All Outputs (3)

An outlier in an independent samples design (2018)
Presentation / Conference
Derrick, B. (2018, September). An outlier in an independent samples design. Paper presented at Royal Statistical Society Conference, Cardiff City Hall, Cardiff, Wales

There is a flaw with some of the most commonly performed statistical tests. A paradox of the one sample t-test is the contrariwise decrease in the p-value as the value of an outlier increases in the direction of the overall effect. Demonstration of t... Read More about An outlier in an independent samples design.

To preliminary test or not to preliminary test, that is the question (2018)
Presentation / Conference
Derrick, B. (2018, July). To preliminary test or not to preliminary test, that is the question

Despite the consequences of violations to the assumptions of statistical tests, there is often neglect in published works to report on the assumptions of the tests performed. Preliminary testing the assumptions prior to preforming the test is done ro... Read More about To preliminary test or not to preliminary test, that is the question.

Statistics: New t-tests for the comparison of two partially overlapping samples (2017)
Presentation / Conference
Derrick, B. (2017, June). Statistics: New t-tests for the comparison of two partially overlapping samples. Poster presented at Faculty of Environment and Technology Degree Show, Frenchay Campus, UWE

This poster presented for the UWE degree show (June 2017), summarises Type I error rates and Power of standard t-tests and a new generalised form of the t-test, for comparing samples which include both paired and independent observations. Material... Read More about Statistics: New t-tests for the comparison of two partially overlapping samples.