Validity of Models of Excellence

A major question for heads of organizations, human resource professionals, and corporate legal departments is:

What is the validity of the Model and
of the tool (iWAM) used to create it?

The Model of Excellence is created on the basis of data from the Inventory for Work Attitude and Motivation. More important, the data are gathered on individuals in your organization who are in the role that is being modeled. A Model is not based on a large national or international sample of individuals in different organizations that have different role requirements than yours.

What is Validity?

Basically, measures of validity provide an indication that a tool or process measures what it says it will measure. There are several kinds of validity that range from how something looks to complex statistical documentation. For a summary of the kinds of reliability and validity measures that exist, see the paper based on the work of Dr. William Trochim (click here to open).

Modeling Methodology

There are three factors in the creation of a Model of Excellence that insure the power and validity:

  1. Models are built for a specified role in a given organization. There is no guarantee of transportability. In fact, a model may not apply to similar roles in the same organization. We have evidence that two sales groups in the same organization (different products and different customers) had significantly different MAPs for high performers.
  2. The Model of Excellence for high performers in an organization is based on your people and your performance ratings, not some national norm that may or may not apply. The resulting Model is yours based on what is required to be successful in the role in your organization.
  3. The analysis leading to a Model is both quantitative and qualitative. The preliminary Model is tested against the iWAM results on a representative group of people in the role and a correlation coefficient (r-value) is computed. The magnitude of the positive (+) correlation indicates if the model scores and performance ratings are significantly related. When the correlation is significant, we compute an r-squared (r2) value that provides an estimate of the amount of variance in performance scores accounted for by MAPs.

The methodology and documentation created in the course of building a Model provides the organization with not only a powerful tool for hiring, managing, and coaching, but also a highly defensible tool as part of selection and succession planning. 

A Note on Reliability

In psychological testing, reliability is a meausre of the extent to which a tool will measure a construct consistently over time.

For some tests–those which measure stable traits such as intelligence, for example–reliability is an important factor.

When, however, one is measuring a construct that can vary the notion of getting the same measure at two points in time is based on a presumption that nothing changed. Well, it is possible, through various means to alter the motivativational and attitudinal patterns of some individuals. That is why the iWAM is such a powerful tool for coaching interventions.

In a case where you have intentionally tried to alter MAPs or circumstances caused a shift in MAPs (changes in context can have that effect), you will not get the same level of consistency in retesting as with other types of instruments.

So what?

Well, those who claim that a test such as the iWAM must meet the classic standards of reliability to be judged as eligible to be used for modeling are simply guilty of comparing apples to oranges. This instrument and its application requires a different, but equally robust, set of standards.

Latest News: