Step 1: Read in csv file (file is already on github so I did not save it there–hopefully the points were for accessing it there rather than just adding it. If not I can easily update).

One thing that I really appreciate about this is that prior to writing this code I would have to repeatedly change the path for finding my file depending on where I was working (at home I have a mac desktop, a mac laptop, and a PC laptop and at work I have a PC laptop).

My understanding is that relative paths of they type in Python aren’t relevant in R because everything is supposed to reference back to a working directory. I’m still trying to work out how to do that without uploading everything to github, which seems unwieldy.

mydata <- read.csv("https://raw.github.com/vincentarelbundock/Rdatasets/master/csv/Zelig/free1.csv",
                  header=TRUE)


Step 2: Convert to dataframe
freedata <- data.frame(mydata)


Step 3:Take a peek at what it looks like
head(freedata)
##        X sex age educ  country y v1 v2 v3 v4 v5 v6
## 1 109276   0  20    4  Eurasia 1  4  3  3  5  3  4
## 2  88178   1  25    4   Oceana 2  3  3  5  5  5  5
## 3 111063   1  56    2 Eastasia 2  3  2  4  5  5  4
## 4 161488   0  65    6 Eastasia 2  3  3  5  5  5  5
## 5  44532   1  50    5   Oceana 1  5  3  5  5  3  5
## 6  95503   0  20    5 Eastasia 1  4  4  3  3  4  4


Step 4:: Get a summary of the data


Please note the following coding definitions:

sex: 0 = female / 1 = male

educ: 1 = No formal education / 2 = Less than primary school education / 3 = Completed primary school / 4 = Completed secondary school / 5 = Completed high school / 6 = Completed college / 7 = Completed post-graduate degree 

summary(freedata)
##        X               sex              age             educ      
##  Min.   :   142   Min.   :0.0000   Min.   : 1.00   Min.   :1.000  
##  1st Qu.: 52723   1st Qu.:0.0000   1st Qu.:27.00   1st Qu.:1.000  
##  Median :108699   Median :1.0000   Median :39.00   Median :3.000  
##  Mean   : 90665   Mean   :0.5568   Mean   :40.74   Mean   :2.942  
##  3rd Qu.:119325   3rd Qu.:1.0000   3rd Qu.:52.00   3rd Qu.:4.000  
##  Max.   :171811   Max.   :1.0000   Max.   :90.00   Max.   :7.000  
##                   NA's   :1        NA's   :4       NA's   :5      
##      country          y              v1              v2       
##  Eastasia:150   Min.   :1.00   Min.   :1.000   Min.   :1.000  
##  Eurasia :150   1st Qu.:3.00   1st Qu.:2.000   1st Qu.:2.000  
##  Oceana  :150   Median :4.00   Median :3.000   Median :2.000  
##                 Mean   :3.52   Mean   :2.649   Mean   :2.536  
##                 3rd Qu.:5.00   3rd Qu.:3.000   3rd Qu.:3.000  
##                 Max.   :5.00   Max.   :5.000   Max.   :5.000  
##                                                               
##        v3              v4              v5              v6      
##  Min.   :1.000   Min.   :1.000   Min.   :1.000   Min.   :1.00  
##  1st Qu.:3.000   1st Qu.:3.000   1st Qu.:3.000   1st Qu.:4.00  
##  Median :4.000   Median :4.000   Median :4.000   Median :5.00  
##  Mean   :3.664   Mean   :4.084   Mean   :3.867   Mean   :4.38  
##  3rd Qu.:4.000   3rd Qu.:5.000   3rd Qu.:5.000   3rd Qu.:5.00  
##  Max.   :5.000   Max.   :5.000   Max.   :5.000   Max.   :5.00  
##


