Lab 03: Data Visualization and Summarization
Markdown Author: Jessie Bell, 2023
Libraries Used: none
Answers: green
caffeineData <- read.csv("caffeine.csv")
level_1 <- caffeineData$exer[caffeineData$caff == 'yes']
level_2 <- caffeineData$exer[caffeineData$caff == 'no']If you are not able to get the data to read into your environment, double check your working directory under Session.
—————————————————————————————————————————-
1. Create a histogram
The histogram we created below is skewed to the right (positive).
hist(level_1, col="#c77cff")
—————————————————————————————————————————-
2. Create a stripchart
a <- caffeineData$caff #this pulls out the caffeine column in the caffeineData
b <- caffeineData$exer #this pulls out the exercise column in the caffeineData
stripchart(b~a, vertical=T, col="#00bfc4") #you can also do this the way that is shown in the lab handout. 
—————————————————————————————————————————-
3. Create a boxplot
boxplot(b ~ a, vertical=T, col="#e68613")
—————————————————————————————————————————-
4. Create a dot and whisker plot.
The code that you were given to complete this task did not work. Graph 4 will not count against you.
# the code for this given in the lab did not work. This problem will not be counted against you.
boxplot(caffeineData$exer ~ caffeineData$caff, col="#ff68a1")
—————————————————————————————————————————-
5. Create a barplot
The code that you were given to complete this task did not work. Graph 5 will not count against you. Some of you chose to use your island data to create bars though, and there is an example of how to do it using tapply() function below.
—————————————————————————————————————————-
Jessie practicing tapply() function, NOT GRADED
islandData <- read.csv("island.csv")
algalDensity <- islandData$AlgalDensity
face <- islandData$Face
island <- islandData$Island
tapply(algalDensity, face, FUN = "mean") #This is actually increadibly useful! It allows you to create a quick table where you call out the column in the data you would like to summarize, then you call out how to separate/categorize the data, then you tell FUN what function you would like to apply to the data## North South West
## 47.43733 42.11800 42.23533tapply(algalDensity, face, FUN="sd") #tells you the SD really quick.## North South West
## 4.288589 8.727338 8.173183barplot(tapply(algalDensity, island, FUN = "mean"), col="#8494ff")
—————————————————————————–
Read in the finch.csv file
finchData <- read.csv("kenyanfinches.csv")—————————————————————————————————————————-
6. Data Types
6a & 6b. What are the columns named? What type of data are these?
a. See data by using head() or other function.
b. The columns are named species is categorical and nominal data, mass is numerical continuous data, and beaklength is also numerical continuous data.
head(finchData)#tells you headers## species mass beaklength
## 1 WB.SPARW 40 10.6
## 2 WB.SPARW 43 10.8
## 3 WB.SPARW 37 10.9
## 4 WB.SPARW 38 11.3
## 5 WB.SPARW 43 10.9
## 6 WB.SPARW 33 10.1summary(finchData) #tells you data type, species is listed as character data. This is categorical, mass and beaklength are numerical. ## species mass beaklength
## Length:45 Min. : 6.00 Min. : 6.700
## Class :character 1st Qu.: 8.00 1st Qu.: 7.600
## Mode :character Median :16.00 Median : 8.000
## Mean :20.44 Mean : 8.744
## 3rd Qu.:36.00 3rd Qu.:10.700
## Max. :43.00 Max. :11.400—————————————————————————————————————————-
7. Separate
7a & 7b. What is each species summary statistic (mean, sd, and cv)?
I don’t know why these were separated into a and b, both a and b are asking for the same thing in my opinion. Please note that there is a long way to do this, and a short way. They are both done and labelled in the code below. There are most likely other ways to complete this task too.
