Measuring the readability of medical research journal abstracts

摘要

This study examines whether the readability of medical research journal abstracts changed from 1960 to 2010. Abstracts from medical journals were downloaded from PubMed.org in ten-year batches (1960s, 1970s, etc.). Abstracts in each decade underwent processing via a custom Python script to determine their Coleman-Liau Index (CLI) readability score. Analysis using one-way ANOVA found statistically significant differences between the mean CLI readability scores of each decade (F(4, 6689135) = 12936.91,p＜0.0001). Post-hoc analysis using Tukey's method also found all pairwise comparisons between decades' mean CLI readability scores to be statistically significant (p<0.001). Readability scores increased from decade to decade beginning with a mean CLI score of 16.0813 in the 1960s and ending with a mean CLI score of 16.8617 in the 2000s. These results indicate a 0.7804 grade level increase in the difficulty of reading medical research journal abstracts over time and raises questions about the accessibility of medical research for broader audiences.