Resources
Removed Questions
On rare occasions, LSAC removes a question from scoring after an exam has been administered. Here's what that means and why it happens.
What does “removed from scoring” mean?
When LSAC removes a question from scoring, that question is excluded when calculating your raw score. Every test-taker is treated as if they never saw the question — it simply does not count, regardless of how you answered it.
Your scaled score and percentile rank are computed only from the questions that remain in scoring.
Why does LSAC remove questions?
LSAC does not always disclose its reasons publicly, but removals typically stem from one or more of the following:
- Ambiguous wordingThe question or answer choices can be reasonably interpreted in more than one way, making it impossible to identify a single defensible correct answer.
- Defective stimulusThe passage or argument contains an error that undermines the question's validity.
- Multiple defensible answersTest-takers raised valid challenges showing that more than one answer choice could be correct under a sound interpretation of the question.
- Administrative errorIn rare cases, a question is removed due to a formatting issue, an incorrect answer key, or a printing error in the test booklet.
Does it affect your score?
Probably not — or if it does, the effect is extremely small and largely unpredictable. LSAT educator Jon Denning has written on this directly:
“My reply when I've been asked about the scaling impact and/or penalty that accompanies the removal of this question? Probably none, or at least extremely minimal and with virtually zero predictability at particular scoring outcomes. I'm sure they'll rescale it, and some scores may move up or down a point (tops imo), but I think it stays mostly identical and there's no telling who's helped or hurt score-wise.”1
The core reason: a single question isn't globally consequential enough to move the needle for everyone. Even the difficulty difference between an easy and hard RC section only produces a two-point scale variance at 170 — and that's for an entire section. One question carries far less weight.
Questions also affect different score levels differently, so even if a removal does shift some raw-to-scaled ratios, there's no reliable way to predict the direction or magnitude for any given score. The practical upshot: if you had a −7 raw, you now need 69 correct out of 76 instead of 70 out of 77 — essentially the same bar.