Institutional Affiliation
Statistical Significance and variables
In a study, the differences between the results of a study may be due to chance or can be attributed to a specific cause. Statistical significance implies that it is more likely that the association between variables is due to something than it is by coincidence. To be exact, statistical significance is used to either refute or agree with the null hypothesis (Benjamin et al., 2018). A null hypothesis assumes that statistical significance does not exist in a given set of results. To determine whether the observations in a particular study is statistically significant, a statistical hypothesis test is used. It incorporates a P-value which, if it falls below 5 percent, then statistical significance is said to exist.
The difference between statistical significance and the clinical importance of the association between variables may affect decision making in practice. Statistical significance differs with clinical significance in the sense that it suggests that the differences in the variables had a cause (Benjamin et al., 2018). On the other hand, the true importance of the association of the results considers why a treatment effect was practically important. The differences might force an individual to think that a particular intervention will be more effective than the other to manage a problem. Novo Nordisk a pharmaceutical company conducted a study to determine whether the new insulin was working. Based on the results, the study presented a statistically significant decrease in type 1 diabetes. The test has been performed on random patients with diabetes for 26 weeks whereby they were put on therapy. A P-value of less than 5 percent indicated that the drop in diabetes type one was based on a specific cause.
As a social worker, the results of the study would make one believe that there must have been something that caused the changes. This is because the patients were selected at random. If there was no defined cause, the medication would not have worked for most of the individuals.
Benjamin, D. J., Berger, J. O., Johannesson, M., Nosek, B. A., Wagenmakers, E. J., Berk, R., & Cesarini, D. (2018). Redefine statistical significance. Nature Human Behaviour2(1), 6.