Bloom's taxonomy
Bloom's taxonomy is a scheme for sorting learning goals by the kind of thinking they require, from recalling a fact to judging or producing something new. Its full name is the Taxonomy of Educational Objectives, and its first volume, on the cognitive domain (knowledge and thinking), was published in 1956 by Benjamin S. Bloom and four co-authors [1]. The original scheme has six levels: knowledge, comprehension, application, analysis, synthesis and evaluation [2].
A revised version appeared in 2001, edited by Lorin W. Anderson and David R. Krathwohl. It renamed the levels as verbs (remember, understand, apply, analyze, evaluate, create), moved "create" to the top, and added a second axis for the type of knowledge involved: factual, conceptual, procedural and metacognitive [3] [2]. A separate handbook, published in 1964, classified goals in the affective domain (feelings, interests and values) [4].
In language learning the levels map onto familiar tasks: remembering that "gato" means "cat" sits at the bottom, and writing an original story in Spanish sits at the top. The taxonomy is widely used to plan lessons and tests, and it has also been criticised, mainly for its claim that each level must be mastered before the next [5] [6].
Contents
Key facts
- The taxonomy began as a way for university examiners to share test items that measured the same goal; Bloom, then Associate Director of the University of Chicago Board of Examinations, started the project, and the group met about twice a year from 1949 [2].
- The original six categories were ordered from simple to complex and from concrete to abstract, and were assumed to form a cumulative hierarchy in which each level is a prerequisite for the next [2].
- The original taxonomy was eventually translated into 22 languages [2].
- Analyses of curricula and tests with the taxonomy have repeatedly found a heavy emphasis on recall, the lowest category [2].
- The 2001 revision swapped the order of synthesis and evaluation, renamed synthesis "create", and split the six categories into 19 specific cognitive processes [2] [3].
- In a 2019 study of middle school and college students, practice quizzes with higher-order questions improved higher-order test scores, while fact-only quizzes did not, contrary to the idea that facts must come first [6].
The 1956 taxonomy
Origins of the project
A common language for learning goals
Bloom saw the result as more than a testing tool. According to Krathwohl, he believed it could serve as a common language about learning goals across subjects and grade levels, as a way to pin down the meaning of broad goals, as a check that objectives, activities and tests in a course match one another, and as a panorama of possible goals against which any one course could be compared [2]. The word "taxonomy" was unfamiliar in education at the time and the book drew little attention at first, but it later became widely known and cited [2].
Levels of the original taxonomy
The six categories
Each of the six original categories had a careful definition, and all except application were divided into subcategories [2]. Knowledge, the lowest level, covered the recall or recognition of specifics (terms and facts), of ways of dealing with specifics (conventions, sequences, classifications, criteria, methods), and of general principles and theories. Comprehension covered translation, interpretation and extrapolation. Application had no subcategories. Analysis covered elements, relationships and organizing principles. Synthesis covered producing a unique communication, a plan, or a set of abstract relations. Evaluation, the top level, covered judgments based on internal evidence and on external criteria [2] [1].
The mixed knowledge category
The knowledge category was unusual because it mixed two things: a type of content (facts, conventions, theories) and an action (recalling it). The 2001 revision separated these into two dimensions [2].
The 2001 revision
The revised taxonomy was developed in much the same way as the original, by a group of specialists, 45 years later [2]. Its book, A Taxonomy for Learning, Teaching, and Assessing, treats every learning objective as a verb plus a noun: the verb names a cognitive process, and the noun names the knowledge it acts on [3].
The cognitive process dimension
Renamed categories
The cognitive process dimension keeps six categories, with these changes [2]:
- Knowledge became Remember, defined as retrieving relevant knowledge from long-term memory.
- Comprehension became Understand, because "understand" is the word teachers commonly use in objectives; its absence had been a frequent criticism of the original.
- Application, Analysis and Evaluation kept their meaning in verb form: Apply, Analyze, Evaluate.
- Synthesis became Create, defined as putting elements together to form a novel, coherent whole or an original product, and moved above Evaluate to the top.
