Scribd
Upload
80 views9 pages

"Cps - TXT" "Education" "South" "SEX" "Experience" "Union" "WAGE" "AGE" "RACE" "Occupat Ion" "Sector" "MARR"

This document demonstrates using R to analyze a dataset called cps. It loads packages, prepares the data, and answers several questions involving exploring and visualizing relationships in the data. For question 1, it selects columns, calculates correlations, filters rows, and calculates means. For question 2, it creates scatter plots and density plots of variables, recodes factors, and makes box plots. Question 3 involves applying functions like mean, sd, and split to randomly generated data and calculating statistics within groups.

Uploaded by

Alper Tamay Arslan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
80 views9 pages

"Cps - TXT" "Education" "South" "SEX" "Experience" "Union" "WAGE" "AGE" "RACE" "Occupat Ion" "Sector" "MARR"

This document demonstrates using R to analyze a dataset called cps. It loads packages, prepares the data, and answers several questions involving exploring and visualizing relationships in the data. For question 1, it selects columns, calculates correlations, filters rows, and calculates means. For question 2, it creates scatter plots and density plots of variables, recodes factors, and makes box plots. Question 3 involves applying functions like mean, sd, and split to randomly generated data and calculating statistics within groups.

Uploaded by

Alper Tamay Arslan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 9

library(magrittr)

library(dplyr)

##
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':


##
## filter, lag

## The following objects are masked from 'package:base':


##
## intersect, setdiff, setequal, union

cps = read.delim(file="cps.txt", header = FALSE)


colnames(cps)=
c("EDUCATION","SOUTH","SEX","EXPERIENCE","UNION","WAGE","AGE","RACE","OCCUPAT
ION","SECTOR","MARR")

#QUESTION 1
#PART A

cps1=data.frame()
cps1= cps %>% select ("SEX","EXPERIENCE","WAGE","AGE","MARR")

#PART B
cps1 %>% cor(cps1)

## SEX EXPERIENCE WAGE AGE MARR


## SEX 1.00000000 0.07522998 -0.20537055 0.07917859 0.01122521
## EXPERIENCE 0.07522998 1.00000000 0.08705953 0.97796125 0.27089961
## WAGE -0.20537055 0.08705953 1.00000000 0.17696688 0.10057887
## AGE 0.07917859 0.97796125 0.17696688 1.00000000 0.27894727
## MARR 0.01122521 0.27089961 0.10057887 0.27894727 1.00000000

#PART C
cps2=data.frame()
cps2= cps %>% select(-"SOUTH",-"UNION",-"MARR") %>% filter(between(AGE, 30,
50), SECTOR==2)

I used the select function to drop the columns and also filter function for subseting the
dataframe.
cps %>% mutate(New_Column = WAGE / AGE) %>% filter(New_Column>0.25) %>% count

## n
## 1 216

Mutate function is for combining two variables and generating a new variable. İ could not
solve how to print the output of count.
#PART E
x=cps %>% filter(SEX==0)
tapply(x$WAGE , x$OCCUPATION , mean)

## 1 2 3 4 5 6
## 13.721765 9.495714 7.489048 7.226471 12.773962 9.068175

#or
cps %>%group_by(OCCUPATION)%>% filter(SEX==0) %>% summarise(mean(WAGE))

## `summarise()` ungrouping output (override with `.groups` argument)

## # A tibble: 6 x 2
## OCCUPATION `mean(WAGE)`
## <int> <dbl>
## 1 1 13.7
## 2 2 9.50
## 3 3 7.49
## 4 4 7.23
## 5 5 12.8
## 6 6 9.07

There are 2 solutions of this question. The difference is my first answer is including tapply
function which is really good to appyling 1 function to 2 variables.
#PART F
#cps %>% table(cps$MARR,cps$SEX)

