Lab 2: Displaying & Describing Data
Lab 2 Outline
1)Run code used to generate different kinds of graphs discussed in class
2)Re-run eagles code from Lab 1
3)Add new data to a state you’ve been assigned
4)Calculate summary stats for that state & fill out in-class worksheet
5)Create polished eagle graph
Based on Chapters 2 & 3, Whitlock and Schulter, 2nd Ed
Example data: Fisher’s Irises
Dataset made popular by R.A. Fisher
See https://en.wikipedia.org/wiki/Iris_flower_data_set for more info.
———————————————-
Loading data into R the easy way:
pre-made data in a “Package”
Getting data into R (or SAS, or ArcGIS…) can be a pain
R comes with many datasets that are pre-loaded into it
There are also many stat. techniques that can easily be added to R
These are contined in “packages”
Load data that is already in the “base” distribution of R
Fisher’s iris dat comes automaically with R. You can load it into memory using the command “data()”
#Load the iris data
data(iris)Look at the iris data
You can check that it was loaded using the ls() command (“list”).
ls()## [1] "iris"You can get info about the nature of the daframe using commands like dim()
dim(iris)## [1] 150 5This tells us that the iris data is essentially a spreadhshett that has 150 rows and 5 columns.
We can get the column names with names()
names(iris)## [1] "Sepal.Length" "Sepal.Width" "Petal.Length" "Petal.Width"
## [5] "Species"Note that the first letter of each word is capitalized. What are the implications of this?
The top of the data and the bottom of the data can be checked with head() and tail
#top of data
head(iris)## Sepal.Length Sepal.Width Petal.Length Petal.Width Species
## 1 5.1 3.5 1.4 0.2 setosa
## 2 4.9 3.0 1.4 0.2 setosa
## 3 4.7 3.2 1.3 0.2 setosa
## 4 4.6 3.1 1.5 0.2 setosa
## 5 5.0 3.6 1.4 0.2 setosa
## 6 5.4 3.9 1.7 0.4 setosa#bottom of data
tail(iris)## Sepal.Length Sepal.Width Petal.Length Petal.Width Species
## 145 6.7 3.3 5.7 2.5 virginica
## 146 6.7 3.0 5.2 2.3 virginica
## 147 6.3 2.5 5.0 1.9 virginica
## 148 6.5 3.0 5.2 2.0 virginica
## 149 6.2 3.4 5.4 2.3 virginica
## 150 5.9 3.0 5.1 1.8 virginicaAnother common R command is is(), which tells you what something is in R land
is(iris)## [1] "data.frame" "list" "oldClass" "vector"R might spew a lot of things out at you; usually the first item is most important. Here, it tells us that the “object” called iris in your workspace is first and foremost a “data.frame”, which is esseentially a spreadsheet of data loaded into R.
You can get basic info about the data themseleves using comamnds like summary()
summary(iris)## Sepal.Length Sepal.Width Petal.Length Petal.Width
## Min. :4.300 Min. :2.000 Min. :1.000 Min. :0.100
## 1st Qu.:5.100 1st Qu.:2.800 1st Qu.:1.600 1st Qu.:0.300
## Median :5.800 Median :3.000 Median :4.350 Median :1.300
## Mean :5.843 Mean :3.057 Mean :3.758 Mean :1.199
## 3rd Qu.:6.400 3rd Qu.:3.300 3rd Qu.:5.100 3rd Qu.:1.800
## Max. :7.900 Max. :4.400 Max. :6.900 Max. :2.500
## Species
## setosa :50
## versicolor:50
## virginica :50
##
##
## If you wanted information on just a single column, you would tell R to isolate that column like this
summary(iris$Sepal.Width)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 2.000 2.800 3.000 3.057 3.300 4.400That is, that name of the dataframe, a dollar sign ($), and the name of the column.
What happens when you don’t capitalize something?
