Better than Bloom?

A recent Google search for “Bloom’s taxonomy” reveals over 50,000 hits.

After almost 50 years, Bloom’s taxonomy is still being used by educators and trainers as a pedagogical tool for the analysis of learning objectives. Originally designed as a method for the development of test questions, the six levels of the cognitive domain (knowledge, comprehension, application, analysis, synthesis, and evaluation) have become almost standard in the “learning business”.

I used Bloom’s taxonomy about ten years ago, while developing an estimate for the cost of CBT development.

We assumed that the higher the level, the higher would be the cost. With hundreds of performance objectives, we quickly reduced the six levels to three, but I now realise that there could have been many other ways to address the problem.

For instance, in Problems With Bloom’s Taxonomy (PDF) Brenda Sugrue states that Bloom’s taxonomy is invalid, unreliable and impractical. According to Sugrue, the six levels of Bloom’s taxonomy for the cognitive domain ” … are not supported by any research on learning.” Basically the taxonomy was a “best guess” by some knowledgeable educators of the time. The six levels make for nice matrices and provide a simple tool for analysis and evaluation, but Sugrue shows an even more effective way to create a Content-Performance matrix. Sugrue is not the only person who considers Bloom’s taxonomy passé. Another critic of the taxonomy is Robert Lewis [dead link] Professor of Knowledge Technology at Lancaster University.

Unfortunately, old chestnuts like Bloom’s taxonomy stay around longer than they should, because after a while we take them for granted. Every once in a while, it’s good to take a long, hard look at our practices, and make sure that we are using proven methods, and not second-rate tools.

Update: Donald Clark adds to the fire, stating that, “The weaknesses are so strong that it is simply sensible to abandon it altogether.”

Also, Carl Bereiter & Marlene Scardamalia of OISE offer Beyond Bloom’s Taxonomy: Rethinking Knowledge for the Knowledge Age.

9 thoughts on “Better than Bloom?”

  1. Bloom’s taxonomy is really not helping us develop education, even if it can act as a memory aid for a person who really have not thought at all about different types of knowledge. But then there’s a risk that the image of the pyramid stays in that person’s mind as The Model of Knowledge.
    If someone wants to use the taxonomy, I would recommend the revised version by Anderson and Krathwohl from 2001. Krathwohl was a member of the original group who developed the taxonomy in the fifties.
    You can see a comparison of the old and the revised taxonomy at:

    Rgds, Romi

  2. Taxonomies, including Bloom’s, can be abused in their apparent simplicity. Bloom’s taxonomy was designed for assessment only, not as a prescriptive measure, which is what is recommended in the College of Education article, that you reference, Romi. As Mary Forehand states in that article:

    “The cumulative hierarchical framework consisting of six categories each requiring achievement of the prior skill or ability before the next, more complex, one, remains easy to understand.”

    Whereas, this Carnegie Foundation article by Lee Shulman shows the weakness of such an approach:

    “Quickly, then, the taxonomies moved from being a scoring rubric and vehicle for communicating about test items, to being a heuristic for instructional design.”

    “Another thing that happens to taxonomies, and it happened to Bloom’s, is that they come to be understood as making a theoretical claim about sequentiality and hierarchy, suggesting that the only legitimate way to learn something is in this particular order. The implication of sequence and hierarchy within taxonomies obscures their true value, because taxonomies are not and should not be treated as theories.”

    I haven’t seen any data that show that people actually learn in the sequence of Bloom’s taxonomy, revised or not. The revised model was developed in the same manner as the original; a bunch of experts getting together and sorting it out. This does not approach the scientific validity of brain-based research using fMRI scans. I would recommend

  3. Thanks Harold for the comments and the valuable links! I’ll use them when I argue AGAINST Bloom’s taxonomy, which I do every now and then because of reasons you and I have mentioned.

    In my work as a HE teacher and project manager I have seen people use Bloom’s taxonomy in different contexts. Here in Europe we have the so called Bologna process going on in higher education. It has to a certain extent meant a regression to a behaviorist view of learning, due to the domination of the big, conservative European countries. Here in the Nordic countries, we look (at least on average) differently at learning and we are leaning more towards late modern/postmodern views, at least in theory. I think the progressive Canadians look at things from approximately the same point of view (that’s why I read your, George Siemens’ and Dave Pollard’s blogs every day :-)).

    So, I do admit that burying Bloom’s taxonomy would be the best solution. But the fact is that it is still used a lot. By pointing conservatives at the new version, I have sometimes been able to get a good conversation going. What does it mean that Evaluation has been replaced by Creating? And that Knowledge has been replaced by Remembering? I feel I have been able to point at some important developments in the traditional view of learning/assessment by using my opponent’s own tool, Bloom’s taxonomy, as a wedge.

    And by pointing at the fact that Krathwohl was involved in both, I can show that even personal views must change, and that “eternal” pedagogical truths have to be looked at very critically.

    But I realize that the way I expressed myself in my first post gives the impression that I actually accept the idea of taxonomies. I don’t.

  4. Sounds like a very pragmatic approach, Romi. Getting a conversation going could be the first step in getting people to think of learning as a complex process instead of filling specific boxes.

  5. Well, it is a pragmatic approach, because I’m a pragmatist, like Dewey ;-). And also an extentialist (at least for the time being…) which means that I have read some of Kirkgaard’s texts, including the famous viewpoint that if you want to teach someone, you have to start at the point where this person stands.

  6. I’ve used the revised Bloom’s taxonomy as a reference tool frequently in my role as an instructional designer. I agree partly with the discussion so far, Bloom’s taxonomy is not a learning theory, or a learning model, nor is it stepped or tiered approach, [each level building on the other]. Did Bloom plan it to be tiered or was this incorporated at a later date? I am not sure, but reading the reference from the link provided by Romi, it appears this tiered concept was added later.

    I use Bloom’s in several ways: 1) as mentioned in the article, as a reference tool when collaborating with other educators – to use a common language. The domains serve as a framework when planning learning activities and assessments in curriculum. For example when designing an assessment tool for a given module of a course, “should we use a multiple choice test to confirm student comprehension (understanding), or are we wanting the student to be engaging critical thinking skills, which would require the student to analyze what he or she learned, perhaps write and essay, and draw comparisons to other concepts and incorporate previous lessons materials”.

    2) As a tool to write instructional objectives, which drives the learning activities and assignments. Here is a link providing an overview of this function:

    Interesting and thought provoking discussion, thank you. Debbie


Leave a Reply to Romi Rancken Cancel reply


This site uses Akismet to reduce spam. Learn how your comment data is processed.