Racial Bias in Approving Standardized Testing Questions
Two 1998 SAT verbal sentence-completion items had similar themes. However, the question correctly answered by more black students than white students was discarded, while the question correctly answered by more white students than black students became an official SAT question. SAT questions are deemed as “reliable [acceptable] questions” if they replicate the outcomes of previous exams, in which black and hispanic students historically scored lower than white students. They justify this by validating questions in which people of "high ability" tend to answer correctly and people of "low ability" tend to answer incorrectly.
1
1
Contributors are:
Who are from:
Tags
Disability Studies
Ch.2 Psychological Research - Psychology @ OpenStax
Psychology @ OpenStax
Introduction to Psychology @ OpenStax Course
Social Science
Empirical Science
Science
OpenStax
OpenStax Psychology (2nd ed.) Textbook
Psychology
Related
Standardized Testing Outcome Disparities
Racial Discrimination in the Verbal Component of Standardized Tests
Racial Bias in Approving Standardized Testing Questions
Racial Bias in Approving Standardized Testing Questions
A school district develops a new standardized test to measure 'critical reasoning skills.' The test includes several passages that use complex vocabulary and scenarios drawn from classical literature and orchestral music. When the results are reviewed, it is found that students from higher-income families consistently score significantly higher than students from lower-income families. Which of the following statements best analyzes this outcome?
De-emphasis of Standardized Tests in College Admissions
Learn After
A standardized test development committee pilots two new questions. On Question X, students from historically high-scoring demographic groups outperform students from historically low-scoring groups. On Question Y, the opposite occurs. The committee decides to include Question X and discard Question Y, arguing that a question is 'reliable' only if its results are consistent with the test's historical performance patterns. What is the most significant flaw in the committee's reasoning?
A standardized test development agency pilots two new questions. The agency's goal is to select questions that are 'reliable,' meaning they produce results consistent with historical test data where Group X has consistently outscored Group Y. Below are the results for the two pilot questions:
- Question 1: 70% of students from Group X answered correctly, while 40% of students from Group Y answered correctly.
- Question 2: 55% of students from Group X answered correctly, while 65% of students from Group Y answered correctly.
Based on the agency's validation process, which outcome is most likely?