Sample Size Estimation

For correlation and regression

Download 0.86 Mb.
Size0.86 Mb.
1   ...   8   9   10   11   12   13   14   15   16
For correlation and regression:
Rarely is a study aiming only to establish a single pairwise correlation between two variables. So the formula for sample size to achieve a significant correlation is not very useful. As well, a test of whether correlation is significant is really only testing whether the correlation is zero or not; it doesn’t’ say anything about the “strength” of the correlation. Correlation is usually just the first step in model-building, especially regression models.
For multiple regression:
It is impossible to estimate regression coefficients before doing the research and data collection study so power studies aren’t really relevant here. Instead, just ensure that the number of data points (i.e. observations or cases) is considerably more than – 5 to 10 times – the number of variables. (A reference for this rule of thumb is: Kleinbaum, Kupper and Muller (1988).)

Download 0.86 Mb.

Share with your friends:
1   ...   8   9   10   11   12   13   14   15   16

The database is protected by copyright © 2023
send message

    Main page