Sleuth3 (version 1.0-3)

case1901: Sex Role Sterotypes and Personnel Decisions

Description

Researchers gave 48 male bank supervisors attending a management institute hypothetical personnel files and asked them whether they would promote the applicant based on the file. The personnel files were identical except that 24 of them listed a male and 24 listed a female applicant. The assignment of managers to receive either a male or female applicant file was carried out at random.

Usage

case1901

Format

A data frame with 2 observations on the following 3 variables.

Gender

a factor with levels "Female" and "Male"

Promoted

the number of managers who promoted the applicant

NotPromoted

the number of managers who did not promote the applicant

References

Rosen, B. and Jerdee, J (1974). Influence of Sex Role Steroetypes on Personnel Decisions, Journal of Applied Psychology 59: 9--14.

Examples

# NOT RUN {
str(case1901)
attach(case1901)
   
## INFERENCE
myTable             <- cbind(Promoted,NotPromoted)    
row.names(myTable)  <- Gender    
myTable   
fisher.test(myTable, alternative="greater")  
  # Alternative: that odds of Promotion in first row (Males) are greater.
fisher.test(myTable)  # Use 2-sided to get confidence interval for odds ratio
prop.test(myTable) # Compare two binomial proportions


## GRAPHICAL DISPLAY FOR PRESENTATION
myTable
#         Promoted NotPromoted
#Male         21           3
#Female       14          10
prop.test(21,(21+3)) # Est = .875; CI = .665 to .967  
prop.test(14,(14+10))# Est = .583; CI = .369 to .772  

pHat   <- c(0.875,0.583)       
lower95 <- c(0.665, 0.369)
upper95 <- c(0.967, 0.772)
if(require(Hmisc)) {      # Use Hmisc library
  myObj<- Cbind(pHat,lower95,upper95) # Cbind: a form of cbind needed for Dotplot 
  Dotplot(Gender ~ myObj,  
  xlab="Probability of Promotion Based on Applicant File (and 95% Confidence Intervals)",
    ylab="Gender Listed in Applicant File", ylim=c(.5,2.5), cex=2)
}

detach(case1901)
# }