FINITE SAMPLE BIAS:

Another term to describe Sparse Data Bias (synonym)^1,2; which is a bias in a statistical estimate, that occurs due to using a statistical test in a scenario when there are too few data points (or no data) in the categories being evaluated by the test. This bias leads to an overestimate of treatment effects, or stronger associations between variables compared to what the true associations, or effects are in nature.

The term finite refers to the fact that the amount of data available in each category (cells of data) becomes smaller as the analysis becomes more complex, and thus there are a limited number of valid comparisons that can be made between categories within the dataset. Also see: Sparse Data Bias, Small Sample Bias, Small Study Bias, and Wrong Sample Size Bias.

References:

1. Richardson DB, Cole SR, Ross RK, Poole C, Chu H, Keil AP. Meta-Analysis and Sparse-Data Bias. Am J Epidemiol. 2021;190(2):336-40. (Link to Reference)

2. Ross RK, Cole SR, Richardson DB. Decreased Susceptibility of Marginal Odds Ratios to Finite-sample Bias. Epidemiology. 2021;32(5):648-52. (Link to Reference)