case0302: Agent Orange

In Sleuth3: Data Sets from Ramsey and Schafer's "Statistical Sleuth (3rd Ed)"

Agent Orange

Description

In 1987, researchers measured the TCDD concentration in blood samples from 646 U.S. veterans of the Vietnam War and from 97 U.S. veterans who did not serve in Vietnam. TCDD is a carcinogenic dioxin in the herbicide called Agent Orange, which was used to clear jungle hiding areas by the U.S. military in the Vietnam War between 1962 and 1970.

Usage

data(case0302)

Format

A data frame with 743 observations on the following 2 variables.

Dioxin

the concentration of TCDD, in parts per trillion

Veteran

factor variable with two levels, "Vietnam" and "Other", to indicate the type of veteran

Source

Ramsey, F.L. and Schafer, D.W. (2013). The Statistical Sleuth: A Course in Methods of Data Analysis (3rd ed), Cengage Learning.

References

Centers for Disease Control Veterans Health Studies: Serum 2,3,7,8-Tetraclorodibenzo-p-dioxin Levels in U.S. Army Vietnam-era Veterans. Journal of the American Medical Association 260: 1249–1254.

Examples

attach(case0302)  
str(case0302)    # Note: Level 1 of Veteran is "Other" (first alphabeticall)

boxplot(Dioxin ~ Veteran)  

t.test(Dioxin ~ Veteran, var.equal=TRUE,
  alternative="less") # 1-sided t-test; alternative: group 1 mean is less  
t.test(Dioxin ~ Veteran, alternative="less", var.equal=TRUE, 
  subset=(Dioxin < 40)) # t-test on subset for which Dioxin < 40  
t.test(Dioxin ~ Veteran, alternative="less", var.equal=TRUE, 
  subset=(Dioxin < 20))    
t.test(Dioxin ~ Veteran, var.equal=TRUE) # 2-sided--to get confidence interval 

## HISTOGRAMS FOR PRESENTATION  
opar <- par(no.readonly=TRUE)  # Store device graphics parameter settings
par(mfrow=c(2,1), mar=c(3,3,1,1)) # 2 by 1 layout of plots; change margins    
myBreaks <- (0:46) - .5    # Make breaks for histogram bins  
hist(Dioxin[Veteran=="Other"], breaks=myBreaks, xlim=range(Dioxin),
  col="green", xlab="", ylab="", main="")     
text(10,25, 
  "Dioxin in 97 'Other' Veterans; Estimated mean =  4.19 ppt (95% CI: 3.72 to 4.65 ppt)",
  pos=4, cex=.75) # CI from 1-sample t-test & subset=(Veteran="Other")    
hist(Dioxin[Veteran=="Vietnam"],breaks=myBreaks,xlim=range(Dioxin),
  col="green", xlab="", ylab="", main="")   
text(10,160,
  "Dioxin in 646 Vietnam Veterans; Estimated mean =  4.26 ppt (95% CI: 4.06 to 4.64 ppt)",
  pos=4, cex=.75)   
text(13,145,"[Estimated Difference in Means: 0.07 ppt (95% CI: -0.63 to 0.48 ppt)]",
  pos=4, cex=.75)  
par(opar) # Restore previous graphics parameter settings
  
detach(case0302)    

Sleuth3 documentation built on May 29, 2024, 2:56 a.m.