Choosing theories about ideas
So you’d like to do a research project which focuses on students’ ideas. Perhaps you’re curious about how they think about a particular topic, or how their thinking evolves over time, or how to help them think about a topic in a new way. You know they enter your class with existing ideas about this topic, and you need a way to conceptualize their existing ideas, the development of their ideas, and the new ideas you’d like them to have. You need a theory of what ideas are.
Different kinds of ideas are sometimes called “concepts”, “resources”, “conceptions”, “knowledge pieces”, and a thousand other similar-sounding vocabulary terms (each of which may have a very particular definition, depending on the theory). There are a lot of linguistic landmines in this space. I’m going to use “ideas” as an umbrella category to mean all of these things.
This article compares two major theories for ideas, Misconceptions and Resources. For the most part, they are very similar. However, there are some subtle differences in these two theories which have deep implications for what kinds of research questions you ask, how you think about students’ ideas, and what good curricula “look like”.
Major similarities
Scope and history
Both Misconceptions and Resources are about what ideas are and how ideas develop. Both theories have long histories in discipline-based education research (and the sister fields of Learning Sciences and the Scholarship of Teaching and Learning). Both theories can – and have! – been used in qualitative and quantitative research studies, to develop curricula, and across many different topical areas in STEM.
There are some historical differences which are no longer true.
- People who use Resources like to say that Misconceptions use a “unitary ontology” (ideas are large and hard to change) but more recent work using Misconceptions has embraced a “manifold ontology” (ideas can be small and context-dependent).
- People who use Misconceptions note that Resources’ ontology was originally too underspecified and vague, but more recent work using Resources has explored the theory in a lot more depth.
Theories generally come in families of related theories. The two families in this article are Knowledge-in-Pieces, of which Resources is a member, and Misconceptions, of which Difficulties is a member. Knowledge-in-Pieces is the common label for its family of theories, but Misconceptions doesn’t really have a common label for the family. Researchers use many labels to ambiguously mean the family or a member of the family: misconceptions, conceptions, concepts, alternative conceptions, difficulties, etc. This is synecdoche in action: the member of the family stands for the family as a whole. I’m going to use “Misconceptions” to mean both the family and the theory within it.
Conceptualizing “ideas”
Ideas are something that an individual has in their head.
- Ideas can be big (like a theory of evolution) or small (like the first four digits of pi) or in between. Some ideas are factual, others are beliefs, and yet others are procedural or relational. The space of ideas is pretty nebulous; that’s ok for now.
- Everyone has ideas about a lot of things. Sometimes, my ideas about something are functionally the same as lots of people’s ideas about that thing. Sometimes, they’re slightly different, and other times they’re radically different.
- Because people grow and change, their ideas grow and change too. The mechanisms for how this happens can be really complicated.
- Some ideas are explicit, meaning that people tell you about them directly, and others are implicit, meaning that you need to infer them from what people are saying or doing.
- Some ideas are really stable across different contexts and times, and others are more evanescent or context-dependent.
- Some researchers (e.g. neuroscientists) are really concerned about biological correlates of ideas, or how groups of people can hold knowledge together, or the differences between knowing (a verb or action) and knowledge (a state or thing). These considerations are outside the scope of this article.
For the most part, Resources and Misconceptions share all of these points for what ideas are. Different studies focus on different sizes of ideas, but the theories as a whole cover the same range. Both theories agree that some ideas are relatively stable, while others are volatile and exceptionally context-dependent. Both theories agree that some ideas are normatively correct while others are not, and that whether an idea is correct sometimes depends on the context.
Fundamental differences
Theories are something that researchers use to understand their participants’ ideas. Resources and Misconceptions, as tools for researchers, have some fundamental differences with implications for research design.
Theories use epistemological commitments (what is good or important about ideas) to suggest methodological choices (how researchers should uncover or change participants’ ideas). That’s a lot of terminology this brief article, but it’s important: this article is about comparing the epistemological commitments and ensuing methodological choices for two different theories.
Theories live or die on whether they are valuable to meaningfully address research questions. If your research question’s epistemological commitments don’t match your theory’s, you won’t be able to meaningfully answer your question.
Valuing students’ ideas
These theories have epistemological commitments about the value of students’ ideas: what kinds of ideas are good, what is important to pay attention to as a researcher, and what are good ways to uncover or change those ideas. While the ontologies of what ideas are are largely the same for Resources and Misconceptions, their epistemological commitments have dramatic differences.
An asset-based epistemological commitment tells us to look for what is good and productive in students’ ideas. This commitment requires that we think about students’ ideas as necessarily having good parts, perhaps misapplied in a particular context, and therefore it requires that students’ ideas are small enough to seek the good ones. Big ideas are made up of smaller ones, and productive reasoning means combining small ideas together to solve problems (or at least make headway on them). There is a corresponding emphasis on whether reasoning is productive (it helps people move forward in solving problems), rather than normatively correct. This is a subtle difference, but an important one: sometimes ideas that aren’t normatively correct can still help us figure out what’s happening, and sometimes what counts as normatively correct is complicated and unhelpful.
In contrast, a deficit-based epistemological commitment tells us to look for what is wrong in students’ ideas. This commitment requires that we pay attention to ideas which have normatively correct versions, so that we can compare the students’ wrong ideas to the normatively correct ones. This commitment doesn’t say much about the size of ideas: big ideas can be wrong if some of their parts are wrong, and small ideas can be wrong alone or they can just be wrongly applied. The emphasis here is on comparing students’ ideas to a normatively correct one, and seeking the ways that the students’ ideas don’t match.
These two epistemological commitments are exactly opposed to each other.
