Skip to main content

Table 7 Content validity of the OCI

From: Development and validation of an occlusal cant index

  

Relevance

    

Clarity

    

Items

Number of experts

Number of ratings of 3 or 4

I-CVIa

PCb

κ*c

Evaluationd

Number of ratings of 3 or 4

I-CVIa

PCb

κ*c

Evaluationd

1

10

9

0.90

0.009

0.90

***

9

0.90

0.009

0.90

***

2

10

10

1.00

0.001

1.00

***

10

1.00

0.001

1.00

***

3

10

10

1.00

0.001

1.00

***

9

0.90

0.009

0.90

***

4

10

9

0.90

0.009

0.90

***

9

0.90

0.009

0.90

***

5

10

10

1.00

0.001

1.00

***

10

1.00

0.001

1.00

***

S-CVI/Avee

 

0.96

    

0.94

    
  1. aI-CVI (item-level content validity index) = number of experts providing a rating of 3 or 4 / number of experts
  2. bPc (probability of chance occurrence) = [N/A * (N - A)] × 0.5 N, N = number of experts; A = number of experts providing a rating of 3 or 4
  3. cκ*(modified kappa) = (I-CVI-Pc)(1-Pc)
  4. dEvaluation criteria for the level of content validity: relationship between I-CVI and k*; excellent validity = I-CVI ≥ 0.78 and k* > 0.74 (****); good validity I-CVI < 0.78 and ≥ 0.60 and k* ≤ 0.74 (***); fair validity I-CVI < 0.6 and ≥ 0.40 and k* ≤ 0.59 (**); and poor validity I-CVI < 0.4 and κ* < 0.40 (*)
  5. eS−CVI/Ave (scale−level content validity index based on the average agreement among experts) = sum of the I−CVI / number of items