Determining a Statistically Valid Sample Size
Calculating the sample size for a chart review follows selected steps adapted from statistical techniques used for descriptive studies with dichotomous variables.
Descriptive studies:
make statistical inferences about the total population such as describing the average and percentage.
Dichotomous variables:
describe the measure in two distinct outcomes – either with the characteristic or without.
Nomogram:
is a graphic representation of numerical relations, or in the case of descriptive studies it is a table showing the sample size most predictive of the value being assessed in the total population.
There are 4 steps to statistically determine a sample size for a chart audit. The process uses a nomogram, or table to identify the desired number. The nomogram looks like this:
Sample size for a descriptive study of a dichotomous variable
95% confidence interval
|
Width of confidence interval |
0.10 |
0.15 |
0.20 |
0.25 |
0.30 |
Expected proportion (P) |
|
|
|
|
|
0.10 |
139 |
|
|
|
|
0.15 |
196 |
88 |
|
|
|
0.20 |
246 |
110 |
62 |
|
|
0.25 |
289 |
128 |
73 |
47 |
|
0.30 |
323 |
144 |
81 |
52 |
36 |
0.40 |
369 |
164 |
93 |
60 |
41 |
0.50 |
384 |
171 |
96 |
62 |
43 |
The 4 steps are:
1. Estimate the expected proportion within the population that will have the measure of interest. (yellow cells)
2. Specify the width of the confidence interval you wish to use. (pink cells)
3. Set the confidence level. (green cells)
4. Use the table to estimate sample size. (blue cells)
Examples
Click on one of the images to view the example
|