May 05, 2016, By: Maher M. El-Masri, RN, PhD
Validity: The extent to which a psychometric instrument measures what it is intended to measure
In the April 2016 column, I discussed psychometric measures, which researchers use to quantify abstract concepts, such as quality of life (for ease of reading, I will use instruments in this column as an umbrella term for all psychometric measures). A key question for researchers and research readers is this: how does one know whether an instrument is valid (i.e., it measures what it is supposed to measure)?
Instrument developers use complex procedures to establish that an instrument is valid. However, research readers need to be familiar with at least the basic principles of instrument validity. Validity indicates the appropriateness of the instrument as a measure of a concept. A valid instrument is one that has been demonstrated to have two characteristics: translational validity (also known as construct validity) and criterion validity.
Translational validity can be defined as the extent to which an abstract concept has been captured by a psychometric instrument. Determining translational validity is a subjective process in which researchers use their theoretical knowledge of the concept to judge the degree to which the instrument reflects our understanding of the concept. Face validity testing and content validity testing are the two most common ways to ascertain translational validity.
Establishing face validity involves looking at the instrument as a whole and evaluating whether “at face value” it seems like a good translation of the concept, as we understand it. This process is entirely subjective; therefore, other testing should also be done to demonstrate validity. Establishing content validity goes a step further: a panel of content experts evaluates whether each item of the instrument is relevant to the concept being measured. For an instrument designed to measure quality of life, for instance, experts will determine whether the items cover the factors that have been shown in the literature to be associated with quality of life, such as family, social and emotional well-being.
Criterion validity, in its simplest form, is the degree to which a new instrument corresponds with one or more existing instruments (known as criterion instruments) that have previously been shown to be valid measures of the concept. To establish criterion validity, researchers will administer the new instrument and a criterion instrument to a group of study participants. The higher the correlation between the scores obtained, the stronger the evidence of criterion validity in the new instrument.
The onus is on researchers to ensure they use valid instruments when measuring abstract concepts in their research and to provide evidence of the validity of these instruments when reporting results. It is the responsibility of readers to critically appraise research articles for validity reporting. Absence of such evidence should raise serious concerns about the appropriateness of the measurements and the overall findings of the study.
NurseONE.ca resource on this topic
Fain, J. A. (2013). Reading, Understanding, and Applying Nursing Research (4th ed.).
Maher M. El-Masri, RN, PhD, is a full professor and research chair in the faculty of nursing, University of Windsor, in Windsor, Ont.