Consolidations as Whole Tasks: In One Cohort, Resource Users Scored Higher

Consolidation is where a tidy second-year module wobbles: goodwill, non-controlling interest, unrealised profit and dividends pile up in one question.

Huw Morgan and Anne Stafford of the University of Manchester built pre-lecture resources around that problem and tracked their use.

What they built

They used four-component instructional design (4C/ID), built around whole, realistic tasks that grow in complexity, to manage cognitive load on limited working memory (pp. 2, 4). They left out its fourth component, part-task practice, as not vital here (p. 6).

The 12 tasks formed three task classes: acquisition-date, post-acquisition and advanced consolidation adjustments, such as unrealised profit and goodwill impairment. Each ended with an unscaffolded final task (pp. 10, 14–16).

"PreView" videos opening each class explained the reasoning, such as control and goodwill. "ReView" videos demonstrated routine steps and stayed "on call" during tasks (pp. 11, 14).

What they found

The study grouped one 2017 cohort of 313 students into seven profiles by when and how often they opened the resources (pp. 10, 20–21).

On the 30-mark consolidation question, the regression's 38 non-users averaged 5.0 marks (p. 23). Adjusting for first-year accounting marks and two engagement measures, every user profile beat non-users by 5.50 to 11.3 marks; the top figure rests on 12 students (p. 23).

The authors read this as favouring preparation-aligned use over late revision (pp. 26–27). The unadjusted averages point the same way (pp. 23–24), but the paper does not test the profiles against each other. After adjustment, one late-revision profile (+6.38) slightly edged one the authors counted as appropriate (+5.96) (pp. 23–24).

Limits and context

Every student had the resources, so there was no comparison group (p. 10). Students chose how to engage, and the authors stress the results are associations, not causal effects (pp. 3, 28).

Platform hits do not show whether a video was watched in full, and the regression excluded 56 students who did not answer the question (pp. 22–23, 26, 28). The data come from one year at one university (p. 28).

Evidence strength: early signal. One cohort's correlational data and a design account, with no comparison group. It links resource use to marks but does not test whether the design works.