If we take the study at face value, then even if the result is technically true (and robust, with a small p value), the effect size is small. The z-scored well-

If we take the study at face value, then even if the result is technically true (and robust, with a small p value), the effect size is small. The z-scored well-being differs barely at all between the median income and the lowest and highest ranges -- it's a difference of .1.