The logic of null hypothesis testing can be applied to real research scenarios. For example, if researchers find a very small difference in talkativeness between men and women ($d = 0.06$), they would determine this result is fairly likely even if there were no true difference in the population, leading them to retain the null hypothesis. Conversely, if researchers find a strong correlation ($r = +.60$) between daily hassles and stress symptoms, they would determine this sample result is highly unlikely if there were no true relationship, leading them to reject the null hypothesis in favor of the alternative.