Cowles and Davis (2016) discussed several factors that contributed to the adoption of the .05 level of statistical significance. The first factor discussed was the discovery that computations can be utilized to measure observational error, by and astronomer, Lambert Adolphe Quetelet. This discovery led to the development of the normal distribution curve utilized today (Cowles & Davis, 2016).
Following the development of the normal distribution curve, a measure of variability known as probable error was discovered (Cowles & Davis, 2016). This discovery was the second factor mentioned by Cowles and Davis (2016) that contributed to the .05 level of statistical significance. Although the term probable error was first used in 1818 by …show more content…
The first criticism stated was, NHST does not provide the information the researchers are looking for (Kirk, 1996). NHST provides the probability that the researchers will obtain certain data if the null hypothesis is true; however, researchers want to know the probability the null hypothesis is true given the data they collected (Kirk, 1996). Kirk (1996) emphasized due to this confusion, researchers often make incorrect assumptions regarding the p value. He stated the utilization of NHST causes researchers to pay more attention to the p value and the rejection of the hypothesis, instead of focusing on if the obtained data supports the initial …show more content…
Practical significance is whether or not the results gathered are applicable or beneficial to the real world. Researchers must determine if the strength of the relationship within the results is strong enough to have real or practical meaning (Kirk, 1996). Kirk (1996) described how the criticisms of NHST led to the development of measurements that established what he referred to as measures of effect magnitude, which measured the strength of association, effect size, or other measures. He argued researchers should utilize confidence intervals and approximations of a point difference to assess practical significance (Kirk, 1996). Kirk (1996) claimed the calculation of the point difference and confidence intervals utilizes the same amount of information as NHST, and provides an exact range of numbers that the difference is probable to occur within. Moreover, it is easier for researchers to interpret results, and more difficult to disregard marginal effects, since the confidence intervals are reported using the same units of measure as the data (Kirk,