The article argues against the practice of testing if data follows a normal distribution. The researchers reject the idea that data must be normal, stating it's unnecessary and a waste of time. They suggest that not conforming to normality doesn't affect the validity of results. By avoiding normality tests, researchers can save time and energy.