Bloom’s Taxonomy: Friend or Foe?
Print 
PDF
At the end of last month, ISPI received an email in regard to a job aid from Iowa State University, Revised Bloom’s Taxonomy Linked to the 4 Classes of Knowledge, tying John Anderson’s 4 classes of knowledge with Bloom’s cognitive domain. This created a wonderful dialogue amongst several ISPI members and resurfaced this article. The article (reprinted below) was written by Brenda Sugrue in the October 2002 issue of PerformanceXpress. Please take a moment to read both items and join the discussion by sharing your comments below. We’d love to keep the conversation going.
Problems with Bloom’s Taxonomy
I did a 99 second critique of Bloom’s taxonomy at the 2002 ISPI conference and it generated more unsolicited feedback than any other presentation I have made. The response varied from those who completely agreed with me and have abandoned Bloom many years ago to those who are still true believers and avid users. In the 99 seconds presentation I criticized the taxonomy but did not have time to present more valid alternatives. This article summarizes the criticisms and presents two alternative strategies for classifying objectives in order to design appropriate instruction and assessment.
Invalidity
Bloom’s taxonomy is almost 50 years old. It was developed before we understood the cognitive processes involved in learning and performance. The categories or “levels” of Bloom’s taxonomy (Knowledge, Comprehension, Application, Analysis, Synthesis, Evaluation) are not supported by any research on learning. The only distinction that is supported by research is the distinction between declarative/conceptual knowledge (which enables recall, comprehension or understanding), and procedural knowledge (which enables application or task performance).
Unreliability
The consistent application of Bloom’s taxonomy across multiple designers/developers is impossible. Given any learning objective, it might be classified into either of the two lowest levels (knowledge or comprehension) or into any of the four highest levels (application, analysis, synthesis, evaluation) by different designers. Equally, there is no consistency in what constitutes instruction or assessment that targets separate levels. A more reliable approach is to separate objectives and practice/assessment items into those that elicit or measure declarative/conceptual knowledge from those that elicit or measure task performance/procedural knowledge.
Impracticality
The distinctions in Bloom’s taxonomy make no practical difference in diagnosing and treating learning and performance gaps. Everything above the “knowledge” level is usually treated as “higher order thinking” anyway, effectively reducing the taxonomy to two levels.
The Content-by-Performance Alternative
Recent taxonomies of objectives and learning object strategies distinguish among types of content (usually facts, concepts, principles, procedures, and processes) as well as levels of performance (usually remember and use). This content-by-performance approach leads to general prescriptions for informational content and practice/assessment such as those presented in Table 1.
Table 1. Prescriptions for Information and Practice Based on Content-Performance Matrix.
| Content Type | Information to Present (Regardless of Level of Performance) |
Practice/Assessment (Depending on Level of Performance) |
|
| Remember | Use | ||
| Fact | the fact | recognize or recall the fact | recognize or recall during task performance |
| Concept | the definition, critical attributes, examples, non-examples | recognize or recall the definition or attributes | Identify, classify, or create examples |
| Principle/ Rule |
the principle/rule, examples, analogies, stories | recognize, recall, or explain the principle | decide if the principle applies, predict an event, apply the principle to solve a problem |
| Procedure | list of steps, demonstration | recognize, recall, or reorder the steps | perform the steps |
| Process | description of stages, inputs, outputs, diagram, examples, stories | recognize, recall, or reorder the stages | identify origins of problems in the process; predict events in the process; solve problems in the process |
The Pure Performance Alternative
A more radical approach would be to have no taxonomy at all, to simply assume that all objectives are at the use level (i.e., “performance” objectives) and that learners will practice or be assessed on the particular performance in representative task situations. If there are “enabling” sub-objectives, those too can be treated as performance objectives without further classification. If, for example, a loan officer needs to be able to distinguish among types of mortgages and describe the pros and cons of each type of mortgage as an enabling skill for matching house buyers with mortgages, then we design/provide opportunities to practice categorizing mortgages and listing their pros and cons before practice on matching buyers to mortgages. If a car salesperson needs to be able to describe the features of different car models as an enabling skill for selling cars, then we design/provide opportunities to practice describing the features of different cars before practice on selling cars.
References
Bereiter, C., & Scardamalia, M. (1998). Beyond Bloom’s taxonomy: Rethinking knowledge for the knowledge age. In A. Hargreaves, A. Lieberman, M.
Fullen, & D. Hopkins, (Eds.), International handbook of educational change. Boston, MA: Kluwer Academic.
Merrill, M.D. (1994). Instructional design theory. Englewood Cliffs, NJ: Educational Technology Publications.
Moore, D.S. (1982). Reconsidering Bloom’s Taxonomy of educational objectives, cognitive domain. Educational Theory, 32(1) 29-34.
Bloom’s Taxonomy: Friend or Foe?
