Table of Contents#### Download Safari Books Online apps: Apple iOS | Android | BlackBerry

### Designing, Conducting, and Analyzing a Multi-Vari Study: A Roadmap

#### A. Designing the Multi-Vari Study

Entire Site

Free Trial

Safari Books Online is a digital library providing on-demand subscription access to thousands of learning resources.

A number of steps must be undertaken in designing, running, interpreting, and analyzing a multi-vari study. These are summarized as follows:

Identify the Green Y (the problem to be solved). If the Green Y is an attribute, try to convert it into a variable using a Likert scale.

Make sure that the accuracy of the measuring instrument is at least five times the accuracy of the product (i.e., its allowed tolerances).

Determine the number of families of probable variation.

Draw a family tree.

Estimate the number of time-to-time samples required.

Determine the number of unit-to-unit samples to be drawn consecutively from the process (generally three to five consecutive units).

Determine the number of samples for each subfamily of within-unit families; e.g., the number of locations, machines, cavities, etc.

Multiply the quantities in steps 3, 4, and 5 to determine the total number of units to be studied.

Design a table to facilitate gathering the multi-vari data.