NotesFAQContact Us
Search Tips
ERIC Number: ED361344
Record Type: Non-Journal
Publication Date: 1993-Apr
Pages: 18
Abstractor: N/A
Reference Count: N/A
Using Linear Regression To Determine the Number of Factors To Retain in Factor Analysis and the Number of Issues To Retain in Delphi Studies and Other Surveys.
Jurs, Stephen; And Others
The scree test and its linear regression technique are reviewed, and results of its use in factor analysis and Delphi data sets are described. The scree test was originally a visual approach for making judgments about eigenvalues, which considered the relationships of the eigenvalues to one another as well as their actual values. The graph that is plotted resembles a mountainside where a base pile of rubble, or scree, is formed. The analysis determines which eigenvalues are salient (mountainside) and which are rubble (scree). A multiple linear regression (MLR) approach has been proposed that would include more data points than the usual scree test and could yield better results. The MLR test provides the same decision as does the visual scree test, but can be easily programed, using an approach in which the ordered eigenvalues are thought of as points in a scatterplot. Examples are presented of the use of the scree test and the MLR approach with Delphi technique data to help decide how many items or issues to retain in a Delphi study or survey. The MLR approach appears to be an effective analytical procedure for the scree test. It usually produces the same number or fewer factors than the visual scree test in factor analysis, but yields more items than the visual scree in factor analysis. Three tables and six figures illustrate the discussion. (Contains 15 references.) (SLD)
Publication Type: Reports - Evaluative; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A