Code complexity: an empirical study based on the index of internal effort

Measuring the complexity of programming code remains a persistent challenge in computer science education, particularly when assessing student-developed projects. This study aims to evaluate the applicability of the Index of Internal Effort (IIE), a metric originally designed to quantify Explicit Intellectual Activities (EIAs), in the context of measuring the complexity of software code. The methodological approach involved analyzing two distinct datasets: (i) simple Arduino programs typically used in electronics and embedded systems courses, and (ii) source code from open-source projects published on public repositories such as GitHub. The study applied the IIEa, a version of the IIE adapted for complexity estimation, and compared its results to well-established software metrics, including LOC (Lines of Code), WMC (Weighted Methods per Class), and CC (Cyclomatic Complexity). The results revealed a strong correlation between IIEa and traditional metrics, supporting its applicability. However, due to the lack of data regarding the number of developers involved in the open-source projects, it was not possible to compute the average team effort using IIE0, a second equation within the IIE framework. The data analysis reinforces the viability of IIE as a complementary metric in programming education, enabling more accurate and fair assessments. Comparisons with prior studies indicate that cognitive complexity is closely associated with the structural properties of code. The findings suggest that the IIE can contribute both to more effective pedagogical practices and to future research aimed at integrating cognitive effort metrics into learning environments.