Often researchers get excited about discovering statistically significant findings without full comprehension of its meaning. It is a phrase that is loaded with meaning, yet often it is hard to explain and harder to understand. Traditionally we set up a null hypothesis (H0) and an alternative hypothesis (H1) and by using a statistical significance test we try to reject the null hypothesis. Using effective visualisation to help to explain the concept of statistical significance, the talk illustrates how easy it can be for a difference between the two samples to occur purely by chance. Without focusing on hypothesis testing and the calculation of some other statistic like a t-ratio, by using two-sample permutation bootstrapping consideration can be given to the actual statistic of interest, ie. the difference between the two samples and its null distribution. Hence, by simply visualising the null distribution and the p-value we can make the procedure of hypothesis testing, which often can be perceived as complex, very natural.
| 
