R Notes for Topic 23: Analyzing Paired Data
R notes for Activity 23-1: Marriage Ages
You can read the data with
MA = read.csv("/home/rebelsky/Stats115/Data/MarriageAges.csv")
23-1 a. A Two-Sample t-test
This is a one-sided t-test, which we can express
with
t.test(MA$HusbandAge,MA$WifeAge,alternative="greater")
23-1 b. Computing a Confidence Interval
You should be able to figure out how tocompute this confidence
interval based on The R Notes for
Topic 22.
You can produce a similar scatterplot using
plot(MA$WifeAge,MA$HusbandAge)
You can add the “same age” line with
abline(0,1)
23-1 e. A Confidence Intervals for Differences
We can use a one-sample t-test of the difference between
the two columns.
t.test(MA$HusbandAge-MA$WifeAge, mu=0, conf.level=.9)
Once again, we need to tell R that this is a one-sided
t-test.
t.test(MA$HusbandAge-MA$WifeAge, mu=0, alternative="greater")
23-1 i. Exploring Technical Conditions
We might want to make a normal probability plot to see if the data
are normal. I'm not sure why, but R gives a somewhat different plot
than does the answer key.
> qqnorm(MA$HusbandAge-MA$WifeAge, datax=T)
> qqline(MA$HusbandAge-MA$WifeAge, datax=T)
R notes for Activity 23-2: Melting times
If all goes well, there will be three files of melting time:
CChips.csv, BChips.csv,
and Chips.csv. You can load them with
CChips = read.csv("/home/rebelsky/Stats115/Data/CChips.csv")
BChips = read.csv("/home/rebelsky/Stats115/Data/BChips.csv")
Chips = read.csv("/home/rebelsky/Stats115/Data/Chips.csv")
You should be able to use your notes on earlier activities to compute the
appropriate summaries. Ask for help if you need it.
R notes for Activity 23-5: Muscle Fatigue
Although the book does not provide the data in electronic format,
I've put it into the file
MuscleFatigue.csv for you. You can load
it with
MF = read.csv("/home/rebelsky/Stats115/Data/MuscleFatigue.csv")
To compute the difference between men and women, you would use
MF$Man-MF$Woman
To build a dotplot for these data, you would use
library(BHH2, lib="/home/rebelsky/Stats115/Packages")
dotPlot(MF$Man-MF$Woman)
You should be able to figure out how to do the t-test
and confidence interval.
