Unverified50% confidenceOpinionExact time
A small-scale experiment with 24 samples was enough to expose systematic differences among judges.
1
Sources
50%
Confidence
Long-term
Relevance
10/18/2026
First Seen
Sources
Related Entities
Related Claims
Unverified25%的风险降幅数字仍需在方法论严格的复制研究中验证，应被视为值得进一步验证的信号而非最终定论72% similarUnverifiedUsage forecasting methodology recommends reserving a 20–30% buffer for experimentation and unexpected consumption71% similarUnverifiedOf the 21 testable semantic changes, the direction of movement in 19 cases is consistent with philological evidence.71% similarUnverifiedThe authors note that this significance may be affected by the small sample size and multiple comparisons, and could be due to chance.70% similarUnverifiedWhen reporting success rates, the sample size must be included, because 19/20 and 950/1000 represent different levels of evidential strength.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1059555API
curl https://kongchang.com/api/v1/knowledge/claims/1059555MCP
get_claim(id=1059555)