In the e-learning world, a common element of course design is the practice of sequencing the learning objectives of courses, modules, and content according to the principles of Bloom’s taxonomy.
Instructional designers typically use Bloom’s taxonomy to structure the learning experience. The course would start with the foundations: things learners need to remember, understand, and apply. As they progress, learners are challenged to complete tasks that require progressively higher orders of thinking — as defined by the taxonomy — until they have mastered the necessary skills. These higher orders challenge learners to analyse, evaluate, and create.
Instructional designers often adopt Bloom’s taxonomy because it is the only scaffolding framework they are taught. But, did you know Bloom’s taxonomy was not intended to serve as a content scaffolding framework? In fact, many curriculum designers find that it's getting in their way. Let's see why that might be the case for your project.
Benjamin Bloom and his team of cognitive psychologists published Taxonomy of Educational Objectives, Handbook 1: Cognitive Domain as the cognitive component of their theory of development in 1956. The original intention was to help educators design assessments that would target several levels of cognitive functioning at the same time, without needing to assess each order of thinking individually.
This theory was then revised in 2001 by another team of scholars, led by Lorin Anderson, to create the version of the taxonomy that is widely used today. Many instructional designers think of this revision as the moment where the taxonomy changed from a hierarchy of nouns to the hierarchy of verbs pictured below, but the more important shift that happened with Anderson’s publication is how the taxonomy is used in practice.
.png)
Whereas Bloom and his associates conceptualised the taxonomy as a framework for designing efficient and effective assessments, Anderson’s version expands Bloom’s taxonomy to incorporate aspects of metacognition and knowledge acquisition, aligning more closely with theories of learning rather than just assessment.
To put it differently: Bloom's original intention was for the taxonomy to be used as a means to categorise the orders of thinking tested within a single question. This way, questions can be designed to test various orders of thinking at the same time. Let's consider an example.
Bloom's taxonomy demonstrates how an instructional designer can design a question at the Apply level and, if they do this correctly, safely assume that Remember and Understand will also be tested in that question. They do not need to set additional questions to test the lower orders. It's a great way to design efficient assessments; you can test the same number of objectives in five questions as you would otherwise have done in 15, and we all know how important it is to reduce "seat time".
But this is rather different to how Bloom’s taxonomy is used by many proponents of Anderson’s version, who apply it as a form of scaffolding learning rather than designing assessments. This is an important shift, as learning and assessment are different processes with distinct purposes, forms, outcomes, and cognitive involvement.
Simply put, learning involves creating and storing knowledge, concepts, and skills. Assessment checks whether these have been stored correctly and challenges learners to retrieve them from long-term memory.
Of course, that is not to say that assessment is not an important component of learning in the broader sense, but it is important to acknowledge that assessment is a special beast that should be handled with extreme care.
What happens when the principles that underpin a theory of assessment are applied as an overall approach to scaffolding a learning pathway? Well, in short, it can go pear-shaped quite quickly.
The concept of scaffolding learning was popularised by the psychologist and social constructivist Lev Vygotsky as part of his theory of the Zone of Proximal Development in 1978. When scaffolding learning, the aim is to provide a learner with just the right amount of support they need to achieve the next goal in their learning. Not so much assistance that they feel bored! Just enough to get them over the line while trusting them to do most of the work.
This type of scaffolding can only be achieved by first checking what the learner already knows, then relating the new content to that existing knowledge, then breaking the new content up into smaller tasks that challenge the learner to construct new knowledge, all with frequent opportunities for constructive feedback. We can scaffold at the level of a degree programme, a course, a module, a lesson, and even within an individual video script. Scaffolding is everywhere, and it's even harder to do meaningfully than it sounds.
What should we take away from this? Regardless of the context, for scaffolding to be effective, the first step must be to check for prior knowledge to determine where the first set of scaffolds may be required. In e-learning, this would manifest as the learner first taking a pre-assessment, after which the learning path is differentiated to exclude that which the learner has already mastered.
If the path is not differentiated, the standard path likely contains too much information for the learner to be challenged. Most instructional designers, empathetic creatures that they are, would rather include too much information than risk omitting something important. The learning experience is then designed for the one or two learners who need the most help, and thereby becomes inappropriately scaffolded for most of the group. The learning is also likely more didactic than it needs to be.
The resulting design thus actively works against the principles foundational to Vygotsky’s theory, and can no longer be called scaffolding. When students are not challenged, they lose all reason to engage; this is even more true of Gen-Z learners.
Any framework applied in the service of scaffolding can only really be effective in e-learning if the material is differentiated based on a pre-assessment. So, in the absence of a pre-assessment, it's bad news not just for our Blooms-based scaffolding approaches, but for scaffolding in general.
