School District Analysis
Purpose
Reanalyzing school district data with Python, Pandas and Jupyter Notebook. Assessing how the disclusion of a set of ninth grade test scores - those suspected of academic dishonesty - will affect the analysis of the particular school in question, as well as the overall analysis of the entire city school district as a whole.
Results
Several areas of the data were affected with the removal of ninth grade math and reading scores from Thomas High School, however, many of these changes were usually quite minor. The changes are as follows:
District Summary Results
(for references see images below - the top image references data without ninth grade scores; bottom with ninth grade scores)
- a decrease in Average Math Score in the district by 0.1 points (78.9 without grade nine scores; 79 with)
- a decrease in % of students passing math in the district by 0.2% (74.8% without grade nine scores; 75% with)
- a decrease in % of students passing reading in the district by 0.3% (85.7% without vs 86% with)
- a decrease in % of students passing overall in the district by 0.1% (64.9% vs 65%)
DataFrames displayed from PyCitySchools_Challenge.ipynb and PyCitySchools.ipynb respectively - see files for more info
School Summary Results
(for references see images below - the top image references data without ninth grade scores; bottom with ninth grade scores)
- a decrease in Average Math scores at Thomas High School by ~0.16 points (83.35 without vs 83.42 with)
- an increase in Average Reading scores at Thomas High School by 0.05 points (83.90 without vs 83.85 without)
- a decrease in % of students passing math at Thomas High School by ~ 0.08% (93.19% vs 93.27%)
- a decrease in % of students passing reading at Thomas High School by ~ 0.3% (97.02% vs 97.31%)
- a decrease in % of students passing overall at Thomas High School by 0.32% (90.63% vs 90.95%)
Relative Performance
- Although there is a ~ 0.3% decrease in % of students passing overall, Thomas High School remains to be the second highest performing school in the city school district based on overall passing % (See below).
Math & Reading Scores by Grade
- Math:
- No changes - Ninth grade scores simply displayed as NaN with no changes to other grades (See below).
- Reading
- Same as above; no changes to other grades; ninth grade scores are displayed as NaN (See below).
Scores by School Spending
- Very minor changes to scores when sorted into spending categories. These changes were seen in the $634-644 spending range category to which Thomas High School falls under. Changes are as follows (reference images below in usual order):
- 0.01 point decrease in Average Math Score
- 0.01 point increase in Average Reading Score
- 0.02% decrease in % of students passing math
- 0.08% decrease in % of students passing reading
- 0.08% decrease in % of students passing overall
- These changes, however, become insignificant once tables are formatted appropriately (See below).
Scores by School Size
- A set of minor changes are seen in scores categorized by school size, similar to those shown in school spending categories. With scores in the medium school-size category, changes are as follows (reference images below in usual order):
- 0.01 point decrease in Average Math Score
- 0.01 point increase in Average Reading Score
- 0.01% decrease in % of students passing math
- 0.06% decrease in % of students passing reading
- 0.07% decrease in % of students passing overall
- Once again, these changes become insignificant once tables are formatted appropriately (See below).
Scores by School Type
- Very minor changes in scores when categorized by school type - these changes were seen in the Charter School categorization to which Thomas High School belongs (reference images below in usual order):
- 0.01 point decrease in Average Math Score
- 0.01 point increase in Average Reading Score
- 0.01% decrease in % of students passing math
- 0.03% decrease in % of students passing reading
- 0.04% decrease in % of students passing overall
- Finally, these changes too, become insignificant once applying appropriate formatting to the data (See below):
Summary
Once the ninth grade scores from Thomas High School were replaced with NaNs and the data was re-analyzed to appropriately represent the new dataset, the new analysis seemed to reveal its major changes generally at the district and school summary levels.
The most noteable change at the district level was the decrease in overall Average Math Scores in the city school district (see District Summary Results above). This change was the most significant of those regarding academic scores, as the change was represented in 3 tenths of a grade point lower without the inclusion of Thomas High School's grade nine scores vs. with grade nine scores. A change in the tenths of a grade point average is typically deemed significant in comparison to changes in the hundreths of grade point. Similarily, a change in Average Math Scores displayed at the school summary level exhibited a significant decrease in Average Math Scores at Thomas High School without the inclusion of their grade nine scores vs. with (see School Summary Results above). Although not as substantial at the school level vs. district level, this decrease is deemed significant at just over one tenth of a grade point lower when discluding the ninth grade data.
A couple of other changes worth noting would be those seen in the percentage of students passing reading and percentage of students passing overall at both the district and school levels. In the overall district summary as well as the summary of performance at Thomas High School, the analysis revealed the percentage of students passing reading decreased by approximately three tenths of a percent (see District Summary and School Summary Results above). This is an interesting finding when compared to the average scores data, suggesting that the proportion of grade nine students passing reading at Thomas High School was significant enough to display a change once removing their data, however, their particular average score was actually not significant enough to reveal a noticeable change in comparison. When assessing the percentage of students passing overall at both district and school levels, a decrease of one tenth and three tenths of a percent was displayed in the new analysis, respectively.
When considering these percentage changes, one could argue here noting that although these changes are represented in the tenths of a value, changes in the tenths of a percentage point represent changes in the hundredths of a grade point, and therefore may not be categorized as significant per se. Nevertheless, these changes are denoted in the final formatting of the data, and if considered, provide interesting insights when set alongside comparative results.