population_bf: Aggregation of Individual Bayes Factors

In multinomineq: Bayesian Inference for Multinomial Models with Inequality Constraints

Aggregation of Individual Bayes Factors

Description

Aggregation of multiple individual (N=1) Bayes factors to obtain the evidence for a hypothesis in a population of persons.

Usage

population_bf(bfs)

Arguments

bfs	a vector with individual Bayes factors, a matrix with one type of Bayes-factor comparison per column, or a list of matrices with a named column "bf" (as returned by bf_multinom/count_to_bf).

Value

a vector or matrix with named elements/columns:

• population_bf: the product of individual BFs

• geometric_bf: the geometric mean of the individual BFs

• evidence_rate: the proportion of BFs>1 (BFs<1) if geometric_bf>1 (<1). Values close to 1.00 indicate homogeneity.

• stability_rate: the proportion bfs>geometric_bf (<) if geometric_bf>1 (<). Values close to 0.50 indicate stability.

References

Klaassen, F., Zedelius, C. M., Veling, H., Aarts, H., & Hoijtink, H. (in press). All for one or some for all? Evaluating informative hypotheses using multiple N = 1 studies. Behavior Research Methods. https://doi.org/10.3758/s13428-017-0992-5

Examples

# consistent evidence across persons:
bfs <- c(2.3, 1.8, 3.3, 2.8, 4.0, 1.9, 2.5)
population_bf(bfs)

# (A) heterogeneous, inconsistent evidence
# (B) heterogeneous, inconsistent evidence
bfs <- cbind(
  A = c(2.3, 1.8, 3.3, 2.8, 4.0, 1.9, 2.5),
  B = c(10.3, .7, 3.3, .8, 14.0, .9, 1.5)
)
population_bf(bfs)

multinomineq documentation built on Nov. 22, 2022, 5:09 p.m.