#you can do this in many ways. I will try to show you two ways.
whitebrow <- subset(finchData, species=="WB.SPARW")
crimson <- subset(finchData, species=="CRU.WAXB")
cutthroat <- subset(finchData, species=="CUTTHROA")
#THE LONG WAY TO DO SUMMARY STATS
whitebrow_mean_mass <- mean(whitebrow$mass, na.rm = F)
whitebrow_mean_beak <- mean(whitebrow$beaklength, na.rm = F)
whitebrow_sd_mass <- sd(whitebrow$mass, na.rm = F)
whitebrow_sd_beak <- sd(whitebrow$beaklength, na.rm = F)
crimson_mean_mass <- mean(crimson$mass, na.rm = F)
crimson_mean_beak <- mean(crimson$beaklength, na.rm = F)
crimson_sd_mass <- sd(crimson$mass, na.rm = F)
crimson_sd_beak <- sd(crimson$beaklength, na.rm = F)
cutthroat_mean_mass <- mean(cutthroat$mass, na.rm = F)
cutthroat_mean_beak <- mean(cutthroat$beaklength, na.rm = F)
cutthroat_sd_mass <- sd(cutthroat$mass, na.rm = F)
cutthroat_sd_beak <- sd(cutthroat$beaklength, na.rm = F)
#THE SHORTER WAY
mean_mass <- tapply(finchData$mass, finchData$species, FUN="mean")
sd_mass <- tapply(finchData$mass, finchData$species, FUN="sd")
cv_mass <- sd_mass/mean_mass #this is the coefficient of variation for finch mass
summarystatstable <- cbind(mean_mass, sd_mass, cv_mass) #bind your tapply data together from above using cbind() function
summarystatstable #print out your table. ## mean_mass sd_mass cv_mass
## CRU.WAXB 7.529412 0.6242643 0.08291010
## CUTTHROA 15.416667 1.2401124 0.08043972
## WB.SPARW 37.937500 3.1084562 0.08193624Summary stats for beak length not provided, but this could be done the same way.
—————————————————————————————————————————-
8. Histograms
8a & 8b. Describe the distributions above. Are there outliers?
Whitebrow finch mass has a right skew (toward positive). Outliers are 1.5 x IQR, and they are hard to see unless you run a boxplot or calculate IQR. We made boxplots in problem 9 and you can see how to calculate outliers below, no outliers shown or calculated.
hist(whitebrow$mass, breaks = "sturges", main = "Distribution of whitebrow finch mass", xlab = "mass", col="#00a9ff")
Crimson finch mass has a left skew (toward negative). Outliers not present.
hist(crimson$mass, main = "Distribution of crimson finch mass", xlab = "mass", col="#f8766d")
Cutthroat finch mass has a bit of a left skew again. That left skew is actually an outlier.
hist(cutthroat$mass, main = "Distribution of cutthroat finch mass", xlab = "mass", col="#aba300")
Please note: IQR can be found by determining the median of the data. Then find the median of both sides of that median. This is your Interquartile Range. Outliers are determined by mutiplying IQR by 1.5. See calculation below.
IQR_whitebrow <- IQR(whitebrow$mass) * 1.5
# the range is on both sides of the median. So the upper range is
median(whitebrow$mass)+IQR_whitebrow #are there datapoints outside of this range? If yes ,its an outlier. If no, no outliers. ## [1] 43.375# now lets look at teh lower quartile
median(whitebrow$mass)-IQR_whitebrow #are there any data points for whitebrow mass that lay outside of this range? if so, its an outlier!## [1] 30.625IQR_cutthroat <- IQR(cutthroat$mass)*1.5
median(cutthroat$mass)+IQR_cutthroat## [1] 17.5median(cutthroat$mass)-IQR_cutthroat## [1] 14.5# there are 3 values of 7 in the mass column for cutthroat finches. These are all outliers. —————————————————————————————————————————-
9. Boxplots
Descriptions of the data should include mean or median, standard deviation or IQR, and outliers if they are present.
boxplot(whitebrow$mass, col="#00a9ff")
boxplot(crimson$mass, col="#f8766d")
boxplot(cutthroat$mass, col="#aba300")