The Unit of Analysis Is the Whole Argument

A colleague once told me that students "just won't use the platform." Three explanations arrived within the same ten minutes. The students are disengaged. The interface is bad. Nobody trained them properly.

Each explanation is plausible. Each one is also doing the same thing, which is why the conversation went nowhere: all three locate the problem inside a single element. One blames the person. One blames the tool. One blames the transmission of skill from tool to person. Change the person, change the tool, or connect them better — those are the only moves available once you've decided what you're looking at.

Activity theory's contribution is not a better answer to that question. It is a refusal of the question's framing. It says: you have chosen your unit of analysis, and your choice has already determined the range of explanations you can reach.

That is the whole idea. Everything else in the tradition is elaboration.

Three enlargements

The history of activity theory is best read as a sequence of arguments about how big the unit should be — and each enlargement changed what counted as an explanation.

Vygotsky made the first move. Human action is not a direct line from stimulus to response; it is mediated by tools and signs. The child who ties a string around her finger has reorganized her own memory by putting something into the world. The mediating artifact isn't an aid to thinking. It is part of the thinking (Vygotsky, 1978). Unit of analysis: a person acting with a tool on an object.

That is already enough to dissolve half of the platform conversation. If the artifact is constitutive rather than instrumental, then "the students plus the platform" is a different cognitive system than "the students," and asking what students can do in the abstract is asking about a system that does not exist.

But it is not enough. A person acting with a tool is still an individual, and individuals do not decide most of the conditions of their own action.

Leont'ev made the second move: individual actions get their sense from the collective activity they belong to. His example is a beater in a primeval hunt, who drives game away from himself — an action that is unintelligible on its own and perfectly rational inside a collective hunt with a shared motive (Leont'ev, 1978). Unit of analysis: collective activity, oriented to an object, within which individual actions become interpretable.

This is the move most people skip, and skipping it is why so much writing that cites activity theory reads like ordinary psychology with unusual vocabulary. If you have not identified the collective activity and its object, you cannot say whether an action is rational. You can only say whether you would have done it.

Engeström made the third move: draw the whole system. Subject, object, mediating instruments — plus the rules that constrain, the community that shares the object, and the division of labor that distributes the work (Engeström, 1987/2015). Then draw two of them, interacting, negotiating a partially shared object.

Now the platform conversation looks different. Whose object is the platform serving — the student's, the advisor's, or the reporting requirement's? What rule makes non-use rational? Who else is in the division of labor, and what does the platform ask them to do that they are not resourced to do? Cole and Engeström (1993) make the general form of this point: cognition is distributed across historically accumulated cultural artifacts, not merely across the people in the room.

Why the choice of unit is the theory

Here is the claim I want to make, because I think it is the thing most often lost when activity theory gets summarized.

"You cannot train an individual out of a systemic contradiction."

If your unit of analysis is the individual, every finding will be a deficit, and every intervention will be a workshop. This is not because the researcher is careless. It is because the unit determines the vocabulary of causes available, and an individual-level unit only has individual-level causes in it. The result is a literature full of interventions that work in the study and evaporate in the setting, and a workforce that has learned to experience structural impossibility as personal inadequacy.

Enlarging the unit is not a gesture toward complexity or a nod to context. It is a substantive empirical bet: that the explanation lives in the arrangement, and that you will not find it by looking harder at the person standing in front of you.

This is also why the tradition insists on cultural-historical activity theory rather than plain activity theory. Culture and history are not the backdrop against which the activity happens. They are inside the unit — sedimented in the artifacts, the rules, and the division of labor, all of which arrived from somewhere and carry the shape of the problems they were built to solve (Cole, 1996).

A three-question test

Because it is easy to borrow this vocabulary without doing the work, here is the diagnostic I use on my own drafts. If I cannot answer all three, I am not doing activity theory. I am decorating.

• What is the object? Not the goal, not the deliverable, not the KPI — the thing the collective activity is working on and for, which is usually contested and rarely stated. If the object in your analysis is the same as the object in the org chart, look again.
• Who is in the division of labor? Including the people who are not in the room, are not on the project, and have not been asked. Especially those.
• What rule makes the irrational action rational? Every persistent behavior that looks like resistance, laziness, or incompetence is rational from somewhere. Find where.

Kuutti (1996) argued this case for human-computer interaction thirty years ago — that the individual information-processing unit could not see what it needed to see — and Kaptelinin and Nardi (2006) built it out into a design framework. The argument has not gotten weaker with age. It has mostly gone unheeded, which is a different thing.

The claim

Activity theory is often introduced as a lens, which makes it sound optional and slightly decorative — a way of seeing the same facts in a warmer light. It is not a lens. It is a decision about where to draw the boundary of the thing you are explaining, made before any evidence is gathered, that determines what your evidence will be able to say.

Roth and Lee (2007) called their survey of the field "Vygotsky's neglected legacy," and the neglect they were describing was not a failure to cite him. It was a failure to accept the cost of his argument: that if mediation is constitutive, the individual is the wrong unit, and a great deal of comfortable research design has to be given up.

So the students are not disengaged, and the platform is probably fine, and nobody needs another training. Somewhere in that activity system there is a rule, an object, or a division of labor that makes non-use the sensible thing to do. That is a harder finding to report and a much better one to act on.

The opening scene is a composite drawn from several settings; no individual or institution is described.

References

Cole, M. (1996). Cultural psychology: A once and future discipline. Belknap Press of Harvard University Press.

Cole, M., & Engeström, Y. (1993). A cultural-historical approach to distributed cognition. In G. Salomon (Ed.), Distributed cognitions: Psychological and educational considerations (pp. 1–46). Cambridge University Press.

Engeström, Y. (2015). Learning by expanding: An activity-theoretical approach to developmental research (2nd ed.). Cambridge University Press. (Original work published 1987)

Kaptelinin, V., & Nardi, B. A. (2006). Acting with technology: Activity theory and interaction design. MIT Press.

Kuutti, K. (1996). Activity theory as a potential framework for human-computer interaction research. In B. A. Nardi (Ed.), Context and consciousness: Activity theory and human-computer interaction (pp. 17–44). MIT Press.

Leont'ev, A. N. (1978). Activity, consciousness, and personality. Prentice-Hall.

Roth, W.-M., & Lee, Y. J. (2007). "Vygotsky's neglected legacy": Cultural-historical activity theory. Review of Educational Research, 77(2), 186–232. https://doi.org/10.3102/0034654306298273

Vygotsky, L. S. (1978). Mind in society: The development of higher psychological processes (M. Cole, V. John-Steiner, S. Scribner, & E. Souberman, Eds.). Harvard University Press.