NotesFAQContact Us
Search Tips
ERIC Number: ED314470
Record Type: Non-Journal
Publication Date: 1989-Oct
Pages: 10
Abstractor: N/A
Reference Count: N/A
Judging the Goodness of Fit of a Multiple Regression Equation for Predicting Program Results across Multiple Sites.
Wright, Joan
An educational program on family resource management was evaluated to judge the goodness of fit of multiple regression equations for predicting program results across multiple sites. The applicability and effectiveness of the 11-lesson MONEY SENSE program was evaluated in six California locations for 190 clients. Indicators of effectiveness were: (1) change in frequency of financial problems; (2) change in self-perception of financial status; (3) change in average weekly food cost per person; (4) food consumption index score; (5) change in percentage of food shopping skills used; and (6) change in percentage of money management skills used. The 16 situational variables posited as influential were concurrently entered into two regression equations (restricted and unrestricted) for each of the outcome measures, and the validity of the two models was assessed. The analysis illustrated the use of a technique for determining whether the unrestricted equation (one specific to a subset) or the restricted equation (a regression equation for all members of a set) represents a better fit. In the example, the fit of the unrestricted equation was superior for three outcome measures, supporting the intuitive understanding that application of a curriculum in multiple locations does not mean that the program operates identically in all locations. Six tables provide study data. (SLD)
Publication Type: Reports - Evaluative; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A