The 19 cognitive processes
The 19 cognitive processes under these categories include recognizing and recalling (Remember), interpreting, exemplifying, classifying, summarizing, inferring, comparing and explaining (Understand), executing and implementing (Apply), differentiating, organizing and attributing (Analyze), checking and critiquing (Evaluate), and generating, planning and producing (Create) [2].
A relaxed hierarchy
The revision still treats the categories as increasing in complexity, but it relaxed the strict hierarchy. The categories may overlap: explaining, a process under Understand, can be more demanding than executing, a process under Apply [2].
The knowledge dimension
Four types of knowledge
The second axis of the revision classifies what is being learned [2] [3]:
- Factual knowledge: the basic elements of a subject, such as terminology and specific details. In Spanish this includes single words such as "gato" or "camión" (the usual Mexican word for "bus").
- Conceptual knowledge: how basic elements fit together in a larger structure, including classifications, principles and theories. The system of grammatical gender or the contrast between Ser and estar is conceptual knowledge.
- Procedural knowledge: how to do something, including skills, methods and the criteria for choosing a method. Conjugating a regular -ar verb is procedural knowledge.
- Metacognitive knowledge: knowledge about thinking in general and awareness of one's own thinking. This category was new in 2001. Krathwohl's own example is knowing which errors are common in one's own writing and how to correct them.
The Taxonomy Table
Combining the two axes gives a grid called the Taxonomy Table, with the four knowledge types as rows and the six processes as columns. Any objective can be placed in one or more cells, and empty cells show which kinds of goals a course leaves out [2].
Comparison with language-learning examples
The table lists the levels of both versions side by side. The definitions follow Krathwohl's 2002 overview; the Spanish examples are illustrations, not taken from the taxonomy books.
The Analyze and Evaluate examples connect to the articles on Preterite and imperfect and Forms of address in Spanish. Producing language, the Create level, is also the focus of research on output and feedback.
| Original level (1956) | Revised level (2001) | Revised definition | Spanish learning example |
|---|---|---|---|
| Knowledge | Remember | Retrieving relevant knowledge from long-term memory [2] | Recall that "gato" means "cat" |
| Comprehension | Understand | Determining the meaning of instructional messages, oral, written or graphic [2] | Explain that "¿Qué onda?" is an informal greeting in Mexico |
| Application | Apply | Carrying out or using a procedure in a given situation [2] | Conjugate a new regular verb such as "caminar" in the present tense |
| Analysis | Analyze | Breaking material into parts and detecting how the parts relate to each other and to the whole [2] | Identify why a story uses the preterite for one event and the imperfect for another |
| Evaluation (6th in 1956) | Evaluate (5th in 2001) | Making judgments based on criteria and standards [2] | Judge whether "tú" or "usted" fits a message to a teacher |
| Synthesis (5th in 1956) | Create (6th in 2001) | Putting elements together to form a novel, coherent whole or original product [2] | Write a short story in Spanish |
Affective and psychomotor domains
The 1956 handbook covered only the cognitive domain. A second handbook, by Krathwohl, Bloom and Bertram Masia, appeared in 1964 and classified affective objectives, which concern interests, attitudes and values [4]. Its five levels are receiving (being willing to pay attention), responding (taking part), valuing (attaching worth to something), organization (fitting values into a system) and characterization by a value or value complex (acting consistently on a set of values) [4]. In language learning, a learner who starts by listening to Mexican music, then joins conversations, and later chooses to keep using Spanish in daily life illustrates a move up these levels.
Use and evaluation
Uses in teaching and assessment
Classifying curricula and tests
One of the most frequent uses of the original taxonomy has been to classify the objectives of a curriculum and the items on a test, to show how far they range across the levels [2]. Such analyses usually find most goals at the recall level, while goals from comprehension to synthesis are generally regarded as the most important outcomes of education; the analyses have therefore been used to push courses and tests toward more complex goals [2].
