S-040 Homework Assignment 3.Rmd

#---
#title: "R Notebook"
#output: html_notebook
#---

install.packages('readstata13') 
library(readstata13)
library(ggplot2)
library(foreign)
library(texreg)

#Read all the packages required to read data (readstata13, foreign) and visualize the inputted data(ggplot2)

#Research Question - Is the association between state achievement scores and college attendance statistically significant?”

rm(list = ls())
madese <- read.dta('https://www.dropbox.com/s/6jes28kc3yuo455/madese.dta?dl=1')
#Clear the environment variables and read the madese data from dataset stored in dropbox to the R environment

#### Testing the relationship between predictor and outcome
madese$score <- (madese$score_math + madese$score_ela)/2
summary(madese$score)
sd(madese$score, na.rm=TRUE)
ggplot(madese, aes(x = score )) + geom_histogram(binwidth = 2) + xlab('Measure of MCAS performance across school districts')
#Measured the center (Mean, Median through Summary()),Spread(Standard deviation through sd() and Interquartile range through Summary()) of MCAS scores across school districts in Massachussets and visualized it via histogram

summary(madese$college_pct)
sd(madese$college_pct, na.rm=TRUE)
ggplot(madese, aes(x = college_pct )) + geom_histogram() + xlab('Percentage of students planning to enrol in College across school districts')
across school districts')
#Measured the center (Mean, Median through Summary()),Spread(Standard deviation through sd() and Interquartile range through Summary()) of college enrolment % across school districts in Massachussets and visualized it via histogram

ggplot(madese, aes(x = score , y = college_pct)) + geom_point() + xlab('Measure of MCAS performance across school districts') + ylab('Percentage of students planning to enrol in College across school districts')
#Plotted a scatterplot through ggplot() to visualize the relationship between college enrolment % and MCAS scores across school districts in Massachussets

regmodel = lm(college_pct ~ score, data = madese, na.action = 'na.exclude')
#Created a statistical model by performing a regression of college percentage on district average MCAS scores from madese dataset
 
regmodelsummary =  summary(lm(college_pct ~ score, data = madese, na.action = 'na.exclude'))
#Summary statistic to explain the regression parameters including the confidence interval
confint(regmodel)

madese[c('ci_lower', 'ci_upper')] <- predict(regmodel, interval = 'confidence')[, 2:3]
# get predicted values of the confidnce interval from the model and cheeck the CI for the below mentioned districts
#Westbridgewater - 384
#Holyoke - 167
#Lexington - 187