# load STAR data
star_df = read.csv("star_data.csv")
star_df = star_df[complete.cases(star_df) , ]

# regress math on school but tabulate school first!
table(star_df$school)

lm( math ~ school, data = star_df)

# predicted value of someone in rural school?
536.34 + 25.43

# predicted value of someone in inner-city school?
536.34

prop.table(table(star_df$lunch, star_df$school))

# add lunch
lm( math ~ school + lunch, data = star_df)

# regress math on star
lm( math ~ star, data = star_df)

m = lm( math ~ star + gender + ethnicity +
        lunch + degree + experience +
        school, data = star_df)

summary(m)
# R2 = 0.086

lm( math ~ gender + experience, star_df)