Step 5:Find the median of a couple of columns


don’t forget to check this against the summary

free_sex_med <- median(freedata$sex, na.rm = TRUE)

free_age_med <- median(freedata$age, na.rm = TRUE)

free_ed_med <- median(freedata$educ, na.rm = TRUE)


free_med <- c(free_sex_med, free_age_med, free_ed_med)

free_med
## [1]  1 39  3


Step6: Find the median of a few elements


don’t forget to check this against the summary

free_mean <- sapply(freedata[,2:4],mean, na.rm = TRUE)

free_mean
##        sex        age       educ 
##  0.5567929 40.7443946  2.9415730


Step7: Cut the original dataframe back to a few columns and rows
freecut <- subset(mydata, y < 5,
select=2:5)

head(freecut)
##   sex age educ  country
## 1   0  20    4  Eurasia
## 2   1  25    4   Oceana
## 3   1  56    2 Eastasia
## 4   0  65    6 Eastasia
## 5   1  50    5   Oceana
## 6   0  20    5 Eastasia


Step8: Change the names of the columns
names(freecut) <- c("gender", "oldness", "student_loaniness", "homeland")


Step9: Check the summary of the new dataframe
summary(freecut)
##      gender          oldness      student_loaniness     homeland  
##  Min.   :0.0000   Min.   : 1.00   Min.   :1.000     Eastasia:106  
##  1st Qu.:0.0000   1st Qu.:27.00   1st Qu.:2.000     Eurasia : 80  
##  Median :1.0000   Median :38.50   Median :3.000     Oceana  :130  
##  Mean   :0.5429   Mean   :41.25   Mean   :3.077                   
##  3rd Qu.:1.0000   3rd Qu.:53.00   3rd Qu.:4.000                   
##  Max.   :1.0000   Max.   :84.00   Max.   :7.000                   
##  NA's   :1        NA's   :2       NA's   :3


Step10: Compare the mean and median of the new DF to the original
fcut_mean <- sapply(freecut[,1:3], mean, na.rm = TRUE)

free_mean
##        sex        age       educ 
##  0.5567929 40.7443946  2.9415730
fcut_mean
##            gender           oldness student_loaniness 
##         0.5428571        41.2452229         3.0766773
fcut_mean - free_mean
##            gender           oldness student_loaniness 
##       -0.01393573        0.50082831        0.13510428

Gender mean for the new dataframe is slightly more female than the original, half a year older, and slightly more educated.


fcut_med <- sapply(freecut[,1:3], median, na.rm = TRUE)
fcut_med
##            gender           oldness student_loaniness 
##               1.0              38.5               3.0
free_med
## [1]  1 39  3
fcut_med
##            gender           oldness student_loaniness 
##               1.0              38.5               3.0
fcut_med - free_med
##            gender           oldness student_loaniness 
##               0.0              -0.5               0.0

As median tends to be a more stable metric, it’s not surprising that the medians for sex and education had no change. Age shifted slightly younger.

Step 11: The new data has different means and medians because the number of observations was cut back. If that hadn’t been done the numbers would match.


Take a look at what levels are in the homeland factor

levels(freecut$homeland)
## [1] "Eastasia" "Eurasia"  "Oceana"


Step 12: Change the levels and verify that they “stuck”
freecut$homeland <- as.character(freecut$homeland)
freecut$homeland[1:nrow(freecut)] <- c("East_Asia","Asiope","Nemoville")
freecut$homeland <- as.factor(freecut$homeland)

# require(plyr)
# 
# revalue(freecut$homeland, c("Eastasia" = "East_Asia", "Eurasia" = "Asiope", "Oceana" = "Nemoville"))

levels(freecut$homeland)
## [1] "Asiope"    "East_Asia" "Nemoville"


Step 13: Take a look at a few rows of the new DF
head(freecut)
##   gender oldness student_loaniness  homeland
## 1      0      20                 4 East_Asia
## 2      1      25                 4    Asiope
## 3      1      56                 2 Nemoville
## 4      0      65                 6 East_Asia
## 5      1      50                 5    Asiope
## 6      0      20                 5 Nemoville


Step 14: Save the new DF as a csv
library(rdrop2)
token<-drop_auth()

require(dropR)

write.csv(freecut, 'freecut.csv')
drop_upload('freecut.csv', path = "cuny_msds/bridge/R")