Discussion topic - Assignment Example

Discussion topic affiliation In any statistical analysis, the alpha level must be determined to guide in data interpretation. In essence, the alpha level is the chance of not accepting the null hypothesis even if the null hypothesis is true. In summary, it can be stated that alpha level is the probability of concluding a wrong decision. If a researcher is worried about the possibility of rejecting the null hypothesis when it is definitely true, then he should use a lower alpha level to minimise the chance that could lead to rejection of null hypothesis incorrectly.

However, reducing alpha level could lead to committing a type II error, the error of not rejecting a null hypothesis when it is false (Gerhan, 2001).There are experiments that require low alpha levels like in cases that may lead to death, maim or serious defects occurring. In such experiments, the researcher wants to avoid a situation whereby the null hypothesis is true and rejected. Therefore, if 0.05 chance of being wrong cannot be tolerated then a much lower level of significance like 0.001 can be used.

For example performing a clinical study of a new drug on people, it will mean that a researcher does not want to take chances if adverse drug reactions are noted on an individual. A Higher level of alpha like 0.10 is rare in practice because it increases the chance of making type I error. Higher levels may be used when doing an experiment the researcher is at ease of accepting the null hypothesis (Shi, Levinson, & Whittemore, 2008). For example, when a researcher insists that there is no significant difference between intelligence among male and female nursing student in a nursing college.

ReferenceGerhan, D. (2001). Statistical Significance: How It Signifies in Statistics Reference. Reference & User Services Quarterly, 40, 361–374.Shi, J., Levinson, D. F., & Whittemore, A. S. (2008). Significance levels for studies with correlated test statistics. Biostatistics, 9, 458–466.