Back To Projects

The Impact of the Covid-19 Pandemic on the Test Scores of Various Demographics

Deniz G.

The research problem that this research paper attempts to address is, How did Covid impact and affect the test scores of different demographics of students in various schools in New York? The overall approach that was used to answer this question involved using statistical analysis to examine the correlation between how certain factors such as being economically disadvantaged or being hispanic affected student proficiency scores on New York State regents exam tests in 2018 and 2021 in different schools.


The research problem that this research paper attempts to address is, How did Covid impact and affect the test scores of different demographics of students in various schools in New York? This issue is significant for the following reasons. By the conclusion of the school year, students had an average of five months’ worth of math and four month’s worth of reading behind them, according to an analysis of the pandemic’s effects on K-12 student learning. For example, student testing in 2021 was about ten points behind in math and nine points behind in reading, compared with matched students in previous years. The pandemic expanded already-existing achievement and opportunity inequities, particularly harming historically underprivileged students. Children in low-income schools had seven months of unfinished arithmetic work at the end of the year, compared to six months for students in majority-Black schools. High school students are now more likely to drop out, and seniors, particularly those from low-income households, are less likely to continue their education after graduation. The crisis also had an effect on pupils’ general health and well-being, with more than 35% of parents expressing severe or great concern for their kids’ mental health. The effects of the epidemic pose a threat to the hopes and possibilities for this generation far into adulthood. Their chances of enrolling in college and subsequently securing a rewarding profession that allows them to sustain a family may be adversely affected by the knock-on effects. Furthermore, analysis indicates that the pandemic’s effects on schooling may result in today’s students earning $49,000 to $61,000 less in their lifetime unless measures are implemented to remedy unfinished learning. The overall approach that was used to answer this question involved using statistical analysis to examine the correlation between how certain factors such as being economically disadvantaged or being hispanic affected student proficiency scores on New York State regents exam tests in 2018 and 2021 in different schools. Some of the significant results that came out of this research paper involved certain demographics such as those who were economically disadvantaged and those who were white. Specifically, in relation to how their demographic correlated with their school’s percent proficiency difference in a subject exam between the years 2018 and 2021. For example, 100% of the students of the Academy for Language and Technology are economically disadvantaged and compared to 2021, the number of students who were proficient in English dropped by 49%. Furthermore, 72% of Akron High school’s students are economically disadvantaged and compared to 2021, the number of students who were proficient in English dropped by 68%. However, on the other side of the spectrum, students of white descent did not fare as badly. For instance, Adirondack Middle School has an approximate demographic makeup of 100% white students. Their student’s percent proficiency when it came to their Algebra I score did not change at all. Additionally, Yonkers Montessori Academy has a demographic makeup where 46% of its students are white. The students from this school showed a 5% increase in their proficiencies in their science scores. The research that was conducted as the basis for this paper was done in an effort to examine the COVID-19 pandemic’s effects on student performance across various schools in New York. Specifically, it affected some demographics disproportionately when compared to others. Due to students being forced to learn from home, some without the necessary technology and adequate resources. Consequently, a lot of students took a significant hit to their academic progress. For instance, 100% of a student’s demographic make up is economically disadvantaged and out of these students, 49% of them were no longer proficient in ELA after the COVID-19 pandemic. Whereas, students of white descent did not fare as badly. For instance, Adirondack Middle School has an approximate demographic makeup of 100% white students. Their student’s percent proficiency when it came to their Algebra I score did not change at all.

Explore More!

Deniz G.
Kush Khosla
MS Computer Science from Stanford, Lead Data Scientist at Retain.ai

Related Projects

Computational Approaches to Enhance Idiopathic Intracranial Hypertension Diagnosis: A Neural Network-Based Framework for Improved Clinical Decision Support

A computer model, trained on eye images, can tell if optic nerves are healthy or not with 97% accuracy. This is important because it messes up less often than people do, making it a useful tool. The model looks at pictures of the back of the eye (fundus imaging) to make these predictions.
Rohan B.
Mentored by
How Does One’s Background Determine Their Mental State?

Mental health issues have become very prevalent in recent times, and although significant progress has been made in terms of treatment in the form of counseling, medication, and other methods, in order to truly find out the root cause of many mental health issues, a correlation has to be drawn between one’s mental state and another factor, such as family history, age, work environment, etc. It is also beneficial to correlate one’s mental state with a multitude of factors to see the compilation.
Sohum T.
Mentored by Akshay Jagadeesh
Detecting Facial Expressions

This paper shows that machine learning algorithms can be used in order to classify emotions into different categories. The most promising results were from the K-nearest neighbor model, with 61.5% accuracy in the training set, and 32.1% accuracy in the testing set.
Bertran M.
Mentored by Odysseas Drosis