Statistical significance is an approach used by statisticians to link an occurrence with a given cause. This approach seeks to come up with a cause-effect relationship from a given set of data analyzed. Through this approach, variables are ranked in the manner in which they affect the outcome. This approach is used to validate the data set collected and analyzed. Through this approach, the probability of the analyzed and reported data being random is highly reduced. This approach helps in determining whether the study’s null hypothesis is to be rejected or accepted (Antonakis & John, 2017) . This approach help ranks variables in order of significance to the study and therefore the researcher can be able to ascertain the meaning of the relationship between variables and results. The significance level is a parameter used to determine the variables which are more relevant to a study and those variables that are of less impact for the topic of consideration. Significance level helps the researcher assign meaning to his/her data after analysis.
Type 1 error arises in a situation whereby the researcher rejects a true null hypothesis instead of accepting it. Type 2 error arises in instances where the researcher, on the other hand, accepts a false null hypothesis and therefore coming up with the wrong conclusion. Statisticians argue that either type of error should be avoided as much as possible since the error may have dire consequences to the issue under consideration (Clarke, et al., 2011) . The significance level of the variable under consideration determines the type of error the researcher is at risk of making. To minimize the occurrence of either type of error the researcher ought to improve the sample size. At 0.05 significant level, the researcher is at a risk of making type 1 error as opposed to when the significance level is at 0.01.
Delegate your assignment to our experts and they will do the rest.
On the other hand, the P-value of 0.05 means that there is probability of 5% of the outcome being statistically significant. Alternatively, it means that 1 in 20 tests would give us a significant value by chance alone. If we set the nominal cutoff at 0.10, we are increasing the probability of finding a false positive result to 10% (1 in 10).
References
Antonakis, & John. (2017). on doing better science: from the thrill of discovery to policy implications. The Leadership Quarterly, 28 (1), 5-21. doi:10.1016/j.leaqua.2017.01.006
Clarke, GM, Anderson, Pettersson, Cardon, LR, . . . Zondervan. (2011). The basic statistical analysis in genetic case-control studies. Nature protocols, 6 (2), 121-133. doi:10.1038/nprot.2010.182
Derrick, Toher, & White. (2016). Why Welchs test is type 1 error robust. The Quantitive Methods for psychology, 12 (1), 30-38. doi:10.20982/tqmp.12.1.p030
