Overview (POWER PARTIALCORR command)
POWER PARTIALCORR invokes the power analysis for the one-sample partial
correlation test.
Options
- Test Assumptions
- Specifies the keywords that control the test assumptions. The keywords specify the sample size, power for sample size estimation, null hypothesis value of the partial correlation parameter, alternative hypothesis value of the partial correlation parameter, test direction, variables assumed to be partialled-out, and the significance level of the type I error rate. For more information, see PARAMETERS Subcommand (POWER PARTIALCORR command).
- Two-Dimensional Plot
- Controls the two-dimensional power by number of partialled-out variables chart, the two-dimensional power by null value chart, the two-dimensional power by alternative value chart, and the two-dimensional power by difference between hypothesized values chart.
- Three-Dimensional Plot
- Controls the three-dimensional power by sample size (x-axis) and difference between hypothesized values (y-axis) chart, the three-dimensional power by sample size (y-axis) and difference between hypothesized values (x-axis) chart, the three-dimensional power by null (x-axis) and alternative value (y-axis) chart, and the three-dimensional power by null (y-axis) and alternative value (x-axis) chart. Also controls the vertical rotation degrees (clockwise from the left) and the horizontal rotation degrees (clockwise from the front) for the three-dimensional chart.
- Precision
- Estimates the sample size based on confidence intervals by specifying the values of the confidence interval half-widths.
Basic specification
The basic specification is the POWER PARTIALCORR command and a PARAMETERS subcommand that specifies valid values for the keywords ALTERNATIVE, PARTIALOUT, and either POWER or N.
Subcommand order
The subcommands can be named in any order.