Skip to content

Use of Pandas with Jupyter Notebook to aggregate data and showcase obvious trends in school performance. Does a bigger budget translate to better test scores?

Notifications You must be signed in to change notification settings

BrittneyOleniacz/Education_Budget_Breakdown

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

36 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Education Budget Breakdown

OBJECTIVE

Analyze the district-wide standardized test results to help the school board and mayor make strategic decisions regarding future school budgets and priorities. You'll be given access to every student's math and reading scores, as well as various information on the schools they attend.

Methods

  1. Read School and Student Data files and store into two seperate Pandas DataFrames and combine the data into a single dataset using merge function
  2. Calculate, the following, and convert into District Summary DataFrame
    • total number of schools
    • total number of students
    • total budget
    • average math score
    • average reading score
    • percentage of students with a passing math score (70 or greater)
    • percentage of students with a passing reading score (70 or greater)
    • percentage of students who passed math and reading (% Overall Passing)
  3. Create an overview table that summarizes key metrics about each school, including:
    • School Name
    • School Type
    • Total Students
    • Total School Budget
    • Per Student Budget
    • Average Math Score
    • Average Reading Score
    • % Passing Math
    • % Passing Reading
    • % Overall Passing (The percentage of students that passed math and reading.)
  4. Sort and display the top five performing schools by % overall passing.
  5. Sort and display the five worst-performing schools by % overall passing.
  6. Create tables that lists the average Reading Score and Math Scores for students of each grade level (9th, 10th, 11th, 12th) at each school by using pandas series for each grade by using a conditional statement, group each series by school, the combine the series into a dataframe.
  7. Create a table that breaks down school performances based on average Spending Ranges (Per Student) by using four bins (<$585, $585-614, $615-644, >$644) to group school spending and contains:
    • Average Math Score
    • Average Reading Score
    • % Passing Math
    • % Passing Reading
    • Overall Passing Rate (Average of the above two)
  8. Create a table that looks at school performances based on the School Size and grouped into bins: Small (<1000), Medium (1000-2000), Large (>2000)
    • Average Math Score
    • Average Reading Score
    • % Passing Math
    • % Passing Reading
    • Overall Passing Rate (Average of the above two)
  9. Create a table that analyzes the school performances based on the type of school (Charter vs. District) and contains:
    • Average Math Score
    • Average Reading Score
    • % Passing Math
    • % Passing Reading
    • Overall Passing Rate (Average of the above two)

Results

District Summary

district_summary

School Summary

school_summary

Top Performing Schools (By % Overall Passing)

top_performing_schools

Bottom Performing Schools (By % Overall Passing)

buttom_performing_schools

Math Scores by Grade

math_scores

Reading Scores by Grade

reading_scores

Scores by School Spending

scores_by_school_spending

Scores by School Size

scores_by_school_size

Scores by School Type

scores_by_school_type

Observations and Findings

1. Charter schools had higher test scores than district schools.

This trend is seen in the top and bottom performing school based on percent of students that passed overall. The top five performing are charter schools and the bottom five are district schools. This trend is corroborated by the scores by school type. While there's less than a six point different in the average math schools and about three point difference in the average reading score, significant differences are seen in percent passing math and the overall passing rate.

2. The quality of education is not dependent on the budget.

The budget for the top performing school ranged from $585,858.00 𝑡𝑜 $1,319,574.00 and the bottom performing schools had budgets between $1,884,411.00 𝑎𝑛𝑑 $3,094,650.00. The assumption was Johnson High School with a budget of $3,094,650.00 is a large school, thus the budget per student would be low. However, by examining the budget per student calculations, Johnson High School has one of the highest budgets per student. This trend is seen again in the test scores grouped by school spending. From this calculation, budget per student and overall passing are inversely related.

3. Size matters.

Schools with a smaller student population produce higher test scores. In general, schools with over 2,000 students had the lowest test scores.

4. Widespread aversion to math.

More student pass reading than math. Generally, reading scores are higher than math scores by looking over the scores by grade, but looking at the test scores by school type this seems to be a trend only in district schools, specifically those of medium to large size. This should be compared to state and national scores over time to detect trends. How is math taught differently at charter schools and how can it be implemented in district schools.

Recommendations and Suggestions

These observations suggest charter schools are of higher education quality and operate more efficiently than district schools. Charter schools with 1000-2000 students grossly outperform the district schools of the same size with larger budgets. I suggest to the Mayor and Schoolboard to examine, adapt, and implement the teaching methods and budget models used at charter schools. The Schoolboard can create working committees to investigate these trends and encourage more collaboration between teachers across the entire school district with a conference, teacher "swap", and peer mentors.

Further analysis

I suggest the additional analyses to provide a detailed, holistic view of the schools and their students to know how to best distribute funds to better serve students.

1. Detailed Budgets

More detailed analysis on charter and district school budgets is needed to compare how each type of school uses and manages their budget. Factors like the quality of food served in the cafetertia or the duration of the school day can impact student performace. We must consider transportation, books/supplies, technological upgrades and other resources. This should shine light on the priorities of the school districts and how the funds are used to ensure student success.

2. Instructional Staff

An analysis of instructional staff would also be helpful to look at the caliber of teaching, the educational level of teachers, the student-to-teacher ratio, teaching styles, and etc. Additionally, salaries of those employed by the school should be carefully analyzed. For instance, if teachers are paid more at charter schools, they will take more pride in their job and will be more attentive to student needs than underpaid instructors at district schools.

3. Student household demographics

I must request a thorough demographic analysis be performed of the student population across the district to look at household size, household income, geographic distribution, learning disability, mental health & wellness of household, education level of head of household, number of daily meals, parent involvement, etc. The quality of a student's homelife commonly factors into academic performance. We need to understand where our students are comming from to know how to best educate them and prepare them for success after graduation. There may be a need for educational outreach in particular communities and collaboration with social workers, career centers, GED courses, food banks, technical training, and other resources.

About

Use of Pandas with Jupyter Notebook to aggregate data and showcase obvious trends in school performance. Does a bigger budget translate to better test scores?

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published