Insensitivity to sample size is a cognitive bias that occurs when individuals evaluate the probability of a sample statistic without taking the sample size into account. People often ignore the fact that results derived from smaller samples are subject to greater variability, leading them to assign similar probabilities to outcomes from vastly different sample sizes. This misconception can skew perceptions of statistical data, as people may view smaller samples as representative despite their inherent instability.
Amos Tversky and Daniel Kahneman found that when asked which hospital—one with 45 daily births or another with 15—would report more days where over 60% of babies born were boys, most people incorrectly thought both hospitals would show similar results. In reality, the smaller hospital is more likely to experience fluctuations due to its limited sample size.
To combat insensitivity to sample size, one can consciously remind themselves to consider the sample size's impact on variability and reliability. Reviewing statistical principles like the law of large numbers can also help make more informed judgments based on sample size.