Presentation of Results

 
Data Tables
When a question calls for a response in the form of a number, responses are reported as the median value and the interquartile range (IQR). The IQR is the range bounded at the low end by the 25th-percentile value, and at the high end by the 75th-percentile value. As shown in the sample question on page 3, the median year 2000 forecast for North American produced passenger car engines incorporating polymer-based fuel rails was 20 percent and the IQR 10-25 percent. This means that one-quarter of the respondents answered 10 percent or less, another one-quarter chose 25 percent or more, and the middle half of all responses ranged between 11 percent and 24 percent. The interquartile range provides an indication of the degree of uncertainty among respondents.

Other questions require responses in the form of a rating, for which mean ratings are computed and presented.

Discussion
Narrative discussions are presented to highlight and explain a particular set of data.

Selected Edited Comments
Selected, edited comments from the Delphi panelists are shown following each data table in order to provide some insight into the deliberative process by which panelists arrived at their forecast.

Such replies may provide important information which is not evident in the numerical data; however, readers should be careful not to overemphasize a particular comment.

Manufacturer/Supplier Comparison
Delphi 9 panelists include respondents from the North American automotive manufacturers, major suppliers of components, parts, and materials, and consultants and academics. A concerted effort is made to obtain an adequate sample of both manufacturer and supplier panelists, to permit comparisons to be made. Where it is considered relevant to a better understanding of or perspective on the forecast, our analyses include such a comparison in an attempt to illustrate where significant agreements or differences exist.

Comparison of Panels
The three groups of Delphi panelists were asked questions that specifically focus on their respective areas of expertise. However, a few questions are considered common to two or more panels. Where differences do exist between the panels, serious consideration should be given to whether the difference reflects the composition and proprietary interest of that particular panel or whether there exists a substantial degree of uncertainty regarding the issue in question. We try to highlight both the differences and similarities.

Trend from Previous Delphi Surveys
A single Delphi survey is a snapshot which collects and presents the opinions and attitudes of a group of experts at a particular point in time. Some questions, in various forms, were asked in previous Delphi surveys, and thus provide trend data. Such trend data can reveal the stability or volatility of a particular market, material or technology issue.

Strategic Considerations
Based on the replies to a particular question, other relevant Delphi X responses, other research and studies, and OSAT's extensive interaction with the automotive industry, this report makes inferences and interpretations as to the core issues in questions and their potential impact on the industry. By no means are they exhaustive statements of critical issues. Rather, they are points that the reader might consider catalysts in formulating their own judgments regarding the implications of the forecast.

Click here for a Sample Question and the Presentation of Results
 

Next: "Value Added to Planning and Research"
Back to Delphi Table of Contents



 

| Home | Staff | Research | Affiliates | Conferences | Newsletters | Other Sites


 

Copyright © OSAT, 2000