Yumin's submission

submissions/final_bootcamp_project.R

@@ -0,0 +1,43 @@
+# Load packages
+library(tidyverse)
+library(here)
+library(reshape2)
+library(dplyr)
+
+## read data files
+nys_school = read_csv(here("data/nys_schools.csv"))
+nys_acs = read_csv(here("data/nys_acs.csv"))
+
+summary(nys_school) #school has a lot of -99 minimums
+summary(nys_acs)
+
+## check missing data
+names(which(colSums(is.na(nys_school)) > 0))
+names(which(colSums(is.na(nys_acs)) > 0))
+
+## check negative values in nys_school dataset
+nrow(nys_school[nys_school$district_name<=0,]) ##1725
+
+## count number of na in column(s)
+sum(is.na(nys_school$district_name)) ##1706
+
+## omit NAs and -99s
+nys_school = na.omit(nys_school)
+nys_school = nys_school[nys_school$district_name >= 0, ]
+
+
+## create a categorical value that groups counties into high, medium and low
+nys_acs = nys_acs %>%
+  mutate(counties_group = case_when(county_per_poverty < quantile(county_per_poverty, 1/3) ~ 'low',
+                                    county_per_poverty >= quantile(county_per_poverty, 1/3) & county_per_poverty <= quantile(county_per_poverty, 2/3) ~ 'med',
+                                    county_per_poverty > quantile(county_per_poverty, 2/3) ~ 'high'))
+
+## merge datasets
+merged = merge(nys_school, nys_acs, by.nys_school = c("county_name","year"), by.nys_acs = c("name","year"))
+
+merged %>%
+  group_by("county_name") %>%
+  select("total_enroll","per_free_lunch","per_reduced_lunch","")
+
+ggplot(nys_school) + 
+  geom_point(aes(x=per_free_lunch,y=mean_ela_score))