So, you might ask, how is that the fault of Bloom's taxonomy? The answer is, it isn't, it's just a challenge with our approach to scaffolding more generally, and I wanted to mention this regardless of the fact that I had to write about Bloom's. However, Bloom’s taxonomy as a scaffolding framework also poses some fairly unique challenges that are not present in other approaches. Let's take a look at those now.
Sequencing lessons or tasks in a course or module based on the order of thinking that those tasks require is not always the best approach to scaffolding the pathway. Consider the below example. I think, depending on their level of understanding of math, most instructional designers would agree that the below module is well-scaffolded:
The foundational principles are taught, after which those principles are applied in a slightly more complex type of equation in Lesson 2, after which both the principles and the more complex type of equation are applied in a range of even more complex combinations and examples. The lessons certainly increase in difficulty as the module progresses.
However, as you may already have observed, all three lessons operate at the “Apply” level of Bloom’s taxonomy. In many cases, Bloom’s taxonomy is a rather unimportant consideration when scaffolding a module, so if this is an instructional designer’s primary approach to scaffolding the course design, they are bound to overlook some important considerations.
One could certainly consider using Bloom’s to scaffold the content within these lessons. For example, one would likely start Lesson 1 by introducing what the five rules of exponents are (remember), followed by some worked examples (understand), and some do-it-yourself examples (apply).
But, this approach will only be effective for specific types of material. Let’s explore what these criteria are next, at which point it will also become clearer why I subjected you to a math example.
There are many learning pathways to which Bloom’s taxonomy can’t be applied meaningfully, regardless of whether you're using it as an approach to assessment design or an approach to scaffolding a 10-minute video script.
Bloom’s taxonomy is only appropriate for application to what Roland Case terms “tightly-aligned” topics. When assessing a “tightly-aligned” topic using Bloom’s taxonomy, one can assume that anything assessed at the “recall” and “understand” levels will also be assessed at the “apply” level. Adapted as a theory of learning, this means that a learner who can successfully complete "Lesson 1: Use the five rules of exponents" can be assumed to recall the five rules and understand how they work. That's all good; but, we know that math is an unusually hygienic subject.
Themes that instructional designers work with often relate to one another much more loosely. Even within a single lesson, using Bloom’s taxonomy to scaffold the content could be a challenge. For example, the fact that I can analyse the results of a usability test (analyse) definitely doesn't mean I can use best practice principles to run a usability test (apply). Slowly, over time, instructional designers everywhere have become lax with thinking critically about how closely related the tasks in their pathway might be, and whether being able to perform X task really does mean I am able to perform Y task simply because X task is front-loaded with a higher-order verb.
Regardless of whether you're using it as a theory of assessment or a principle by which to scaffold your learning pathway, be very strict with yourself on how tightly aligned those tasks really are before you declare that a learner has mastered an objective.
I've always felt that "understanding" has an unfair reputation, and I blame Bloom's taxonomy for this. Can one truly understand a theory without putting it into practice? I would imagine I can convince myself I understand how generative AI works if I can explain it to you, but my explanation would sadly be rooted in a combination of recall and my natural aptitude for paraphrasing. If that's all it takes to "understand", I'm sure you'll agree this a very depressing view of understanding.
I would prefer if we, as a community of practice, could agree that understanding requires a bit of experience and an ability to work with a concept from unexpected directions, including performing tasks such as analysis and evaluation. "Lizzy can help you with scaffolding, she's the one who really understands these things" should mean more than "Lizzy can paraphrase a definition of scaffolding that she can remember." I would expect Lizzy to be able to tell me what is wrong with the way that I scaffolded my module, in words that I understand, using examples she constructed from her own experiences with scaffolding.
Is this just a matter of semantics? Well, yes, but that doesn't mean it's not important. If my goal is to ensure all the learners "understand", and I enlist this sad view of "understanding" as my measure of success, then surely my learners are worse off than those whose educational experience was designed by someone with a bit more respect for the verb.
The "understand" of Bloom's taxonomy, which is something to be achieved prior to "apply", is not my "understand", and education thinkers like Sam Wineburg and Jack Schneider agree.
Although Bloom’s taxonomy can at times be a useful tool for ensuring all learning objectives are assessed efficiently, there are multiple problems with how it is often used as a method for scaffolding learning experiences:
Does this mean that Bloom’s taxonomy no longer has a place in learning design? Certainly not. It remains a useful way to think about assessment design, and even the development of exercises and activities. But, when it comes to scaffolding a learning experience more broadly, our approach needs to be more robust.
