case1401: Chimp Learning Times

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

Description

Researchers taught each of 4 chimps to learn 10 words in American sign language and recorded the learning time for each word for each chimp. They wished to describe chimp differences and word differences.

Usage

case1401

Format

A data frame with 40 observations on the following 4 variables.

Minutes

learning time in minutes

Chimp

a factor indicating chimp, with four levels "Booee", "Cindy", "Bruno" and "Thelma"

Sign

a factor indicating word taught, with 10 levels

Order

the order in which the sign was taught

Source

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

References

Fouts, R.S. (1973). Acquisition and Testing of Gestural Signs in Four Young Chimpanzees, Science 180: 978–980.

Examples

str(case1401)
attach(case1401)
 
## EXPLORATION AND MODEL DEVELOPMENT
plot(Minutes ~ Sign)
myLm1 <- lm(Minutes ~ Chimp + Sign)   
plot(myLm1,which=1)  # Plot residuals (indicates a need for transformation).
logMinutes <- log(Minutes)
myLm2 <- lm(logMinutes ~ Chimp + Sign)
plot(myLm2, which=1)  # This looks fine.
if(require(car)){   # Use the car library                                   
  crPlots(myLm2) # Partial residual plots
}

## INFERENCE AND INTERPRETATION
myLm3  <- update(myLm2, ~ . - Chimp)  # Fit reduced model without Chimp.
anova(myLm3, myLm2)   # Test for Chimp effect.
myLm4 <- update(myLm2, ~ . - Sign)  # Fit reduced model without Sign.
anova(myLm4, myLm2)    # Test for Sign effect.
# Fit 2-way model without intercept to order signs from easiest to hardest 
myAov1 <- aov(logMinutes ~ Sign + Chimp - 1)  
sort(myAov1$coef[1:10]) # Show the ordering of Signs
orderedSign <- factor(Sign,levels=c("listen","drink","shoe","key","more",
      "food","fruit","hat","look","string") )  # Re-order signs, easiest 1st
myAov2 <- aov(logMinutes ~ orderedSign + Chimp - 1)  # Refit
opar <- par(no.readonly=TRUE)  # Store current graphics parameters settings
par(mar=c(4.1,7.1,4.1,2.1)) # Adjust margins to allow room for y-axis labels

## takes too long to run
if(require(multcomp)){   # Use the multcomp library   
  myMultComp    <- glht(myAov2, linfct = mcp(orderedSign = "Tukey"))              
  plot(myMultComp)  # Plot Tukey-adjusted confidence intervals.
  summary(myMultComp)   # Show Tukey-adjusted p-values  pairwise comparisons  
  confint(myMultComp)   # Show Tukey-adjusted 95% confidence intervals  
}

par(opar) # Restore original graphics parameters settings

## DISPLAY FOR PRESENTATION
myYLab <- "Log Learning Time, Adjusted for Chimp Effect"
myXLab <- "Sign Learned"      
if(require(car)){   # Use the car library  
crPlots(myAov2, ylab=myYLab, xlab=myXLab,
  main="Learning Times by Sign, Adjusted for Chimp Effects",
  layout=c(1,1)) # Click on graph area to show next page (Just use first one.) 
}  

detach(case1401)

Sleuth3 documentation built on May 2, 2019, 6:41 a.m.