Abstract:Many studies use logistic regression models to investigate the ability of complexity metrics to predict fault-prone classes. However, it is not uncommon to see the inappropriate use of performance indictors such as odds ratio in previous studies. In particular, a recent study by Olague et al. uses the odds ratio associated with one unit increase in a metric to compare the relative magnitude of the associations between individual metrics and fault-proneness. In addition, the percents of concordant, discordant, and tied pairs are used to evaluate the predictive effectiveness of a univariate logistic regression model. Their results suggest that lesser known complexity metrics such as standard deviation method complexity (SDMC) and average method complexity (AMC) are better predictors than the two commonly used metrics: lines of code (LOC) and weighted method McCabe complexity (WMC). In this paper, however, we show that (1) the odds ratio associated with one standard deviation increase, rather than one unit increase, in a metric should be used to compare the relative magnitudes of the effects of individual metrics on fault-proneness. Otherwise, misleading results may be obtained; and that (2) the connection of the percents of concordant, discordant, and tied pairs with the predictive effectiveness of a univariate logistic regression model is false, as they indeed do not depend on the model. Furthermore, we use the data collected from three versions of Eclipse to re-examine the ability of complexity metrics to predict fault-proneness. Our experimental results reveal that: (1) many metrics exhibit moderate or almost moderate ability in discriminating between fault-prone and not fault-prone classes; (2) LOC and WMC are indeed better fault-proneness predictors than SDMC and AMC; and (3) the explanatory power of other complexity metrics in addition to LOC is limited.

Notes on Ott's cohesion measures for classes

Comments on ‘A Cohesion Measure for Object‐oriented Classes’ by Heung Seok Chae, Yong Rae Kwon and Doo Hwan Bae (softw. Pract. Exper. 2000; 30:1405–1431)

Measuring class cohesion based on dependence analysis

More comments on 'A cohesion measure for object-oriented classes' by Heung-Seok Chae, Yong-Rae Kwon and Doo-Hwan Bae

A comparative study of graph theory-based class cohesion measures.

A Novel Approach To Measuring Class Cohesion Based On Dependence Analysis

DMC: a more precise cohesion measure for classes

ICBMC: an Improved Cohesion Measure for Classes

A Pedagogical Evaluation and Discussion about the Lack of Cohesion in Method (LCOM) Metric Using Field Experiment

Family of Coherence Measures and Duality Between Quantum Coherence and Path Distinguishability

Quantifying Dynamical Coherence with Coherence Measures

An In-Depth Investigation into the Relationships Between Structural Metrics and Unit Testability in Object-Oriented Systems

Measuring Aspect Cohesion

Meta-analysis of Potentially Confounding Effect of Class Size on Associations Between Object-Oriented Metrics and Maintainability

Measuring Cohesion of Packages in Ada95

Effect of slice scope on data slice-based class cohesion metrics

Measuring Package Cohesion Based on Client Usages

Mathematical Foundations of Data Cohesion

Measuring Package Cohesion Based on Context

On the Ability of Complexity Metrics to Predict Fault-Prone Classes in Object-Oriented Systems

Object-Oriented Memory Consistency Model-OC