. University of Pittsburgh,, Pittsburgh, PA, USA

Publication History

  1. Published Online: 15 DEC 2010

Abstract

The reliability and validity of the measurements that we use in marketing research are crucial if we are to have meaningful results. Reliability is whether the individual measure taken is close to the true actual measure (precise) or if there is so much noise, variance, or deviation from the true measure that the measure that we take has no precision. Alternative ways to assess reliability are test–retest reliability where we correlate a second measure (retest) with an original measure. Split half reliability splits the sample into two subsamples and compares the correlation. Validity is concerned with whether the measure one takes really is measuring what one wants it to measure. There are alternative types of validity – face validity, construct validity, and criterion validity. Assessment of validity is a bit of an art, but researchers need to make a good effort to insure that their results have meaning. The absence of either reliability or validity makes marketing research either meaningless (in the case of validity problems) or not very precise and therefore unable to show differences the research might be trying to detect.

Keywords:

  • validity;
  • reliability;
  • scales;
  • measurement;
  • constructs