load("lesson1_education.RData")

Introduction

In this data exploration, I visualize Baltimore City Public School graduation and dropout rates by race for the classes of 2016-17. District-wide data was retreived from https://www.baltimorecityschools.org/Page/24385.

Here is a sample of what the data looks like:

head(data)
##    School.Number                                      School.Name     CLN
## 3            178             Excel Academy @ Francis M. Wood High      12
## 7            239 Benjamin Franklin High School at Masonville Cove      10
## 8            239 Benjamin Franklin High School at Masonville Cove      10
## 9            239 Benjamin Franklin High School at Masonville Cove      10
## 13           307                                 Claremont School       8
## 15           325       ConneXions: A  Community Based Arts School Charter
##    Cohort Class.of Category X4..or.5.Year.Outcome
## 3    2013     2016    Black                5-Year
## 7    2013     2016    Black                5-Year
## 8    2013     2016   Latino                5-Year
## 9    2013     2016    White                5-Year
## 13   2013     2016    Black                5-Year
## 15   2013     2016    Black                5-Year
##    Number.of.Students.in.Cohort Number.of.Graduates Graduation.Rate
## 3                           200                  74            37.0
## 7                            54                  33            61.1
## 8                            28                  13            46.4
## 9                            28                  13            46.4
## 13                           10                   0             0.0
## 15                           50                  41            82.0
##    Number.of.Dropouts Dropout.Rate
## 3                  81         40.5
## 7                  20         37.0
## 8                  10         35.7
## 9                  15         53.6
## 13                  1         10.0
## 15                  1          2.0

In this overview, I visualize graduation and dropout rates for the three dominant race categories: Black or African American, Hispanic/Latino of any race, and White. In future installments I will statistically test some of the hypotheses I draw here based on visual inspection of the data.

Graduation Rates

Below we see the distribution of 4- and 5-year graduation rates for Black, Latino and White students in the classes of 2016-17 across all City Schools. While I haven’t yet tested for signifance, it is visually apparent that Latino students have the lowest rates of graduation.

multiplot(grad4_boxplot, grad5_boxplot, cols = 2)

Another way to visualize this data is to draw lines connecting the students from the same schools together, so that we can see how Black, Latino and White students compare within each school. Below I have plotted the 4- and 5-year graduation rates for Black, Latino and White students grouped together by school. It appears that there is variation in the relationship between race and graduation rate across schools. Based on apparent trends in this data, one hypothesis that could be tested is that Black and Latino students are more likely than their White counterparts to continue on to a 5th year of school given that they failed to graduate in their 4th year.

multiplot(grad4_lineplot, grad5_lineplot, cols = 2)

Dropout Rates

Below we see dropout rates for Black, Latino and White students. It appears that on average, White and Latino students are more likely to drop out of school than Black students.

multiplot(dropout4_boxplot, dropout5_boxplot, cols = 2)

Plotting dropout rates grouped by schools, we see that in almost all of the schools, the 4-year dropout rate for Latino students is higher than for Black students. In 9 out of the 12 schools that have White students, White students have the highest 4-year dropout rate of any group. However, the average dropout rate for schools that do not have White students appears to be higher than that of schools that do. Across the board, Black students most often have the lowest 4- and 5-year dropout rates in their schools.

multiplot(dropout4_lineplot, dropout5_lineplot, cols = 2)

Note: For the sake of visualization, one data point with a 100% dropout rate was omitted from the plot above; the now closed Heritage High School had a cohort of 11 Black students, all of whom dropped out in 2016.

Conclusions

While I haven’t performed any statistical tests yet, I have a few hypotheses that I will test in the next installments based on my observations:

  1. Latino students have significantly lower rates of graduation than Black or White students.
  2. Black and Latino students are more likely than their White counterparts to continue on to a 5th year of school given that they failed to graduate in their 4th year.
  3. The 4-year dropout rate for Latino students is higher than for Black students.
  4. In schools that have White students, the White students have significantly higher 4-year dropout rates than their Black and Latino peers.
  5. The average dropout rate in segregated Black/Latino schools is significantly higher than in schools with Black, Latino and White students.
  6. Black students have significantly lower 4- and 5-year dropout rates than other students in their schools.

Did you see anything interesting in the data that I may have missed? Do you have hypotheses that you think I should test? Send me an email at sbluher@gmail.com!