Instructions.
Use R Markdown to create an html document with the homework tasks.
As always, any plots should have appropriate axis and overall labels.
1 Data
Data for this HW assignment come from a randomized experiment to study the efficacy
of acupuncture for treating headaches. Results of the trial were published in the British
Medical Journal in 2004. You may view the paper at the following link: http://www.
bmj.com/content/328/7442/744.full. The data set includes 301 cases, 140 control (no
acupuncture) and 161 treated (acupuncture). Participants were randomly assigned to groups.
Variable names and descriptions are as follows:
• age; age in years
• sex; male = 0, female = 1
• migraine; diagnosis of migraines = 1, diagnosis of tension-type headaches = 0
• chronicity; number of years of headache disorder at baseline
• acupuncturist; ID for acupuncture provider
• group; acupuncture treatment group = 1, control group = 0
• pk1; headache severity rating at baseline
• pk5; headache severity rating 1 year later
Homework problems:
Import the data using read_csv() and call it acu. Note that the data have a header
row.
Use base R to calculate the mean and sd for each variable in the data set using apply().
Use the by() function to calculate the mean pk5 score by treatment group. Are you
comfortable making a conclusion about the efficacy of acupuncture based on these
means? Why or why not?
Use base R to create a new variable called diff by calculating the difference score
(post – pre) for headache severity rating.
Use base R to create a new variable called remission that is a 1 if a person’s score
dropped by 10 points or more and a 0 if not.
Usetidyversefunctionstoreplicate(3)-(5),callingthevariablesdiff2andremission2.
Create a scatterplot of baseline headache severity rating (horizontal axis) vs 1-year
headache severity rating (vertical axis) using base R.
Use ggplot2 to replicate (7).
In either base R or ggplot2, take your pick, incorporate information on treatment
group and sex into the plot. Consider using color, point character, point size, or other
graphical elements. Be sure to include a legend.
Try to replicate what you did in (9) with the other plotting platform. I.e., if you used
base R, now try to replicate using ggplot (or vice versa).