#all lower case
# summary(iris$sepal.width) # this won't work
#just "s" in "sepal" lower case
# summary(iris$sepal.Width) #this won't work either
#or what if you capitalize "i" in "Iris"?
# summary(Iris$Sepal.Width) #won't work eitherThe 1st two error messages are not very informative; the 3rd one does make a little sense.
Load data that is in another R package part 1
Packages that come w/ R
Many scientists develop software for R, and they often include datasets to demonstrate how the software works. Some of this software, called a “package” comes with R already and just needs to be loaded. This is done with the library() command.
The MASS package comes with R when you download it.
#Load the MASS package
library(MASS)MASS contains a dataset called called “mammals”
data(mammals)You can confirm that the mammals data is in your workspace using ls()
ls()## [1] "iris" "mammals"You should now have the iris and the mammals data.
What is in the mammals dataset? Datasets actually usually have useful help files. Access help using the “?” function.
?mammals## starting httpd help server ...## doneThe help screen you pop up. It tells us that mammals is “A data frame with average brain and body weights for 62 species of land mammals.” At tghe bottom we can see that these data come from the paper
“Selected from: Allison, T. and Cicchetti, D. V. (1976) Sleep in mammals: ecological and constitutional correlates. Science 194, 732-734.”
We can learn about the mammals data usig the usual commands
dim(mammals)## [1] 62 2names(mammals)## [1] "body" "brain"head(mammals)## body brain
## Arctic fox 3.385 44.5
## Owl monkey 0.480 15.5
## Mountain beaver 1.350 8.1
## Cow 465.000 423.0
## Grey wolf 36.330 119.5
## Goat 27.660 115.0tail(mammals)## body brain
## Echidna 3.000 25.0
## Brazilian tapir 160.000 169.0
## Tenrec 0.900 2.6
## Phalanger 1.620 11.4
## Tree shrew 0.104 2.5
## Red fox 4.235 50.4summary(mammals)## body brain
## Min. : 0.005 Min. : 0.14
## 1st Qu.: 0.600 1st Qu.: 4.25
## Median : 3.342 Median : 17.25
## Mean : 198.790 Mean : 283.13
## 3rd Qu.: 48.203 3rd Qu.: 166.00
## Max. :6654.000 Max. :5712.00Load data that is in another R package part 2
Packages from CRAN
Most packages don’t come with R when you download it but are stored in a central site called CRAN. We’ll load the “doBy” package.
Loading packages using R-Studio
RStudio makes it easy to find and load packages.
* In the panel of RStudio that has the tabs “Plots”, “Packages”,“Help”, “Viewer” click on packages * Onthe next line it says “Install” and “Update”. Click on “Install” * A window will pop up. In the white field in the middle of the window under “Packages” type the name of the package you want. * RStudio will automatically bring up potential packages as you type. * Finish typing “doBy” or click on the name. * Click on the “Install” button. * In the source viewere some misc. test should show up. Most of the time this works. If it doesn’t, talk to the professor!
Loading packages using code
You can also use the install.packages() command to try to load the package. I already have the “doBy” download so I have “commented out” the code with a “#”. To run the code, remove the “#”. If you already followed the instructions above you don’t need to run the code.
#install.packages("doBy")What if you tell R to install a package you already have downloaded?
If you already have the package downloaded to your computer then a window will pop up asking you if you want to restart your computer. Normally this isn’t necessary; just click “no”. You might see a “warning” message pop up in the console such as “Warning in install.packages: package ‘doBy’ is in use and will not be installed”. This isn’t a problem
What if I can’t get a package I need loaded?
- Talk to someone who is good w/R (eg, your professor)
- Google something like “how to install R package” for general info
- Google something like “problem loading R package”
- Copy and paste any error message you might be getting into Google and see if anyone has written about this problem
Boxplots
Now we’re going to finally plot some data
R function: boxplot(…)
Very very easy to make boxplots in R
Very very very very hard to make in Excel
You should be making these all the time to explore your data
Also good way to present data in a final paper
Making a basic plot is faily easy. This basic structure is used frequently in R so we will discuss it in some detial.
