By Owen Reyes · Republished 2026-05-16 · Originally reported by Matthew Fleischer on FishbowlLA, September 2010
In late September 2010, FishbowlLA covered LA Times reporter Jason Felch sitting on a UC Berkeley panel to discuss the paper’s controversial series on teacher evaluations — and taking criticism in the wake of a teacher’s death. The original framing tracked the journalism community processing one of the era’s most contested editorial decisions.
Then
The LA Times’ 2010 teacher-evaluation series was one of the most consequential and most contested pieces of data journalism of its era. Using a statistical method known as “value-added” modeling — which estimated individual teachers’ effectiveness from their students’ standardized-test-score gains — the Times made the editorial decision to publish the names and ratings of thousands of individual Los Angeles teachers.
The controversy intensified enormously after the death of fifth-grade teacher Rigoberto Ruelas; family members and teachers-union officials said the publication of his evaluation scores may have contributed to it. The Berkeley panel was, in effect, the journalism and education communities working through the ethics question in real time. A Berkeley statistician on the panel, Sophia Rabe-Hesketh, argued that value-added data sets were too crude for high-stakes decisions or for publicly naming and shaming individuals. A former Oakland teacher, Anthony Cody, framed the controversy as part of a broader hostility toward the profession.
The core question — was publishing individual scores accountability journalism, or a harm-causing overreach? — was the one Felch was made to answer in person.
Now
The value-added measurement approach the series championed has been substantially discredited in the years since. Education researchers and statisticians broadly concluded that value-added scores are too statistically noisy and too confounded by factors outside a teacher’s control — school leadership, resources, student circumstances — to fairly rate individual teachers. Most districts and states that adopted value-added evaluation systems in the early 2010s later walked them back.
The LA Times series itself drew major recognition at the time, but it is now frequently cited in journalism-ethics discussions as a cautionary case about publishing algorithmically derived ratings of named individuals. The deeper question the Berkeley panel raised — the responsibility that comes with data journalism that identifies people by name — has only grown more central as data-driven and algorithmic journalism has expanded.
The 2010 piece reads now as a documented moment of the press examining itself: a reporter facing his peers over an editorial decision whose consequences, and whose underlying method, the following decade would substantially reassess.
This article references a death by suicide and the journalism-ethics debate that followed. It is presented as media-history documentation. If you are struggling, support is available — in the US, call or text 988.
Original report archived on the Wayback Machine.