Abstract:

High failure rates in Fundamentals of Electrical Engineering courses, reaching up to 70% in some engineering programs, have motivated extensive research into the sources of assessment difficulty. Prior empirical work established student proficiency models using Item Response Theory (IRT), identifying temporality and mathematical abstraction as competence barriers, but without a structural explanation for why certain item properties produce difficulty.

The present study analyses the content and task structure of assessment items in Fundamentals of Electrical Engineering and introduces two complementary analytical instruments developed within a cognitive load framework. The first is a Dimensional Content Analysis, which identifies and operationalizes six content dimensions of assessment items through observable indicators. Essentially, the scale levels within each dimension are defined not as content categories but as qualitatively distinct cognitive demands, where each level represents a different type of cognitive processing required, rather than a greater quantity of content. The sum of the six dimension scores forms the Index of Dimensional Complexity (IDC), capturing the breadth of cognitive demands present in an item.

The second instrument is the Functional Order (OF), which measures the depth of qualitatively distinct cognitive function chaining required to solve an item. Each time the output of one cognitive function must be stored and reused as input to a qualitatively different cognitive function, the functional order increases. This measure thus captures the cognitive depth an exercise demands, reflecting the degree to which the students must integrate to reach the solution, and contributes to the total cognitive load imposed by the task.

Both instruments were applied retrospectively to eleven structurally independent items from an authentic Fundamentals of Electrical Engineering exam whose IRT difficulty parameters were empirically known (n = 196 students). Spearman rank-order correlations between each predictor and IRT difficulty yielded: IDC alone rs = .460 (p = .155); OF alone rs = .737 (p = .010). When both factors were combined, the correlation increased substantially to rs = .804 (p = .003), confirming that dimensional content analysis and functional task structure represent distinct but complementary perspectives on item difficulty. While each predictor captures a different aspect of cognitive demand, together they account for a substantially larger portion of the empirically observed difficulty variation, yielding a Spearman correlation of rs = .804.

The findings offer a theoretically grounded basis for predicting and calibrating item difficulty prior to assessment, with practical implications for examination design and the didactic sequencing of content in engineering education.

Keywords:

Fundamentals of electrical engineering, item difficulty, cognitive load, assessment design, item response theory, content analysis, content dimensions.