#Load the iris data using the data() command
data(iris) #load iris data in case you haven't
#make a very basic boxplot
boxplot(Sepal.Length ~ Species, #the plotting formula of y ~ x
data = iris) #the data to plot
Anatomy of an R command
Key features of the comman are * the “~”" * the “comma” * the “data = iris” * capitalization
Box plot with labels
R will usually generate labels for the x and y axes based on the command. These need to be changed.
“xlab =” sets the labels for the x-axis, “ylab” for the y axis. Note that these both occur inside the paretheses, and that the text for the labels goes in quotes. Forgetting the quotes will cause the code to fail.
boxplot(Sepal.Length ~ Species, #the plotting formula of y ~ x
data = iris, #the data to plot
xlab = "Iris Species", #Label for x axis
ylab = "Sepal Length (mm)" ) #Label for y axis, w/ units
Changing colors in R plots part 1
If we wanted we could change the color of the boxplots using the universal command “col =”. This code can be used to change the color of most types of plots in R. This doesn’t increase the information content of the figure but maybe makes it nicer to look at.
boxplot(Sepal.Length ~ Species, #the plotting formula of y ~ x
data = iris, #the data to plot
xlab = "Iris Species", #Label for x axis
ylab = "Sepal Length",
col = 3) 
Changing colors in R plots part 2
If we want we could set the color of each box different. This accents that each box is a differnet Iris species. The code “col = 2:4” tells R to use the sequence of colors “2, 3, 4”. I skip “1” b/c its black and will obscure the bar that indicates the median
#boxplot w/ each color different
boxplot(Sepal.Length ~ Species, #the plotting formula of y ~ x
data = iris, #the data to plot
xlab = "Iris Species", #Label for x axis
ylab = "Sepal Length",
col = 2:4) 
This next code uses a slight variant. I’ve reversed the order of the colorrs so you can see the difference. I’ve use “col = c(4,3,2)” which is longhand for “col = 4:2”
boxplot(Sepal.Length ~ Species, #the plotting formula of y ~ x
data = iris, #the data to plot
xlab = "Iris Species", #Label for x axis
ylab = "Sepal Length",
col = c(4,3,2)) 
We could also do this
boxplot(Sepal.Length ~ Species, #the plotting formula of y ~ x
data = iris, #the data to plot
xlab = "Iris Species", #Label for x axis
ylab = "Sepal Length",
col = c("red","green","blue")) 
The “c(…)” Function
This shows up a lot in R code.
can have numbers inside it, separate by a colon
c(2:4)
can have numbers in it, sep. by commas
c(4,3,2)
can have text in it, sep by commas AND w/ " "
c(“red”,“green”,“blue”)
Plotting Means w/ Error Bars
library(MASS)
data(mammals)
#The errbar() function used below is in the Hmisc package
library(Hmisc)## Loading required package: lattice## Loading required package: survival## Loading required package: Formula## Loading required package: ggplot2##
## Attaching package: 'Hmisc'## The following objects are masked from 'package:base':
##
## format.pval, round.POSIXt, trunc.POSIXt, unitsn <- dim(mammals)[1]
u.body <- mean(mammals$body)
u.brain <- mean(mammals$brain)
se.body <- sd(mammals$body)/sqrt(n)
se.brain <- sd(mammals$brain)/sqrt(n)
mam.df <- data.frame(body.part = c("body","brain"),
mean = c(u.body,u.brain),
SE = c(se.body,se.brain))
errbar(x = 1:2,
y = mam.df$mean,
yplus =mam.df$mean+ mam.df$SE,yminus = mam.df$mean-mam.df$SE,
xlab = "Body Part",ylab = "Mass",
xlim=c(0.5,2.5),
xaxt="n",
cex =3)
axis(side=1,at=1:2,labels=mam.df$body.part)
legend("topleft", legend = "Error bars = SE",bty = "n")
legend("bottomleft", legend = "n = 62 spp",bty = "n") 
Histograms of DISCRETE/CATEGORICAL data
Number of extinct birds from each Hawaiian island
-Shows the frequency of each category
-Error bars are not possible w/these data
-This type of plot is useful for general descriptions of data
#Data from Science paper by biogeographer...not Gaston...
