A two‐dimensional and non‐hierarchical framework of Bloom's taxonomy for biology

Abstract

Bloom's taxonomy is a classification of learning objectives originally developed for general educational purposes. The taxonomy was subsequently revised to expand beyond cognitive processes and to include an additional knowledge dimension. The revision was prompted by psychometric measurements indicating inconsistencies in the original taxonomy, and the revised taxonomy has been adapted for use in many disciplines. However, biology education researchers and practitioners continue to use the original taxonomy despite the concerns of validity. To facilitate the use of the revised Bloom's taxonomy in biology, we developed a discipline‐specific rubric from an analysis of assessment items from diverse biology disciplines.Implicit learning objectives of 1,432 assessment items were coded by two researchers. Items were collected from AP Biology, MCAT, and both introductory and advanced undergraduate biology courses. A biology‐specific rubric was generated using the revised taxonomy as a model. Problems were coded independently by two raters in batches of 50–100, followed by a consensus process. After coding 831 items, a revised rubric was developed and tested on the remaining 601 items. Inter‐rater reliability was 0.76 and 0.70 (Cohen's κ) for the cognitive‐process and knowledge dimensions respectively.Using thematic analysis as the methodology over two versions of our rubric, we identified distinctive features among biology assessment items that define the six cognitive processes and four types of knowledge in the revised Bloom's taxonomy. Quantitative analysis indicates that our data set contains problems mostly in the remember or understand categories (43.9% and 37.3% respectively) but rarely in the apply, analyze, evaluate, or create categories (7.0%, 5.2%, 4.7%, 2.0% respectively). These results are aligned with existing biology education research literature using the original Bloom's taxonomy. In the knowledge dimension, factual, conceptual, and procedural knowledge are over‐represented (38.4%, 48.7%, 12.8% respectively), with essentially no metacognitive knowledge (0.2%). This suggests a potential area for change in biology education, as metacognition is critical to how people learn.We performed statistical analyses on the 601 items coded with our revised rubric to examine the Bloom's taxonomy framework. Contingency analysis indicates that the cognitive processes and knowledge dimensions are not independent (χ2 = 249.05, df = 10, p < 0.0001). Correspondence analysis identifies two components that account for 100% of the inertia or variation in how the two dimensions are related, with three distinct clusters of knowledge and cognitive process: factual knowledge with the remember cognitive process, procedural knowledge with the apply cognitive process, and conceptual knowledge with the remaining four cognitive processes. Together, these results provide a two‐dimensional, non‐hierarchical framework of Bloom's taxonomy for use of in biology.Support or Funding InformationThis project was supported in part by an institutional award for undergraduate education from the Howard Hughes Medical Institute and the Office of Undergraduate Research at Northwestern University.