Friend! Benjamin Bloom’s work in mastery learning theory (criterion-referenced testing) and his work in naming and organizing cognitive learning levels continue to influence 21st Century instructional design. Bloom is not alone in his understanding that “there are different types of learning outcomes and each type calls for a different type of instruction” (Richey, Klein & Tracey, 2011, p. 105). Robert Gagné cited Bloom as a forerunner to his work. Gagne expanded the types of learning tasks and demonstrated the relationships between them. A contemporary configuration, The Performance-Content Matrix (Robert Gagné and David Merrill,1990 as cited in Richey et al., p. 108) closely resembles the Content-Performance Matrix, Table 1 (Sugrue, B., October 2002, PerformanceXpress). David Krathwohl published Bloom’s Revised Taxonomy in the same year. Our critic asks “All this and for what real purpose?”
Bloom’s end goal was mastery learning and therefore his hierarchy was a tool to sequence instruction by demonstrating the relationships between learning tasks, their prerequisites and the overall learning outcomes.
Recently, we used Bloom’s Revised Taxonomy as an alignment tool to design a Kirkpatrick Level 2 workshop for a corporate L&D team. The focus was assessment design and development. After aligning learning goals to the needs of the organization, we introduced Bloom’s Revised as a design tool to assist learners to write a majority of their performance-based objectives at higher skill levels – above recall and understand. Then,we used Bloom’s Revised again to guide learners to compose test questions at that appropriate skill level.It was eye-opening to the practitioners as they realized they had been “teaching high” and “testing low.”
References
Richey, R. C., Klein, J. D., & Tracey, M. W. (2011). The instructional design knowledge base: theory, research, and practice. New York: Routledge.
Krathwohl, D. K. (2002). A revision of Bloom’s Taxonomy: An overview. Theory into practice, 41(4), 212-218.
Submitted by: Bonnie Beresford and Claudia VonDrak
I can see how Bloom’s taxonomy is helpful to those new to the field. When I first got started, learning about the taxonomy helped reinforce for me the need to align instructional methods with objectives.
But, as others have stated, I’ve found the taxonomy to be of little practical use. I agree with many of Brenda Sugrue’s points. And from the comments, I especially agree with Guy Wallace and his comment about assuming that all objectives are at the use level. If we’ve written an objective well (i.e., it identifies the desired behavior precisely), then it doesn’t really matter what level we think it is, as long as we get learners to perform that exact behavior in training.
I agree with Karen Carpenter. I believe Bloom’s Taxonomy is very relevant in a classroom/training environment when determining/framing the level of evaluation. It helps to determine to what level you need a learner to learn a concept. Sometimes I need my learners to really look at a statement and analyze the situation given and give me an answer based on what I am asking. Sometimes I only need my learners to be able to “recall” an appropriate response. True, the lesson objectives should determine the level of assessment. (check and balance) I use it as a job aid to develop the “test”. For me, the taxonomy frames how to develop the assessment.
As a teacher by profession, Bloom (original version) was very useful in the classroom for asking increasingly higher order questions to achieve learning objectives by engaging students. Ensuring meaningful, relevant and applied learning lends itself to memorable learning and more or less using the taxonomy. Much of secondary and post-secondary education (theory) courses are not performance related (few performance based objectives). Bloom’s taxonomy (revised or original) is not a clear-cut fit for all workplace learning or contexts in my opinion, but it’s not worth throwing out either.
Bloom’s taxonomy was helpful in grad school when I was beginning to “think” about the process of developing instruction. But, yes, it is cumbersome to actually “practice” and try explaining it to a superior or client who has absolutely no background in the field!
Training, learning – whatever label you apply – is about change and the “change process” needs to be managed as closely as the instructional development. And people fear change! Improved performance means “more work” to many employees and they simply are not embracing that concept!
We can continue to deconstruct these old models but where is the part about considering modern workplace life? What level does that fall into? I find that designers discuss learners as if they were all eager to attend/take a course or module. And we know that simply isn’t true.
I really like Brenda’s… “The Pure Performance Alternative” A more radical approach would be to have no taxonomy at all, to simply assume that all objectives are at the use level (i.e., “performance” objectives)…for an Enterprise Learning Context. Thanks for republishing this!
If you extend the content-performance matrix to include the additional two categories from Robert Horn’s original formulation of the Information Mapping® Method, you also have STRUCTUREs (things that have parts and functions) and CLASSIFICATIONS (types of things). While both of these might seem derivative, in some respects, they tend to help one think through the prerequisite knowledge topics one might need in order to carry out (“do”) a procedure.
It’s such a relief to discover I’m not the only ID who finds Bloom’s taxonomy cumbersome and redundant. I strongly believe the reason T&D has had such a difficult uphill climb proving value is just this: measuring learning is futile. Change in performance (behavior) is the only thing that affects the bottom line. I also believe that the recent research on how our brains motivate our behaviors will lead to even more refinements e.g. behavior/performance change is difficult and rare and therefore not likely to actually be affected by training or learning. Rather, a change in thinking is the only way to change behavior and that happens individually and with much effort. Coaching that provides one-on-one exploration and reflection of thinking will prove to be far more effective than training in helping individuals improve performance.