The article describes the importance of statistical power analysis in the field of psychology. Jacob Cohen was quick to note the critical role that applied statistics plays in this field. He also found out that mean power is key in identifying the medium effect sizes. The experiment was carried out using a fair cone which was tossed severally to obtain the results. The author accredited that the researchers ignored power analysis due to low levels of consciousness of its magnitude. They fail to realize the magnitude of power analysis to characterize psychology. This neglect was obvious through casual observation and confirmation by other power reviews.
The exposition of verbal intuitive served by many examples carried out from across the spectrum of behavioral sciences related to psychology. The study of statistical power and its effect on power of studies by Seldlmeier and Gigerenzer. The two endeavored to shared their thoughts about a Journal of Abnormal Psychology, in 1989 reviewing on the power studies. They studied and viewed an article published in 1984 volume of the Journal of Abnormal Psychology.
A null hypothesis method of research was carried on a large population base and its effects measured. The effects on the population were also accounted for. The choice of the method of study was found to be no significant and researchers ignored power analysis. Surprisingly there is no corresponding evidence of controversy between methodologists on the significance of power analysis. Sedlimier and Gigerenzer attribute the passiveness of researchers towards power analysis to the accident of historical Fisherian theory. Its hybridization with the contradictory Neyman-Pearson theory and apparent completeness of Fisherian null hypothesis.
Statistical power analysis exploits the relationship among sample size, significance criterion, population effect size, and statistical power. For all statistical model, the relationship between these variables must be clearly and considerably defined. When ES is not equal to zero H0 is false. This causes a type two error on the data illustration. Sample size N will significantly increase with increase in power desired. Specifying the ES is the most difficult part of power analysis. The power analysis spectrum requires a high degree of accuracy and a significant illustration. The relationship between the variables must be clearly illustrated for considerable counter checking by future studies.
The Neyman-pearson method of statistical interference shows relationships between hypothesis. It involves alternate hypothesis H0 counterpoised against H1. There are also other tests considerably used in psychological research. The test for the difference between two independent means. It deals with corresponding formulae. The test for product moment correlation coefficient, the test of the difference between two independents accomplished as a normal curve test through fisher Z transformation of r. The binomial distribution for large samples that uses nonparametric sign tests for differences between paired observations. The normal curve test for the difference in two independent proportions is also accomplished through arcsine transformation. Lastly the chi square test for goodness of fit that analyzes data in two contingency tables. These test help in the comprehensive, elaborate study and research of psychological phenomena