This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
A U.S. federal judge allowed a lawsuit to proceed against the Houston Independent School District regarding its use of the SAS Institute's EVAAS algorithmic teacher evaluation system. Teachers argued that the system's proprietary nature violated their due process rights by preventing them from challenging termination decisions based on potentially erroneous scores. The court ruled that teachers should have the opportunity to test the accuracy of their scores, noting that algorithms are subject to human error.
On May 4, 2017, a U.S. federal judge advanced teachers’ claims that the Houston Independent School District’s algorithmic teacher evaluations violated their due process rights to their jobs by not allowing them to review the grounds of their termination.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
Challenges in understanding or explaining the decision-making processes of AI systems, which can lead to mistrust, difficulty in enforcing compliance standards or holding relevant actors accountable for harms, and the inability to identify and correct errors.
Also classified as
The school district delegated high-stakes employment evaluation and termination decisions to an automated system without providing a meaningful way for humans to review or appeal the algorithmic outputs.
The automated evaluation system created job insecurity and systemic pressure on teachers, negatively impacting their employment quality and working conditions.
The core of the dispute is the proprietary, 'trade secret' nature of the EVAAS algorithm, which prevented teachers and the school district from understanding or verifying how performance scores were calculated.
AI system
Due to a decision or action made by an AI system
Unintentional
Due to an unexpected outcome from pursuing a goal
Post-deployment
Occurring after the AI model has been trained and deployed
The due process violations and potential wrongful terminations arose from the post-deployment operation of the EVAAS scoring algorithm, which was deployed to evaluate teachers.
Added categories peak at 2: Minor (Psychological Harm, Inferred)
A domestic legal dispute in the United States concerning a school district's use of a proprietary teacher evaluation algorithm. The federal court ruled that using secret algorithms for high-stakes employment decisions without a way to challenge the scores violates procedural due process. The national security impact is negligible to minor, primarily touching on domestic civil rights and administrative law.
The EVAAS system acts as an Assistant by analyzing student test scores to generate performance ratings for teachers, which the school district then uses to guide its employment and termination decisions.
Risk Level 2: High Risk. The system is an employment and worker management system used to evaluate teacher performance and make termination decisions, which directly affects employment.
The EVAAS system was explicitly deployed to track and evaluate teacher performance based on student standardized test scores.
The reports clearly outline the legal dispute, the parties involved, the AI system (EVAAS), and the specific civil rights issues (due process). The facts regarding the lawsuit and the court's ruling are well-documented.
None. The incident is clearly a legal dispute over algorithmic transparency in employment evaluation.
Technology
Proprietary Algorithm
SAS kept the EVAAS algorithm a trade secret, preventing verification.
Potential Software Glitches
Code errors or data entry mistakes could erroneously calculate scores.
Data Inputs
Standardized Test Scores
System relied heavily on student standardized test scores for evaluation.
Data-Entry Mistakes
Incorrect input data could lead to erroneous teacher evaluation scores.
Human Factors
Blind Trust in Algorithm
District relied on the system without providing a way to verify accuracy.
Process and Methods
Lack of Verification Process
No process existed for teachers to independently test or verify their scores.
Inadequate Dispute Mechanism
Teachers could not challenge the evidence used for termination decisions.
Management
Aggressive Termination Goal
HISD set a goal to fire 85 percent of teachers rated ineffective.
Lack of Transparency
Management adopted a system without ensuring explainability for employees.
Regulatory Environment
Due Process Violations
Evaluation system failed to provide teachers with detailed basis for firing.
AI system involved
Identified by automated extraction; can differ from the alleged developers above.
