14.03.2014 Views

Basic Analysis and Graphing - SAS

Basic Analysis and Graphing - SAS

Basic Analysis and Graphing - SAS

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

Chapter 5 Performing Oneway <strong>Analysis</strong> 145<br />

<strong>Analysis</strong> of Means Methods<br />

Table 5.9 Descriptions of Methods for Comparing Means<br />

ANOM<br />

ANOM with Transformed Ranks<br />

Compares group means to the overall mean. This method<br />

assumes that your data is approximately normally distributed.<br />

See “Example of an <strong>Analysis</strong> of Means Chart” on page 162.<br />

This is the nonparametric version of the ANOM analysis. Use<br />

this method if your data is clearly non-normal <strong>and</strong> cannot be<br />

transformed to normality. Compares each group mean<br />

transformed rank to the overall mean transformed rank.<br />

Compare St<strong>and</strong>ard Deviations (or Variances)<br />

Use the ANOM for Variances <strong>and</strong> ANOM for Variances with Levene (ADM) options to compare group<br />

st<strong>and</strong>ard deviations to the root mean square error. This is a type of variance heterogeneity test.<br />

Table 5.10 Descriptions of Methods for Comparing St<strong>and</strong>ard Deviations (or Variances)<br />

ANOM for Variances<br />

ANOM for Variances with Levene<br />

(ADM)<br />

Compares group st<strong>and</strong>ard deviations to the root mean square<br />

error. This method assumes that your data is approximately<br />

normally distributed. To use this method, each group must have<br />

at least four observations. See “Example of an <strong>Analysis</strong> of Means<br />

for Variances Chart” on page 163.<br />

This is the nonparametric version of the ANOM for Variances<br />

analysis. Use this method if you suspect your data is<br />

non-normal <strong>and</strong> cannot be transformed to normality. Compares<br />

the group means of the absolute deviation from the median<br />

(ADM) to the overall mean ADM.<br />

<strong>Analysis</strong> of Means Charts<br />

Each <strong>Analysis</strong> of Means Methods option adds a chart to the report window, that shows the following:<br />

• a center line indicating the overall mean or root mean square error (or MSE when in variance scale)<br />

• upper decision limits (UDL)<br />

• lower decision limits (LDL)<br />

If a group mean falls outside of the decision limits, then that mean is significantly different from the overall<br />

mean. If a group st<strong>and</strong>ard deviation falls outside of the decision limits, then that st<strong>and</strong>ard deviation is<br />

significantly different from the root mean square error.<br />

<strong>Analysis</strong> of Means Options<br />

Each <strong>Analysis</strong> of Means Methods option adds an <strong>Analysis</strong> of Means red triangle menu to the report window.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!