Document Type

Dissertation

Publication Date

8-1992

Abstract

A paper and pencil test was constructed to investigate the existence of phases as identified by the Service Learning Model (Delve, Mintz, & Stewart, 1990). Subjects with no structured exposure to service-learning activities residing in the residence halls at the University of Northern Colorado (URC) (n =70) and subjects involved in service-learning programs coordinated by the Office of Community Service at Colorado State University (CSU) (n=65) participated in the study during the spring of 1992.

Research objectives provided guidelines for developing the test format and Writing clear and understandable items that reflected the affective, behavioral and cognitive ways in which students experience and understand what it is to be a responsible citizen. No significant relationship was observed between the phases of the test instrument and parallel learning modes as measured by the Learning-Styles Inventory. It was suggested that learning styles transcend the different ways to experience and understand what it is to be a responsible citizen, and therefore, do not provide an appropriate measure of concurrent validity. The internal reliability tor each phase of the test instrument was lower than desired, however, the coefficients of internal consistency showed promise ranging from r=.5266 for Phase 1 to r=.7582 for Phase 3.

Post-hoc tests comparing mean phase scores between colleges and between categories of self-reported duration of involvement in helping activities were conducted to supplement a statement of validity for the test instrument. Differences tor mean phase scores between subjects at UNC and CSU were significant and in the predicted direction. Similar results were observed when mean phase scores tor subjects with eight or fewer contacts over the prior academic year were compared to subjects with nine or more contacts over the same period. However, it was unclear it these differences were influenced by involvement in service-learning activities or personal variables such as age and class status.

More research needs to be conducted before the Service-Learning Model can be empirically confirmed or rejected. Recommendations identity several suggestions for further investigating the validity and reliability of the test instrument.

Comments

Copyright 1992 by Christopher Anthony Payne.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS