Analysis of data for the City School System, to aggregate and, find patterns and insights in standardised testing and school performance so that the superintendent and school district can make strategic decisions regarding budgeting and allotment at the district and school level.
We have been told that the students_complete.csv file shows evidence of academic dishonesty; specifically, reading and math grades for Thomas High School ninth graders appear to have been altered. So we will have to remove the grades for Thomas High School - ninth graders.
Maria, the City School Systems chief data scientist wants the analysis to include the following
- District Summary
- School Summary
- High and Low Performing Schools
- Math and Reading Scores by Grade
- Scores by School Spending¶
- Scores by School Size
- Scores by School Type
- Data Source:
- School Funding : schools_complete.csv
- Standardized Testing : students_complete.csv
- Software: Python 3.8.8, Pandas Dataframe
- Is the district summary affected?
There is no significant difference for the district summary for the average scores or percentages in the following tables.
- Is the school summary affected?
There was a significant change in Thomas High School summary after removing Thomas High School 9 grader scores but that was solved by substituting the agrregate scores and percentages for 10th, 11th and 12th graders in the Thomas High School Summary line. So the final school summary shows only the aggregate values for grades other than 9th for Thomas High School.
Thomas High School Summary after 9th grade was removed
- Does replacing the ninth graders’ math and reading scores affect Thomas High School’s performance relative to the other schools?
As you can see from the image below, Thomas High School was affected with a drop from 90% to 65% in the overall percentage. This affects the schools statistics significantly which inturn will affect a decrease in the school budget from the district and subsequently the district budget from the state. The analysis will also be wrong as only 10th, 11th and 12th graders for Thomas High School are to be included in the analysis. This was solved by substituting the agrregate scores and percentages for 10th, 11th and 12th graders in the Thomas High School Summary line.
- Does replacing the ninth-grade scores affect the following:
These were the scores of Thomas High School before replacing with agrregate scores and percentages for 10th, 11th and 12th graders in the Thomas High School Summary line. And these are the score after replacing with aggregate values.
-
Math and reading scores by grade : You see a big difference in the % Passing and % Overall Passing scores from ~90% to ~ 65%. These values will skew the results for making decisions.
-
Scores by school spending : Replacing with the aggregate scores affects the School spending range table. If you don't replace with aggregate scores, the calculation of Average and Overall scores will significantly lower for the $640-$644 range which is where Thomas High School falls under.
- Scores by school size : Similar to the above, the results will be skewed for the Medium(1000-2000)size schools which Thomas High School falls under if aggregate scores were not sustituted.
- Scores by school type : Thomas High School is a Charter School and the scores for the calculation of averages and overall percentages will show a significant drop for the Charter School category.
The School District and Superintendent can now review the results of the PyCitySchools analysis along with Maria, the chief data scientist and make informed decsions for budgeting and allotment.
By replacing the Thomas High School ninth grade scores with NaNs, we were able to eliminate the faulty scores from giving an accurate picture. However, we still had to modify the calculation for Thomas High School to take into account only the agrregate scores and percentages for 10th, 11th and 12th graders for its school summary.
The updated analysis shows the following
- NaN for all scores and percentages for Thomas High School - ninth grade
- Math and Reading pecentages are calculated ONLY for 10th, 11th and 12th graders not ninth graders for Thomas High School
- Scores by school type, size and spending show an accurate picture of the schools performance.
- The School and District summary are not skewed due to dropping the Thomas High School - ninth-grader scores and represent the rest of the schools performance accurately.
- Top 5 Schools are:
- Bottom 5 Schools are: