What is I-CVI?

Item-level Content Validity Index: measures expert agreement on individual item relevance

I-CVI (Item-level Content Validity Index) is a validity metric that measures the proportion of experts who agree that a specific item on a measurement instrument is relevant and should be retained. Unlike CVR, I-CVI provides a straightforward probability-based measure ranging from 0 to 1.

I-CVI is particularly useful because it:

I-CVI Formula

I-CVI = Number rating item as essential / Total number of experts

Interpretation

I-CVI directly represents the proportion of experts who rated the item as essential:

Acceptable I-CVI Thresholds

Different evaluation criteria apply depending on the number of experts on your panel. The most commonly cited standards are from Lynn (1986) and Polit & Beck (2006):

ExpertsCriterionStatusAction
1-2I-CVI = 1.0Valid ✓Retain item
3+I-CVI ≥ 0.78Valid ✓Retain item
3+0.70 to 0.78QuestionableRevise or gather more data
3+I-CVI < 0.70InvalidRemove or substantially revise

Why These Thresholds?

With 1-2 experts, you need perfect agreement (1.0) because the expert panel is so small. With 3+ experts, 0.78 is considered the minimum threshold because this represents statistically significant agreement. The research shows that items rated essential by 78% of experts are consistently validated in later testing.

Practical Example

Example: Nursing competency scale

Situation: A nursing board develops a competency assessment with 5 experts reviewing a patient care item.

Ratings received:

  • Expert 1: Essential
  • Expert 2: Essential
  • Expert 3: Useful but not essential
  • Expert 4: Essential
  • Expert 5: Essential

Calculation:

I-CVI = 4 / 5 = 0.80 ✓ Valid

Interpretation: With 5 experts and minimum threshold 0.78, this item's I-CVI of 0.80 is valid. The item should be retained.

I-CVI vs CVR

FeatureI-CVICVR
Range0 to 1-1 to +1
InterpretationProportion of expertsProportion difference from chance
Ease of understandingVery intuitiveLess intuitive
Threshold depends onNumber of expertsNumber of experts
Best used whenYou want simplicityYou want statistical rigor
Note: I-CVI is increasingly preferred in modern research because it's easier to interpret and explain to stakeholders. Many researchers now report both I-CVI and CVR for transparency.

Common Mistakes to Avoid

References

Lynn, M. R. (1986). Determination and quantification of content validity. Nursing Research, 35(6), 382-385.

Polit, D. F., & Beck, C. T. (2006). The content validity index: Are you sure you know what's being reported? Research in Nursing & Health, 29(5), 489-497.