Skip to main content

Table 2 Details of informative priors used in Bayesian latent class analysis for test parameters. Alpha and Beta denote the parameters of the beta distribution and the mean, standard deviation (SD), 95% greater than and mode indicate the properties of the distribution with the given parameters. The sum of Alpha and Beta is often known as the ‘sample size equivalent’ and the effect of the prior can be thought of as adding alpha + beta samples to the analysis, with alpha samples being positive

From: Latent class analysis to evaluate performance of point-of-care CCA for low-intensity Schistosoma mansoni infections in Burundi

Parameter

Test

Alpha

Beta

Mean (%)

SD (%)

95% greater than (%)

Mode (%)

Sample size equivalent

Sensitivity

KK

1.43

1.29

53

26

10

60

2.72

CCAB

3.05

1.51

53

26

30

80

4.56

CCAL

3.05

1.51

53

26

30

80

4.56

CAA

3.05

1.51

53

26

30

80

4.56

Specificity

KK

21.2

2.06

91

6

80

95

23.26

CCAB

5.38

1.49

53

26

50

90

6.87

CCAL

5.38

1.49

53

26

50

90

6.87

CAA

5.38

1.49

53

26

50

90

6.87