Notice the echo=FALSE, message=FALSE here. If you want to hide your R code you can do that by adding those inside the {}. We are going to load some packages we need for making nice looking histograms.

We are going to be working with the addhealth data so let’s attach it.

attach(addhealth)

The first thing we are going to do is to create some new variables where the hours over 50 are rounded down to 50. If you want to round to a different number change the 50 to that number.
In English the first line of code means: If tvhrs is greater than 50, set the new variable to 50, otherwise set it to tvhrs.

tvhrs_rounded<-ifelse (tvhrs > 50, 50, tvhrs)
radiohrs_rounded<-ifelse (radiohrs > 50, 50, radiohrs)
compgamehrs_rounded<-ifelse (compgamehrs > 50, 50, compgamehrs)

Let’s get summary statistics for the original variables and for the new variables.

summary(tvhrs)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##    0.00    5.00   10.00   16.03   21.00   99.00

summary(tvhrs_rounded)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##    0.00    5.00   10.00   15.45   21.00   50.00

Now let’s get the distribution by race and ethnicity as coded by addhealth

aggregate(tvhrs, by=list(raceeth), summary)

##              Group.1 x.Min. x.1st Qu. x.Median x.Mean x.3rd Qu. x.Max.
## 1    Latino/Hispanic   0.00      5.00    13.50  16.76     24.25  99.00
## 2              Asian   0.00      5.00    10.00  13.49     20.00  70.00
## 3    Native American   0.00      6.00    12.00  17.35     27.25  60.00
## 4 Non-Hispanic Black   0.00      7.00    15.00  20.98     30.00  99.00
## 5 Non-Hispanic White   0.00      5.00    10.00  14.23     20.00  99.00
## 6              Other   0.00      5.00    11.00  15.23     21.00  72.00

Now what if we create separate histograms for each race/ethnic group as coded by addhealth

ggplot(addhealth, aes(tvhrs)) + 
  geom_histogram(binwidth=1, fill="black") + 
  facet_wrap(~raceeth) +
  xlab("") +
  ylab("") + 
  ggtitle("Figure #")

Now we will get proportion distributions (known as density) for each group. Notice two changes. First, the addition of the aes(y=..density..) inside the geom_histogram. Second we use the scale_y_continuous with the option of percent to change the display proportions into percents. You can try the same graph without the scale option to see the difference.

There are many ways you can customize appearance of the graph. If you are interested in exploring you can search for information on ggplot.

ggplot(addhealth, aes(tvhrs)) + 
  geom_histogram(aes(y=..density..),binwidth=1, fill="black") +   
  facet_wrap(~raceeth) +
  scale_y_continuous(labels=percent) +
  xlab("") + 
  ylab("") + 
  ggtitle("Figure #") 

We can also plot the rounded data.

ggplot(addhealth, aes(tvhrs_rounded)) + 
  geom_histogram(aes(y=..density..),binwidth=1, fill="black") +   
  facet_wrap(~raceeth) +
  scale_y_continuous(labels=percent) +
  xlab("") + 
  ylab("") + 
  ggtitle("Figure #") 

Summarize the results here.

Now do the same kind of analysis but focused on radiohrs and compgamehrs.

1. Comparing Figures 2-4, please describe 2-3 MOST interesting and/or important differences in how members of these racial/ethnic groups spend their leisure time consuming media. (Note: consider what makes the most coherent and interesting story)
2. Why did you choose those 2-3 differences? (Note: your reasons should be based on the empirical information, not personal preferences.)