PLSquared_DataAnalysis.rmd

---
title: "PLSquared- Basic Analysis"
author: "Arun Sharma"
date: "12/15/2019"
output: html_document
---
```{r global_options, include=TRUE}
#knitr::opts_chunk$set(echo=FALSE, warning=FALSE, message=FALSE)
```


```{r setup, include=FALSE}

# Loading libraries
library(tidyverse)
library(data.table)
library(ggplot2)

```

```{r}

#loading data for the year 17-18
Fall1718 <- fread("2018_NWEA/NWEA Fall 2017-hashed.csv")
Winter1718 <- fread("2018_NWEA/NWEA_Winter_2017-2018-hashed.csv")
Spring1718 <- fread("2018_NWEA/NWEA_Spring_2017-2018-hashed.csv")

#loading data for the year 19-18
Fall1819 <- fread("2019_NWEA/NWEA Fall 2018-hashed.csv")
Winter1819 <- fread("2019_NWEA/NWEA_Winter_2018-2019-hashed.csv")
Spring1819 <- fread("2019_NWEA/NWEA_Spring_2018-2019-hashed.csv")
```


```{r}
#merging both years to get thecommulative data

Complete_1718 <- rbind(Fall1718,Winter1718,Spring1718)
Complete_1819 <- rbind(Fall1819,Winter1819,Spring1819)

```

```{r}
#playing with data
#summary(Complete_1718)
```

```{r}
#Summarising the data by student performance

Summary_Student_1718 <- Complete_1718 %>%
  group_by(StudentID, TermName, Discipline) %>%
  summarise(count = n(), avg.RIT = mean(TestRITScore), percentile =  mean(TestPercentile))

head(Summary_Student_1718)

```

```{r}

Summary_school_1718 <- Complete_1718 %>%
  group_by(SchoolName, TestName) %>%
  summarise(count = n(), avg.RIT = mean(TestRITScore), percentile =  mean(TestPercentile))

Summary_school_1718
```

```{r, fig.width= 10, fig.height=10}
#removing the middle school, comparing only elementary schools, cleaning with little data
Elementaryschools_1718 <- Complete_1718[!Complete_1718$SchoolName %in% "Elizabeth Forward Middle School"&
                                        !Complete_1718$TestName %in% "Screening: Math 2-5 PA 2013" &
                                        !Complete_1718$TestName %in% "Screening: Reading 2-5 PA 2013" &
                                        !Complete_1718$TestName %in% "Growth: Language 2-12 PA 2013" ,]
Elementaryschools_1718 <-  Elementaryschools_1718 %>%
  filter(!is.na(TestRITScore))

#Summarising school wise

Box_plot_schools <- Elementaryschools_1718 %>%
  group_by(SchoolName) %>%
  ggplot(aes(x= SchoolName, y = TestRITScore, fill = SchoolName))+ 
  facet_wrap("TestName")+
  geom_boxplot() + 
  theme(axis.text.x = element_text(angle = 60, hjust =1, vjust =1)) 

Box_plot_schools

```

```{r, fig.width=10, fig.height=10}
Elementary.maths.2_5_1718 <- Complete_1718[Complete_1718$TestName %in% "Growth: Math 2-5 PA 2013",]
Elementary.maths.2_5_1718.summary <- Elementary.maths.2_5_1718 %>%
  group_by(SchoolName) %>%
  summarise( count = n(), avg.rit.score = mean (TestRITScore), lower = t.test(TestRITScore, mu =0)$conf.int[1], upper = t.test(TestRITScore, mu =0)$conf.int[2])

Elementary.maths.2_5_1718.summary %>%
  ggplot(aes( x= SchoolName, y= avg.rit.score, fill = SchoolName)) +
  geom_bar(stat = "identity") +
  geom_errorbar(aes(ymin = lower, ymax = upper),
                width =0.4,
                position = position_dodge(0.9)) +
  coord_cartesian(ylim = c(175,215)) +
  theme(axis.text.x = element_text(angle = 60, hjust =1, vjust =1)) +
  ggtitle("Growth : Math 2-5 PA 2013 Scores distribution by schools")

#anova for variation with in Math score across schools

aov(data = Elementary.maths.2_5_1718, TestRITScore ~ SchoolName)
  
```

```{r}
demographicInfo <- fread(input = "Studnet ID_IEP_Econ_Ethn_EF-hashed.csv")
names(demographicInfo)[1] <- names(Summary_Student_1718)[1]
Summary_Student_1718 <- merge(Summary_Student_1718, demographicInfo)



# combining the entire data with demographic characteristics
Complete_1718 <- merge(Complete_1718 , demographicInfo)
Complete_1819 <- merge(Complete_1819, demographicInfo)

```

```{r}
#distribution by schools

demographic.distribution <- Complete_1718 %>%
  count(SchoolName, Ethnicity) %>%
  mutate(prop = prop.table(n))


demographic.distribution %>%
  ggplot(aes( x= SchoolName, y = n, fill = Ethnicity)) +
  geom_bar(stat= "identity", position = "dodge") +
  theme(axis.text.x = element_text(angle = 60, hjust =1, vjust =1))
```

```{r, fig.width= 10, fig.height=10}

#Ethnicity wise RIT scores
Elementaryschools.demo_1718 <- Complete_1718[!Complete_1718$SchoolName %in% "Elizabeth Forward Middle School"&
                                        !Complete_1718$TestName %in% "Screening: Math 2-5 PA 2013" &
                                        !Complete_1718$TestName %in% "Screening: Reading 2-5 PA 2013" &
                                        !Complete_1718$TestName %in% "Growth: Language 2-12 PA 2013" ,]

boxplot.demo_1718 <- Elementaryschools.demo_1718 %>%
  group_by(SchoolName,TestName, Ethnicity) %>%
   ggplot( aes( x = Ethnicity, y = TestRITScore, fill = Ethnicity)) +
  geom_boxplot() +
  facet_wrap(vars(SchoolName, TestName)) +
    ggtitle("BoxPlots depicting average RIT scores for different races across Schools")
boxplot.demo_1718 

```
```{r, fig.width=10, fig.height=10}
Elementaryschools.demo_1718 <- mutate(
  Elementaryschools.demo_1718, EconDis = recode_factor(Elementaryschools.demo_1718$EconDis, `1` ="Not Ecnonmically Disadvantage", `0` = "Ecnonmically Disadvantage"))

boxplot.demo_1718 <- Elementaryschools.demo_1718 %>%
  group_by(SchoolName,TestName, EconDis) %>%
   ggplot( aes( x = EconDis, y = TestRITScore, fill = EconDis)) +
  geom_boxplot() +
  facet_wrap(vars(SchoolName, TestName)) +
    ggtitle("BoxPlots depicting average RIT scores for different Econmic Disadvantageous Kids across Schools")+
   theme(axis.text.x = element_text(angle = 60, hjust =1, vjust =1))

boxplot.demo_1718 

```