In thıs question, in the begining the code was working but, i could not understand why it is
not working now. It says lenghts must be same. I checked the lenghts of them but its not the
problem. In addition the table functions is generating a contingency table for 2 variables.
#QUESTION 2
#PART A

library(ggplot2)
ggplot(cps, aes(x = AGE, y = WAGE)) +
geom_point(aes(color = factor(SEX))) + scale_color_brewer(palette =
"Dark2") + labs (title="Distribution of age and wage by sex")
It representing the distribution of age and wage accordingly to sex variable. I used the
factor function to convert continous values to discrete values.
ggplot(cps, aes(WAGE)) +
geom_density(aes(color = factor(RACE)),size=2,linetype="dashed") + labs
(title="The density plot of the wage by race")
The graph shows the density plot of the wage by race.
cps$RACE[cps$RACE=="1"]="Other"
cps$RACE[cps$RACE=="2"]="Hispanic"
cps$RACE[cps$RACE=="3"]="White"
cps$OCCUPATION[cps$OCCUPATION=="1"]="Management"
cps$OCCUPATION[cps$OCCUPATION=="2"]="Sales"
cps$OCCUPATION[cps$OCCUPATION=="3"]="Clerical"
cps$OCCUPATION[cps$OCCUPATION=="4"]="Service"
cps$OCCUPATION[cps$OCCUPATION=="5"]="Professional"
cps$OCCUPATION[cps$OCCUPATION=="6"]="Other"
cps$RACE = factor(cps$RACE , levels=c("Other", "Hispanic", "White"))
cps$OCCUPATION= factor(cps$OCCUPATION,
levels=c("Management","Sales","Clerical","Service","Professional","Other"))

ggplot(cps, aes(x = RACE, y = WAGE ,fill =RACE)) +


geom_boxplot()+
facet_wrap(~ OCCUPATION, ncol = 3) +ylab("WAGE ($/hour)")
I noticed that i should have assigned the names to values while working on this question. At
the begining it was not the same order as in the homework pdf, so i listed them again to get
same order.
#PART D
male_hist=ggplot(cps[cps$SEX==0,], aes(x=AGE ))+
geom_histogram(fill="blue",bins=200) + labs(title="Ages of Males")
male_hist
female_hist=ggplot(cps[cps$SEX==1,], aes(x=AGE ))+
geom_histogram(fill="pink",bins=200) + labs(title="Ages of Females")
female_hist
gridExtra::grid.arrange(male_hist,female_hist,ncol=2)

Ages of males and ages of females in the same window.


#PART E
library(ggmosaic)

## Warning: package 'ggmosaic' was built under R version 4.0.5

ggplot(data = cps) +
geom_mosaic(aes(x = product(RACE), fill = OCCUPATION)) +
labs(title="Relationship between occupation and race")
Mosaic plot of relationship between occupation and race variables.
#QUESTION 3
#PART A
set.seed(124)
X = rnorm(1000)
Y = rnorm(50,10,2)
Z = runif(200,-5,20)
obj1=list(X,Y,Z)
lapply(obj1,mean)

## [[1]]
## [1] -0.0653552
##
## [[2]]
## [1] 10.58706
##
## [[3]]
## [1] 6.639656

Rnorm for normal dist. and runif for uniform dist.


#PART B
obj2=matrix(X,nrow=50,ncol=20)
apply(obj2, 2, sd)
## [1] 0.7978615 0.9672100 0.9389446 0.8773311 1.1407254 0.9078230 1.0698188
## [8] 0.9932702 0.9363832 0.9563979 0.8518203 0.8733509 1.0939566 0.9817683
## [15] 1.0541420 1.1166804 0.9214657 1.1430706 0.9734228 0.9550968

#PART C
h=LETTERS[1:4]
g=rep(h,50)
obj3=data.frame(Z,g)
mean(obj3$Z[which(obj3$g=="A")])

## [1] 6.839135

mean(obj3$Z[which(obj3$g=="B")])

## [1] 6.986212

mean(obj3$Z[which(obj3$g=="C")])

## [1] 5.560432

mean(obj3$Z[which(obj3$g=="D")])

## [1] 7.172845

“tapply(obj3Z , o b j3g["A"],mean)” I tried this but did not work.


#PART D

j=split(Y,gl(5,10))
matrix(lapply(j,max))

## [,1]
## [1,] 12.63274
## [2,] 13.20395
## [3,] 14.82519
## [4,] 15.30867
## [5,] 14.68656

matrix(lapply(j,min))

## [,1]
## [1,] 5.565061
## [2,] 9.320027
## [3,] 7.280937
## [4,] 4.97121
## [5,] 7.33229

You might also like