stat_ecs {callback}

Exclusive callback shares

Description

Computes the callback shares and their confidence intervals. The analysis is restricted to the tests with discrimination cases.

Usage

stat_ecs(x, level = 0.95)

Arguments

x

a callback object.

level

a number, containing the level of the confidence intervals (0.95 by default).

Value

A list with class "stat_ecs" containing 8 components : level, shares, cp, wilson, student, t.student, t.pearson and t.fisher.

level: the level of the confidence intervals.

shares: a data frame containing the following variables.

• disc: number of discrimination cases.

• c10: number of tests with the 1st candidate preferred (2nd candidate discriminated against).

• c01: number of tests with the 2nd candidate preferred (1st candidate discriminated against).

• cdif: net discrimination c10-c01.

• p_cand1: 1st candidate callback share (c10/disc).

• p_cand2: 2nd candidate callback share (c01/disc).

• p_cand_dif: p_cand1-1/2.

cp: a data frame containing the Clopper-Pearson confidence intervals, from binom.test(), and the p-value of the Fisher test of independence between the candidate type and the callback variable, from fisher.test().

• inf_p_cand1: 1st candidate callback rate, lower bound.

• sup_p_cand1: 1st candidate callback rate, upper bound.

• inf_p_cand2: 2nd candidate callback rate, lower bound.

• sup_p_cand2: 2nd candidate callback rate, upper bound.

wilson: a data frame containing the Wilson confidence intervals and the p-value of the equality test of callback shares between the two candidates, from prop.test().

• inf_p_cand1: 1st candidate callback share, lower bound.

• sup_p_cand1: 1st candidate callback share, upper bound.

• inf_p_cand2: 2nd candidate callback share, lower bound.

• sup_p_cand2: 2nd candidate callback share, upper bound.

• inf_cand_dif: p_c10-p_c01, lower bound.

• sup_cand_dif: p_c10-p_c01, upper bound.

student: a data frame containing the Student confidence intervals and the p-value of the equality test of callback shares between the two candidates.

• inf_p_cand1: 1st candidate callback share, lower bound.

• sup_p_cand1: 1st candidate callback share, upper bound.

• inf_p_cand2: 2nd candidate callback share, lower bound.

• sup_p_cand2: 2nd candidate callback share, upper bound.

• inf_cand_dif: p_c10-p_c01, lower bound.

• sup_cand_dif: p_c10-p_c01, upper bound.

t.fisher: a data frame containing the statistics of the Fisher test.

• p_cand_dif: 1st candidate callback share - 1/2.

• p_Fisher: the p-value of the Fisher test.

• s_Fisher: the significance code of the Fisher test.

t.pearson: a data frame containing the statistics of the Pearson test.

• p_cand_dif: 1st candidate callback share - 1/2.

• Pearson: the value of Pearson's chi-squared test statistic.

• p_Pearson: the p-value of the Pearson test.

• s_Pearson: the significance code of the Pearson test.

t.student: A data frame containing the statistics of the Student test.

• p_cand_dif: 1st candidate callback share - 1/2.

• Student: the value of Student's test statistic.

• p_Student: the p-value of the Student test.

• s_Student: the significance code of the Student test.

Author(s)

Emmanuel Duguet

References

Clopper, C. J. & Pearson, E. S. (1934). The use of confidence or fiducial limits illustrated in the case of the binomial. Biometrika, 26, 404–413. doi:10.2307/2331986.

Wilson, E.B. (1927). Probable inference, the law of succession, and statistical inference. Journal of the American Statistical Association, 22, 209–212. doi:10.2307/2276774.

Examples

data(labour1)
x <- callback(data=labour1,cluster="offer",candid="hist",callback="callback")
str(stat_ecs(x,level=0.9))