ALERT! Warning: your browser isn't supported. Please install a modern one, like Firefox, Opera, Safari, Chrome or the latest Internet Explorer. Thank you!
Log inLog in to ACORN Wiki. | RegisterSign on to ACORN Wiki.

Items or Questionnaires?

See also Outcomes Measurement 2.0

This topic focuses on the relative merits of two alternative methods of choosing measures to evaluate the outcome of treatment. The traditional, and most widely used method, is to select questionnaires with known psychometric properties and presumptively demonstrated reliability and validity for the intended use.

The advantages of this method are many and obvious. Primarily, it assures that the measures meet a minimum standard for scientific credibility, and save the researcher the difficulty of developing new questionnaires.

However, there are also disadvantages to the restriction of using questionnaires with known psychometric properties. Almost invariably, such questionnaires are copyrighted. In many instances there are licensing fees for use of the questionnaires; but even in the absence of licensing fees, the copyright places limitations on creating alternative versions of the questionnaires by selectively adding or dropping items.

All questionnaires are a collection of items from a potentially larger pool of similar items assessing a similar domain (symptoms of depression for example). Once a questionnaire has been copyrighted, there is a naturally tendency to restrict further development or modification of the questionnaire. The user loses the flexibility to select only the best items from the questionnaire for a particular purpose, to add items of special interest, or in any other ways to alter the number and type of items presented to the patient. From a pure measurement point of view, this puts the end user at a significant disadvantage when trying to extract the greatest amount of useful information for the time and cost associated with collecting the data.

The alternative measurement strategy is to begin with an inventory of items rather than complete questionnaires. The critical requirement for this strategy is that the organization has the flexibility to construct questionnaires for its use by drawing from an inventory of individual items. This strategy is preferable if the organization has the capacity to capture and analyze data on the items on an ongoing basis from a sufficiently large sample of clients or patients. As data accumulates, the organization has the ability to continuously evaluate the psychometric properties of any questionnaires at the item level, and continuously refine the items (and questionnaires) so that the primary domains of interest are measured as efficiently as possible, using no more items than is necessary to the measurement task.

This strategy requires that the organization possesses or have access to individuals with the necessary statistical expertise to perform ongoing item analyzes using methods based on both classical test theory and item response theory.

Primary topics:

You may also view all topics in alphabetical order in WebTopicList.

-- JebBrown - 05 Jan 2007
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding ACORN Wiki? Send feedback
Syndicate this site RSS