“Researchers generally believe that the consequence of making a Type I error are more serious than those associated with a Type II error” (Cozby, 2009, 259). There are a couple of statistical power issues that may arise, 1) a Type I error or 2) a Type II error. A Type I error occurs when the null hypothesis, although true, gets rejected by the researcher whereas a Type II error occurs when the null hypothesis is accepted, but is false. It is better for researchers to obtain a Type II error because a Type I error would mislead those who are reading about these studies or experiments. Type II errors are viewed as less severe than Type I.
Factors that might influence statistical power include: choosing a significance level. Significance levels such as: 05 and .01 are commonly used among researchers. According to Cozby (2009), “The significance level chosen by the researcher usually is dependent on the consequence of making a Type I error versus a Type II error” (pg. 259). Random errors usually carry a lower significance level than those previously mentioned, however “a meaningful result is more likely to be overlooked when the significance level is very low” (Cozby, 2009, pg. 260). Aside from choosing a significance level, a Type II error could be due to a small sample size or a small effect size. If the sample size is too small there is a possibility that no relationship may be determined and if there is a relationship, it may be easily overlooked. The problem with a small effect size is that according to Cozby (2009), “Very small effect sizes are difficult to detect without a large sample size” (pg. 260). Therefore, researchers should make sure that the significance level, effect size, and sample size reflect an appropriate size in order to receive accurate results thus eliminating the possibility of Type I and/or Type II errors occurring.

Cozby, P.C. (2009).