For accurate study data, you need a tool that correctly & comprehensively measures the outcome of interest (concept). If a tool measures your outcome of interest accurately it has strong validity. If it measures that outcome consistently, it has high reliability.

For now, let’s focus on validity.

Again, validity is how well a research tool measures what it is intended to measure. 

The four (4) types of validity are 1) face, 2) content, 3) construct, & 4) criterion-related. Click here to read my blog on face validity–the weakest type. Now, let’s step it up a notch to content validity.

Content validity is the comprehensiveness of a data collection survey tool. In other words, does the instrument include items that measure all aspects of the thing (concept) you are studying–whether that thing be professional quality of life, drug toxicity, spiritual health, pain, or something else.

When you find a tool that you want to use, look for documented content validity. Content validity means that the tool creators:

• 1) adopted a specific definition of the concept they want to measure,
• 2) generated a list of all possible items from a review of literature and/or other sources,
• 3) gave both their definition and item list to 3-5+ experts on the topic, &
• 4) asked those experts independently to rate how well each item represents the adopted concept definition (or not). Often experts are asked to evaluate item clarity as well.

When a majority of the expert panel agrees that an item matches the definition, then that item becomes part of the new tool. Items without agreement are tossed. Experts may also edit items or add items to the list, and the tool creator may choose to submit edited and new items to the whole expert panel for evaluation.

Optionally tool creators  may statistically calculate a content validity index (CVI) for items and/or for the tool as a whole, but content validity is still based on experts’ judgment. Some tool authors are just more comfortable with having a number to represent that judgment. An acceptable CVI > 0.78; the “>” means “greater than or equal to.” (Click here for more on item & scale CVIs. )

When reading a research article, you might see content validity reported for the tool. Here’s an example: Content…validity of the nurse and patient [Spiritual Health] Inventories…[was] based on literature review [and] expert panel input….Using a religious-existential needs framework, 59 items for the nurse SHI were identified from the literature with the assistance of a panel of theology and psychology experts…. Parallel patient items were developed, and a series of testing and revisions was completed resulting in two 31-item tools (p. 4, Highfield, 1992).

For more, check out this  quick explanation of content validity: 3 minute YouTube video. If you are trying to establish content validity for your own new tool, consult a mentor and a research text like Polit & Beck’s Nursing research: Generating and assessing evidence for nursing practice.

Critical thinking: What is the difference between face and content validity? How are they alike. (Hint: check out the video.) What other questions do you have?