'Selection bias' as a reason why schools with many high-achieving students do not necessarily provide superior education
by
Some parents may choose or recommend a school for their children, thinking, 'This school has produced many excellent graduates, so if I send my child there, he will be successful in the future.' However, Freddie deBoer , a teacher working in New York and a member of the City University of New York faculty and teachers' union, explains that selection bias is hidden throughout the education system, and that 'schools with many high-achieving students do not necessarily provide a superior education.'
why selection bias is the most powerful force in education – the ANOVA
https://fredrikdeboer.com/2017/03/29/why-selection-bias-is-the-most-powerful-force-in-education/
In the United States, it is said that scores on the SAT , the college entrance exam, have a significant impact on students' academic performance after entering university, and students with higher SAT scores tend to perform better in college, are less likely to drop out in their first year, and are more likely to successfully complete four or six years of school. Furthermore, students with high SAT scores are more likely to find full-time employment after graduation.
Large-scale studies have also confirmed that 'people with higher SAT and spatial reasoning scores in their youth are more likely to become scientists, scholars, CEOs of Fortune 500 companies, federal judges, senators, and billionaires in the future.' It has also been found that the number of patents obtained and papers published correlates with SAT scores and spatial reasoning. A study focusing on 'precocious geniuses' called SMPY initially defined geniuses as those in the top 1% of college entrance exam scores, and this group included well-known figures such as Google founder Sergey Brin, Facebook founder Mark Zuckerberg, and artist Lady Gaga.
A 45-year follow-up study of 5,000 geniuses found that 'precocious geniuses' are more likely to achieve social and economic success - GIGAZINE
Here's a comparison of the average SAT scores for the two states of Connecticut and Mississippi:
The average SAT score in Connecticut is 450 in math and 480 in English, for a total of 930. On the other hand, the average score in Mississippi is 530 in math and 550 in English, for a total of 1080. Looking at the scores alone, Mississippi students appear to be higher achieving and more accomplished than Connecticut students.
Considering that SAT scores have a significant impact on people's lives, what would happen if a governor who wanted to revitalize his state's economy analyzed this data and implemented reforms to the education system? Connecticut has a strong education union in which most of the state's teachers belong, while many teachers working in Mississippi's public schools do not belong to a public union. Based solely on the fact that Mississippi has higher average SAT scores, the governor might adopt a policy of opposing unions, arguing that children's academic ability would improve without unions.
However, Connecticut is a state with a successful education policy, ranking second in the nation for the academic performance of its fourth and eighth grade students according to the National Assessment of Educational Performance (NAEP). Meanwhile, Mississippi ranks second from the bottom in the nation for the academic performance of its fourth and eighth grade students.
The reason why these results appear to be opposites to reality is due to the relationship between the SAT test-taking rate and average scores in each state. Mississippi's SAT test-taking rate is 3%, and those who do are college-ready, so to speak, the top students in the state. On the other hand, Connecticut's SAT test-taking rate is 88%, meaning nearly all students have the academic ability to take the SAT. While the large number of test-takers lowers the overall average score, the top 20% of SAT-takers in Connecticut actually perform better than the top 20% of SAT-takers in Mississippi.
by
This suggests that our understanding of the 'quality of education' may be influenced by selection bias .
For example, imagine a city has one public high school and one private high school that prides itself on the high scores of its students on academic tests. If you simply compare the students' grades, the educational content of the private high school may appear superior to that of the public high school. However, being private, it is of course very expensive, and only a limited number of children can attend. In other words, the grades of students attending private high schools are influenced by their family income, so it cannot necessarily be said that the quality of the education offered at private high schools is sufficient to ensure good grades for students. It is impossible to say whether the quality of education is superior or not without taking into account the screening process that is used to screen the students who attend each school.
The graph below shows the average SAT scores at the time of college admissions on the horizontal axis, and the average scores on a standardized test called the Collegiate Learning Assessment (CLA) administered at each university on the vertical axis. The blue dots represent the average scores of first-year students at each university, and the red dots represent the average grades of fourth-year students. The difference between the blue and red lines represents the level of learning over four years of university. This graph shows that there is a strong correlation between SAT and CLA scores in both the first and fourth years of university, indicating that selection bias already exists at the time of university admissions.
Harvard graduates excel academically and professionally, but that doesn't automatically mean that a Harvard education is superior. Harvard spends millions of dollars carefully 'selecting its students,' and many of the kids who make it through Harvard's admissions process excel wherever they go. Having a high number of highly accomplished graduates doesn't answer the question: Is a Harvard education superior?
The existence of selection bias also changes our view of causality in education. Many advocates for education reform argue that the system is such that students from poor families or minorities cannot live in expensive or predominantly white areas and therefore cannot enter good schools. However, this causality may also be reversed. In other words, the idea is that these areas are able to maintain high academic performance because they can efficiently exclude poor and minority students. Schools that educate students from socially advantaged backgrounds naturally appear to be good schools. However, this idea is rarely discussed in the media or among liberals.
by JodyHongFilms
Selection bias in education is ubiquitous. In the United States, charter schools—elementary and secondary schools established by parents, teachers, and community groups with state or school district approval—have been on the rise since the 1990s. They are often described as 'revolutionary' due to their educational considerations for a diverse range of students. However, a thorough investigation by Reuters in 2013 revealed that some charter schools have systems in place to intentionally exclude students who are difficult to educate. While these charter schools appear to accept all students, just like public schools, they often employ hidden 'filters,' such as holding meetings at times when working parents cannot attend and requiring students to attend those meetings as a condition for admission.
In Stanford University's CREDO Project, which investigated the effectiveness of charter schools, researchers claim to have taken into account potential 'filtering' mechanisms that would be hidden in large-scale studies. However, as Campbell's Law states, 'The more powerful any quantitative indicator is, the more it will be subject to pressure to degrade by those who seek to undermine the system,' the modern American education system is subject to many biases.
'When we think about schools and the quality of education, we rarely consider selection bias,' deBoer said. 'Many parents assume, 'The students at this school are doing very well. My child, who is a struggling student, will do well if I send him to this school.'' However, he said, 'if we want to use limited educational resources fairly and efficiently, we need to look deeper and recognize the dynamics of the modern education system.'
Related Posts:
