From the Editor… Ruth Huntley Bahr You have probably heard that you can prove anything with statistics. Weil, maybe not everything, but if you run enough subjects in any experiment, eventually you will get a statistically significant difference between groups. However, is this difference really an important one? For instance, I could ... Editorial
Editorial  |   April 01, 2002
From the Editor…
 
Author Notes
Article Information
Editorial
Editorial   |   April 01, 2002
From the Editor…
Language, Speech, and Hearing Services in Schools, April 2002, Vol. 33, 83. doi:10.1044/0161-1461.3302.83
 
Language, Speech, and Hearing Services in Schools, April 2002, Vol. 33, 83. doi:10.1044/0161-1461.3302.83
Ruth Huntley Bahr
You have probably heard that you can prove anything with statistics. Weil, maybe not everything, but if you run enough subjects in any experiment, eventually you will get a statistically significant difference between groups. However, is this difference really an important one? For instance, I could run a study designed to determine if there is a statistically significant difference in the birth weights of children based on their sex. Given a large enough sample, my results could show that, on average, baby boys weigh more than baby girls. Would you then go out and buy a gift for a new baby knowing only the birth weight? Of course not. There are male infants who weigh less than the average female infant and vice versa. So, in this case, one would say that the difference is statistically significant, but it is not practically significant, In other words, the data do not change the way you do things. However, that is exactly why we need research—to help us support or change the way we do things.
First Page Preview
First page PDF preview
First page PDF preview ×
View Large
Order a Subscription
Pay Per View
Entire Language, Speech, and Hearing Services in Schools content & archive
24-hour access
This Article
24-hour access