Include your answers in this document in the sections below the rubric.

# Rubric

Answer the questions with the two data examples.

# read the data
adm$Acceptance <- relevel(adm$Acceptance, "Success")
str(adm)
## 'data.frame':    560 obs. of  3 variables:
##  $Gender : Factor w/ 2 levels "Female","Male": 1 2 1 1 2 1 2 1 1 1 ... ##$ Acceptance: Factor w/ 2 levels "Success","Failure": 2 1 2 2 2 1 2 2 1 2 ...
##  $School : Factor w/ 2 levels "Art","Business": 2 1 2 2 2 2 2 2 1 2 ... head(adm) ## Gender Acceptance School ## 1 Female Failure Business ## 2 Male Success Art ## 3 Female Failure Business ## 4 Female Failure Business ## 5 Male Failure Business ## 6 Female Success Business tail(adm) ## Gender Acceptance School ## 555 Male Failure Art ## 556 Male Success Art ## 557 Female Success Art ## 558 Female Failure Business ## 559 Female Success Business ## 560 Male Success Business ## Admission rates by gender ## Gender ## Acceptance Female Male ## Success 88 198 ## Failure 112 162 ## Gender ## Acceptance Female Male ## Success 0.44 0.55 ## Failure 0.56 0.45 ## Joining by: Acceptance, Gender 1. (2 p) Interpret the table and associated plot. For example, “The proportion of Females is XX, which is XX higher/lower than Males.” ## Admission rates by gender, for each School ## [1] "Art" ## Gender ## Acceptance Female Male ## Success 64 180 ## Failure 16 60 ## Gender ## Acceptance Female Male ## Success 0.80 0.75 ## Failure 0.20 0.25 ## [1] "Business" ## Gender ## Acceptance Female Male ## Success 24 18 ## Failure 96 102 ## Gender ## Acceptance Female Male ## Success 0.20 0.15 ## Failure 0.80 0.85 ## Joining by: Acceptance, Gender, School 1. (1 p) Interpret the table and assocated plot. 2. (1 p) Compare the Acceptance by Gender (where schools were combined) to when Schools were treated separately. What do you observe? 3. (1 p) Does this surprise you? Why? 4. (1 p) What is the name of the “paradox” given to this data phenomenon? # Death penalty Example 1 - Death Penalty A 2-by-2-by-2 in from Agresti (1990) studied effects of racial characteristics on whether individuals convicted of homicide received the death penalty. The 326 subjects were defendants in homicide indictments in 20 Florida counties during 1976-1977. Is there an association between death penalty, defendant’s race and victim’s race? What kind of association? We can display this table in long format: # read the data penalty <- read.table(text=" Defendant Victim Death Freq white white yes 19 white white no 132 white black yes 0 white black no 9 black white yes 11 black white no 52 black black yes 6 black black no 97 ", header = TRUE) penalty$Death <- relevel(penalty$Death, "yes") str(penalty) ## 'data.frame': 8 obs. of 4 variables: ##$ Defendant: Factor w/ 2 levels "black","white": 2 2 2 2 1 1 1 1
##  $Victim : Factor w/ 2 levels "black","white": 2 2 1 1 2 2 1 1 ##$ Death    : Factor w/ 2 levels "yes","no": 1 2 1 2 1 2 1 2
##  \$ Freq     : int  19 132 0 9 11 52 6 97

## Death penalty by Defendant race

##      Defendant
## Death black white
##   yes    17    19
##   no    149   141
##      Defendant
## Death     black     white
##   yes 0.1024096 0.1187500
##   no  0.8975904 0.8812500
## Joining by: Death, Defendant

1. (2 p) Interpret the table and assocated plot.

## Death penalty by Defendant race, by Victim’s race

## [1] "black"
##      Defendant
## Death black white
##   yes     6     0
##   no     97     9
##      Defendant
## Death      black      white
##   yes 0.05825243 0.00000000
##   no  0.94174757 1.00000000
## [1] "white"
##      Defendant
## Death black white
##   yes    11    19
##   no     52   132
##      Defendant
## Death     black     white
##   yes 0.1746032 0.1258278
##   no  0.8253968 0.8741722
## Joining by: Death, Defendant, Victim

1. (1 p) Interpret the tables and assocated plot.

2. (1 p) Compare the Death by Defendant (where victim’s race were combined) to when victim’s race were treated separately. What do you observe?