Researchers are always on the lookout for data that are ‘Statistically Significant’. But what does ‘Statistically Significant’ really mean in the scientific context? This has baffled both the scientists and the general public. Here statistically significant does not hold the same weight age as words such as significant or insignificant. ‘Statistically Significant’ in the scientific context means that there is a strong case for a relationship or correlation between two or more variables and that they are not related to each other by randomness or chance.
In statistical testing, a p-value is determined. This p-value indicates whether the results observed can be observed by pure chance or randomness or whether there is actually a relationship between the two. As a rule of thumb p-value of 5% or lower is considered to be statistically significant. In other words, there is a correlation between the variables. p-values are extremely important when dealing with quantitative methods. Whether a result is ‘Statistically Significant’ or has a p-value of 5% or lower, will be determined by the sample size of the experiment. With the increase in sample size, the likelihood of obtaining statistically significant results increases. In a large sample size it is impossible for a relationship to exist between variables simple by chance, rather it is more likely that a concrete relationship exists between the variables under study. Small sample sizes (say fewer than 50 users) are ineffective in providing statistically significant results. When the sample size is large it is possible to represent the phenomenon on the population sample being studied. A data set is considered statistically significant if the probability of the phenomenon being random is less than 1/20 (one out of every twenty). This explains why p-value is set at 5%. In other words, if the p-value is less than 5% then the result is statistically significant.
For example, a pharmaceutical company has a discovered a drug to cure type II diabetes. It is found that the drug reduced type 1 diabetes and had a p-value of less than 5%. This automatically means that the lowering of type 1 diabetes is due to the drug itself and was not due to random chance.
Subscribe today and get future blog posts your email.