Both Field (2005) and Cohen (1994) emphasize the importance of effect sizes in psychology, discussing their benefits while recognizing some key flaws. Effect sizes provide a standardized way to measure and determine if an effect is small, medium, or large (according to Cohen’s guidelines) (Fields, 2005) . This standardization helps researchers compare findings across different studies. Effect size also provides a structured approach to summarizing results from multiple studies. Field (2005) points out that effect sizes are essential for meta-analyses, which support building knowledge in psychology rather than producing results that are difficult to compare. Cohen (1994) advises routinely reporting effect sizes with confidence intervals, as they contain all the information given by significance tests and help move the field beyond “significant/nonsignificant” outcomes toward more practical interpretations. Also, effect sizes enable researchers to distinguish between statistically significant and practically meaningful differences, promoting a more nuanced understanding in psychological research (Field, 2005).
Field (2005) also proposed that attention should be given to the common misinterpretation of the p-value. For example, a UK study found that 80% of undergraduate students incorrectly thought that p represented the probability that the null hypothesis was true, when in fact, p represents the likelihood of the observed data assuming the null hypothesis is true. Cohen (1994) argued that the assumption of a true “zero effect” null hypothesis is unlikely, since even small differences can become statistically significant with large enough samples.
Overall, when effect sizes are combined with confidence intervals, they help researchers in psychology to gather meaningful insights (Cohen, 1994). This approach supports the goal of enabling psychological science to build on existing knowledge and develop a more reliable body of research.
References
Cohen, J. (1994). The Earth is round (p < .05). American Psychologist, 49(12), 997–1003. https://doi.org/10.1037//0003-066x.49.12.997
Field, A. P. (2005). Discovering statistics using SPSS (2nd ed.).