Island <- c("Hawaii","Kauai","Lana","Maui","Molokai","Oahu")
Extinctions <- c(11,1,7,4,7,7)
i <- order(Extinctions,decreasing = T)
barplot(height = Extinctions[i], names.arg = Island[i],
col = 1:length(Island))
Histograms of CONTINUOUS data
Histograms are very useful when you
-explore datasets you are seeing for the 1st time
-display data that is skewed or oddly shaped
-Convey similar idea as boxplot
-vertical lines often added to show mean, median, etc
-made with “hist()”" in R
A basic histogram
This plot shows the distribution of birthweights from a set of babies born in a hospital.
library(MASS)
data(birthwt)
hist(birthwt$bwt)
Add labels
library(MASS)
hist(birthwt$bwt,
xlab = "Birthweight (grams)",
ylab = "Frequency")
Add reference lines using the abline() command
library(MASS)
hist(birthwt$bwt,
xlab = "Birthweight (grams)",
ylab = "Frequency")
abline(v = mean(birthwt$bwt), col = 2, lwd = 3,lty =3)
Add reference line for the size of my son
hist(birthwt$bwt,
xlab = "Birthweight (grams)",
ylab = "Frequency")
abline(v = mean(birthwt$bwt), col = 2, lwd = 3,lty =3) #first reference line for the mean
abline(v = 4422.525607500001, col = 3, lwd = 3,lty =1) #Dr. Brouwer's son jude
Two Continuous variables: Scatter Plots
Scatter plots are standard way of for plotting two continous variables against each other
Frequently used to visualize data for “regression” analysis
We frequently take the log of numerical data - more on this later
A basic plot
#"Mammal body-brain allometry"
library(MASS)
data(mammals)
plot(log(brain) ~ log(body),
data = mammals)
Add some labels
#"Mammal body-brain allometry"
library(MASS)
data(mammals)
plot(log(brain) ~ log(body),
data = mammals,
xlab = "Log of animal body mass",
ylab = "log(brain mass)")
Increase the size of the points wiht cex = 2
plot(log(brain) ~ log(body),
data = mammals,
cex = 2,
xlab = "Log of animal body mass",
ylab = "log(brain mass)")
Change the shape of the point being plotted w/ “pch = 2”
plot(log(brain) ~ log(body),
data = mammals,
cex = 2,
pch =2,
xlab = "Log of animal body mass",
ylab = "log(brain mass)")
Increase the thickness of the lines used to draw the points with “lwd=2”
plot(log(brain) ~ log(body),
data = mammals,
cex = 2,
pch =2,
lwd = 2,
xlab = "Log of animal body mass",
ylab = "log(brain mass)")
Trends over time
#install.packages("primer")
library(primer)## Loading required package: deSolve##
## Attaching package: 'deSolve'## The following object is masked from 'package:graphics':
##
## matplotdata(sparrows)
plot(Count ~ Year, data = sparrows, type = "b",
lwd = 2, pch = 2, xlab = "Year",ylab = "Sparrows on Route")
sparrow.lm <- lm(Count ~ Year, data = sparrows)
abline(sparrow.lm, col = 2, lwd = 3, lty = 2)
Song Sparrow (Melospiza melodia) counts in Darrtown, OH, USA. From USGUS Breeding Bird Survey (BBS)
Measures of variation
Misc. References
Websites www.biostat.wisc.edu/~kbroman/topten_worstgraphs/ www.americanscientist.org/issues/pub/population-growth-technology-and-tricky-graphs
Papers Wainer. 1984. How to Display Data Badly. Am. Statistician.