Note: This document was converted to R-Markdown from this page by M. Drew LaMar. You can download the R-Markdown here.
Download the R code on this page as a single file here
Hover over a function argument for a short description of its meaning. The variable names are plucked from the examples further below.
One-sample \(t\)-test:
t.test(heat$temperature, mu = 98.6)
Other new methods:
Confidence intervals for the variance and the standard deviation.
Confidence intervals for the population mean, variance, and standard deviation using eye span measurements from a sample of stalk-eyed flies.
Read and inspect the data.
stalkie <- read.csv(url("http://whitlockschluter.zoology.ubc.ca/wp-content/data/chapter11/chap11e2Stalkies.csv"))
stalkie
## individual eyespan
## 1 1 8.69
## 2 2 8.15
## 3 3 9.25
## 4 4 9.45
## 5 5 8.96
## 6 6 8.65
## 7 7 8.43
## 8 8 8.79
## 9 9 8.63
Histogram with options
hist(stalkie$eyespan, right = FALSE, col = "firebrick", las = 1,xlab = "Eye span (mm)", ylab = "Frequency", main = "")
95% confidence interval for the mean. Adding $conf.int
after the function t.test
causes R to give the 95% confidence interval for the mean.
t.test(stalkie$eyespan)$conf.int
## [1] 8.471616 9.083940
## attr(,"conf.level")
## [1] 0.95
99% confidence interval for the mean. Adding the argument conf.level=0.99
changes the confidence level of the confidence interval.
t.test(stalkie$eyespan, conf.level = 0.99)$conf.int
## [1] 8.332292 9.223264
## attr(,"conf.level")
## [1] 0.99
95% confidence interval for variance. R has no built-in function for the confidence interval of a variance, so must we compute it using the formula in the book:
df <- length(stalkie$eyespan) - 1
varStalkie <- var(stalkie$eyespan)
lower = varStalkie * df / qchisq(0.05/2, df, lower.tail = FALSE)
upper = varStalkie * df / qchisq(1 - 0.05/2, df, lower.tail = FALSE)
c(lower = lower, variance = varStalkie, upper = upper)
## lower variance upper
## 0.07238029 0.15864444 0.58225336
95% confidence interval for standard deviation. Calculated from the confidence interval of the variance, which we just calculated above.
c(lower = sqrt(lower), std.dev = sqrt(varStalkie), upper = sqrt(upper))
## lower std.dev upper
## 0.2690359 0.3983020 0.7630553
Uses a one-sample \(t\)-test to compare body temperature in a random sample of people with the “expected” temperature 98.6\(^\circ\)F.
Read and inspect the data.
heat <- read.csv(url("http://whitlockschluter.zoology.ubc.ca/wp-content/data/chapter11/chap11e3Temperature.csv"))
head(heat)
## individual temperature
## 1 1 98.4
## 2 2 98.6
## 3 3 97.8
## 4 4 98.8
## 5 5 97.9
## 6 6 99.0
Histogram with options.
hist(heat$temperature, right = FALSE, breaks = seq(97, 100.5, by = 0.5),
col = "firebrick", las = 1, xlab = "Body temperature (degrees F)",
ylab = "Frequency", main = "")
One-sample \(t\)-test can be calculate using t.test
. The mu
argument gives the value stated in the null hypothesis.
t.test(heat$temperature, mu = 98.6)
##
## One Sample t-test
##
## data: heat$temperature
## t = -0.56065, df = 24, p-value = 0.5802
## alternative hypothesis: true mean is not equal to 98.6
## 95 percent confidence interval:
## 98.24422 98.80378
## sample estimates:
## mean of x
## 98.524