Resources
Resources firmly requires an asset-based epistemological commitment. Stories about student learning that use Resources focus on (a) identifying elements of student thought; (b) describing how students put those elements together to make arguments and solve problems; and (c) highlighting what’s interesting or cool about this reasoning, not what’s wrong.
For Resources analysis to really shine, you want to ask questions that permit messy reasoning and multiple pathways to the right answer. You don’t want to ask “gotcha” questions where there’s only one right way to arrive at a single right answer, because questions like these might inhibit students from reasoning productively, and therefore you might miss some of their great ideas. In contrast, a question with multiple pathways allows you to celebrate the diversity in student reasoning.
It’s definitely possible to analyze data from single-path questions using Resources. (See: Bahar Modir’s paper using secondary analysis of other people’s data on quantum mechanics.), but more nuanced analyses are possible when you have more open-ended problems (See: Savannah Mitchem’s paper on how students think about electric and magnetic fields in materials.). This data collection emphasis on multiple pathways to success is a direct consequence of the epistemological commitment to looking for productivity and connections in students’ reasoning.
Misconceptions
In contrast, Misconceptions comes very firmly from a deficit-based perspective of student knowledge. It’s right there in the title: misconceptions are conceptions that are wrong! As a consequence of this epistemological commitment, research with this theory tends to compare students’ answers to a normative “complete and correct” answer. Students’ answers are graded by how far they fall from the normative answer, and how resistant they are to changing their ideas after instruction. Another consequence of this commitment is that research using Misconceptions can struggle to to figure out why students are wrong: are they wrong because their ideas are wrong-yet-stable, or are they wrong because their ideas are wrongly-applied-and-flexible? Some research using Misconceptions attempts to delve into the context-dependence of students’ ideas, while other research aims to minimize their flexibility to find out what students “really think”.
In Misconceptions research, there’s a substantial emphasis on asking questions to elicit specific ways of being wrong, and of tuning your questions so that (a) students cannot “accidentally” get the right answer; (b) surface features of the problem are designed to lure students into wrong answers; and (c) there is a single correct answer and a single correct path to getting that answer. There is a ton of expertise that goes into creating questions like this and showing that they can get the same results with many different students (validity and reliability testing). This data collection emphasis on tightly constrained questions with attractive distractors is a direct consequence of the epistemological commitment to seeking deficits in students’ reasoning.
Complete studies
Another set of epistemological commitments are about what education research studies need to accomplish in order to be a complete study. Unlike asset-based and deficit-based epistemological commitments, the commitments in this group are not mutually exclusive.
Education research needs to teach students. Under this commitment, it is not enough to merely figure out what ideas people have or how they apply them; you must develop an intervention to change their ideas, ideally for the better. Usually interventions look like curricula, but other options are possible. Because curricula are supposed to help lots of students, not just a few, research with this epistemological commitment almost always includes quantitative components as well as before-and-after comparisons of student knowledge.
Education research needs to engage with the humanity of students. It’s not enough to identify what ideas come up as a function of context; you must engage with the narrative work of deeply understanding students’ ideas and exploring the variation among them. While some research with this epistemological commitment is quantitative in nature, engaging with humanity is a messy process. Complete studies are either wholly qualitative or mix both quantitative and qualitative components.
Education research needs to suggest mechanisms by which different contexts can elicit different ideas. It’s not enough to merely tabulate different ideas in different contexts; you need to have underlying reasons why those contexts elicit different ideas. Some research which takes up this epistemological commitment reaches to neuroscience for brain-based mechanisms; other research reaches to the social sciences for epistemic or social mechanisms.
Misconceptions
Misconceptions research almost always takes up the epistemological commitment to teaching students. Coupled to the commitment to deficit-based models, research using Misconceptions tends to have a substantial emphasis on “fixing” students, usually by developing curricula to improve their success rate on the carefully chosen questions which elicit their misconceptions. Naturally, this research tends to be quantitative and driven by counting statistics. For example, we gave this test to 100 students, or we interviewed 30 students using these questions. Our results show that students are 60% were wrong in this way; 20% wrong in that way; 10% correct; 10% other.
Occasionally, Misconceptions research engages with the humanity of students, usually in support of the primary goal to teach students.
Resources
Resources, because it seeks productive “seeds of goodness”, tends to seek robust descriptions of how students reason, either alone or together. Research using Resources can be qualitative or quantitative, but it tends to be qualitative and driven by narratives. You can definitely support those narratives with counting statistics, but the central work of the research study is to deeply understand students’ ideas and explore the variation among them, not to tabulate responses or reduce them to comparisons with a normative response. This emphasis on understanding students’ responses comes from the epistemological commitment that education research needs to engage with the inherent humanity of the participants in order to be a complete study.
Research using Resources sometimes takes up the epistemological commitment of teaching students. There are some brilliant curricula that were developed using Resources. However, most studies using Resources do not take up this epistemological commitment.
So, what’s better?
The best choice is the one that you make mindfully; that promotes coherence between your research question, methods, and theory; and that gets you where you want to go. In this short overview of these two theoretical families, there’s so much nuance that I’ve lightly skipped over.
But.
I have a personal epistemological commitment to asset-based models for humans, so I use Resources. It’s just better.
No story here.
If Imani uses Resources, she’ll look for how students bring ideas together, and she’ll try to build questions in pairs so that they can show off how ideas connect together. If she uses Misconceptions, she’ll look for how students bring in wrong ideas, and she’ll try to build questions with a lot of appealing distractors to induce them into wrong answers. Imani’s choice of theory tells her how to build her instrument.
Imani’s choice of theory also suggests how she should interact with her students and their ideas, because it tells her what might be broken and a path forward to fixing it.
Additional topics to consider
History
This article was first written on July 23, 2023, and last modified on March 13, 2024.