Assessing the practical significance of empirical results in accounting education research: The use of effect size information

Robust Statistical Methods for Empirical Software Engineering

“…We agree with Arcuri and Briand (2014) that effect sizes are extremely useful, as they provide an objective measure of the importance of the experimental effect, regardless of the statistical significance of the test statistic. Furthermore, effect sizes are much less affected by sample size than statistical significance and, as a result, are better indicators of practical significance (Madeyski 2010;Urdan 2005;Stout and Ruble 1995). Cohen (1988Cohen ( , 1992 was the first person to propose interpretation guidelines for effect sizes, by suggesting criteria to define a small, a medium or a large effect for use in the behavioural sciences.…”

Preparing accounting students for ethical decision making: Developing individual codes of conduct based on personal values

“…The effect size (eta 2 ) is large according to Cohen (1988) and this likely represents a practically significant finding given that the relatively high pre-test mean may have created ceiling effects that limited the extent to which participant perceptions could change during the course of the exercise (see Stout & Ruble, 1995 for a discussion of practical significance in the accounting education literature). The significant interaction with student level indicates that self-awareness increased more for the undergraduate students than the MBA students, which is likely due to MBA students reporting higher awareness of their value system prior to completing the exercise.…”

The effect of the 150-hour requirement on new accountants’ professional commitment, Ethical orientation, and Professionalism

“…The final column of Table 3 reports the eta-squared statistic for each dependent variable. Eta-squared measures the proportion of variance shared by two variables (Stout & Ruble, 1995). In this study, all the eta-squared statistics are less than 0.01, indicating that less than 1% of the variance in each dependent variable can be explained by education level.…”

“…The 5-year graduates had marginally higher mean score than the 4-year graduates (t = 1.54, p = 0.06, one-tailed). Stout and Ruble (1995) recommended reporting effect size to evaluate the "practical" significance of educational treatments. The final column of Table 3 reports the eta-squared statistic for each dependent variable.…”