Glossary

Content Validity

Content validity is the extent to which the items of a measurement instrument represent the entire range of the construct they are intended to measure. It is typically assessed by expert judges who review whether the item pool adequately covers all relevant domains and exclude...

Definition

Content validity is the extent to which the items of a measurement instrument represent the entire range of the construct they are intended to measure. It is typically assessed by expert judges who review whether the item pool adequately covers all relevant domains and excludes irrelevant material. Unlike statistical validity coefficients, content validity is primarily a qualitative judgment.

Why It Matters

Even a highly reliable scale is useless if it omits key aspects of the concept it claims to measure. A customer-satisfaction survey that only asks about product quality but ignores service, price, and delivery is missing critical content. Content validity ensures comprehensiveness and relevance at the earliest stage of instrument development, before any data are collected.

Example

A hospital develops a patient-satisfaction questionnaire. A panel of five experts — including nurses, physicians, and health-service researchers — independently rates each item on its relevance to patient experience. They compute the Content Validity Index (CVI), finding that three of twenty items fall below the 0.78 threshold. Those items are revised or removed. The final 17-item questionnaire is then field-tested, confident that it covers the full breadth of the patient-experience domain.

Related Terms

Software Notes

  • SPSS: Content validity is not computed automatically in SPSS because it relies on expert ratings. However, you can use SPSS to calculate the Content Validity Index (CVI) once experts have rated items: enter ratings as 0/1 variables and compute the proportion of experts endorsing each item.
  • R: No dedicated package is required. Use rowMeans(expert_ratings) to compute item-level CVI, and mean(item_cvi) for the scale-level CVI. The psych package can then be used for subsequent reliability and factor analyses.
  • Stata: Content validity is computed manually from expert rating data. egen item_cvi = rowmean(expert1 expert2 expert3 expert4 expert5) computes item-level CVI. summarize item_cvi, detail provides the distribution of CVI values across items.