Links to mastery learning
The taxonomy also connects to Bloom's later work. His research on Mastery learning and the "2 sigma problem" listed "higher order questions" among the teaching variables that raise achievement [7]. A 2025 randomized trial of an AI tutor in a Harvard physics course described its lessons as targeting the understand, apply and analyze levels of the taxonomy [8]. In language teaching, the levels are often used as a rough guide to sequencing: learners meet a word, understand it in context, use it, and eventually create with it. How fast a learner should move up depends on what is already known, the subject of Knowledge frontier. The role of understood listening and reading in building the lower levels of language knowledge is covered under Comprehensible input.
Criticisms and limits
Furst's 1981 review
A 1981 review by Edward Furst, one of the original authors, examined four claims made for the taxonomy: that it is neutral with respect to educational values, that it is comprehensive, that its categories form a cumulative hierarchy, and that it is useful [5]. The review drew on about 25 years of published and unpublished studies [5].
The hierarchy claim
The hierarchy claim has drawn the most attention. The 2001 revision itself loosened it, accepting that the categories overlap and that some "lower" processes are harder than some "higher" ones [2]. Placing Create above Evaluate also reversed the order the original authors had chosen. A common reading of the taxonomy holds that learners must build a base of facts before attempting higher-order thinking. Pooja Agarwal tested this directly in 2019: in laboratory and school experiments, retrieval practice with higher-order questions or a mix of questions improved higher-order test performance, while practice with fact questions alone did not [6]. She concluded that fact-first practice may be less effective for higher-order learning than practicing higher-order questions directly. The link between practice and memory is covered under Retrieval practice and Spaced repetition.
Limits of scope
Other limits follow from the taxonomy's purpose. It classifies goals, not methods of teaching, so it says little about how a learner reaches a level. It was not designed specifically for language learning, and a learner may be able to explain a grammar rule (Understand) yet still be unable to use it quickly in conversation. Frameworks built for languages, such as the CEFR levels, describe what a learner can do in the language rather than classifying thinking processes. The Stages of talent development described in Bloom's later research offer a different, developmental view of how learners progress over years.
See also
Pages that link here
References
- Bloom, B. S., Engelhart, M. D., Furst, E. J., Hill, W. H., Krathwohl, D. R. (1956). Taxonomy of Educational Objectives: The Classification of Educational Goals. Handbook I: Cognitive Domain. Longmans, Green / David McKay. archive.org/details/taxonomyofeducat0000unse_t6n4
- Krathwohl, D. R. (2002). A Revision of Bloom's Taxonomy: An Overview. Theory Into Practice, 41(4), 212-218. doi:10.1207/s15430421tip4104_2
- Anderson, L. W., Krathwohl, D. R. (eds.) (2001). A Taxonomy for Learning, Teaching, and Assessing: A Revision of Bloom's Taxonomy of Educational Objectives. Longman. archive.org/details/taxonomyforlearn00unse
- Krathwohl, D. R., Bloom, B. S., Masia, B. B. (1964). Taxonomy of Educational Objectives: The Classification of Educational Goals. Handbook II: Affective Domain. David McKay. archive.org/details/taxonomyofeducat0000unse
- Furst, E. J. (1981). Bloom's Taxonomy of Educational Objectives for the Cognitive Domain: Philosophical and Educational Issues. Review of Educational Research, 51(4), 441-453. doi:10.3102/00346543051004441
- Agarwal, P. K. (2019). Retrieval practice & Bloom's taxonomy: Do students need fact knowledge before higher order learning?. Journal of Educational Psychology, 111(2), 189-209. doi:10.1037/edu0000282
- Bloom, B. S. (1984). The 2 Sigma Problem: The Search for Methods of Group Instruction as Effective as One-to-One Tutoring. Educational Researcher, 13(6), 4-16. doi:10.3102/0013189X013006004
- Kestin, G., Miller, K., Klales, A., Milbourne, T., Ponti, G. (2025). AI tutoring outperforms in-class active learning: an RCT introducing a novel research-based design in an authentic educational setting. Scientific Reports, 15, 17458. doi:10.1038/s41598-025-97652-6