From Theory to Spreadsheet: Evaluating the Pedagogical Effectiveness of Excel Solver in Teaching Modern Portfolio Theory

Spreadsheet optimisers have made the Markowitz model easy to compute in class, but computing a portfolio is not the same as understanding it. This study asked how far Excel Solver supports learning of modern portfolio theory and whether the design of a Solver task changes what students learn. Three dimensions were evaluated, namely procedural accuracy, construction of the efficient frontier, and reasoning about Solver output. Using qualitative document analysis, the study examined practicum reports written by student groups in two accounting classes. In the first class, groups set their own risk targets, compared Solver portfolios with randomly weighted ones and chose a best portfolio. In the second, groups traced fifteen efficient portfolios and drew the frontier. Every report was coded on the three dimensions, and its key figures were reproduced from the students’ own price data. Solver performed reliably as a calculator, but its value for learning fell sharply when students had to place the frontier correctly and interpret the results. The two designs helped in complementary ways. Drawing the frontier led students to anchor their targets at the lowest-risk portfolio, whereas random portfolios supplied a ready benchmark that invited dominance comparisons. Weights that Solver drove to zero were the strongest prompt for conceptual explanation. Solver therefore becomes a learning tool rather than a black box only when the task obliges students to begin at the minimum-variance portfolio, benchmark the output and justify every weight.