The purpose of this study was to construct and gather data on an "Observation of Group Self-Concept Schedule" and a parallel "How I Am A t School Self-Report," in an attempt to produce valid and reliable scales of measurement of group self-concept. The significance of the study was founded in the lack of instruments assessing group self-concept.Group self-concept was defined as: A social structure exhibited by interactions which indicate values, attitudes and beliefs that determine the self-perception of a group, wherein individuals mirror the composite group perceptions demonstrated through group behavior. It is an attribute that is capable of change.The related literature supported the term group self-concept through references to the organization and structure of the self, as well as in the recommendations for analyzing the data in terms of group assessments. The conclusions of several studies on self-concept indicated a need for parallel instruments to gather data on the complex structure of the self.The construction of both the observation and self-report instruments was based on criteria that promoted descriptive rather than judgmental items. The process of instrument development included close examination of students in an operational setting and review of literature relevant to self-concept, instrument design in the affective domain and analytical procedures for reliability and validity estimates. The instruments were exposed to field-testing and constructive criticism for in-depth clarification and refinement. Observers and instrument administrators were trained until the researcher was comfortable with their skills in the data collection procedure.The sample included fourth, fifth and sixth grade pupils who were involved in the Westinghouse *PLAN which included computer assisted instruction, open-education and team-teaching.The data collection process included five pairs of trained observers rating each pupil group in each grade level, wherein pairs of observers rated a group simultaneously for twenty minutes. Twenty-five per cent of the ratings of pairs of raters obtained a correlation coefficient of .80 or greater and seventy-four per cent of the ratings of the pairs of raters obtained a correlation coefficient of .50 or greater.Statistical analysis of the data produced a reliability coefficient of .84 (L.05) for test-retest procedures using a three week interim. A stability coefficient of .72 (<.05) was obtained overa ten month period.A panel of Educational Psychology Faculty assessed the content validity of both instruments and agreed that both contained a pool of items which related to individual indicators of public self-concept.The analysis of the data attempting to interpret the construct validity was an intellectual and analytical number of students to factor analyze the data. Oblique and orthogonal simple structure rotations produced factors that could be interpreted.A correlation coefficient of .55 was obtained between the observation schedule and the self-report.The problems of negligible variation and weak items as revealed through statistical analysis were pursued. The revised instruments contain expanded response systems and syntactically refined items in an attempt to resolve the aforementioned problems.It is recommended that if similar studies are carried out that there be a balance maintained between intellectual and statistical approaches in constructing the instruments and analyzing the data.The primary stage of this study has produced guidelines and has supplied implications for further research.
| Identifer | oai:union.ndltd.org:BSU/oai:cardinalscholar.bsu.edu:handle/175996 |
| Date | January 1975 |
| Creators | Eckard, Pamela Jane |
| Contributors | McElhinney, James H. and Lane, Conrad C. |
| Source Sets | Ball State University |
| Detected Language | English |
| Format | vii, 94, [25] leaves ; 28 cm. |
| Source | Virtual Press |
Page generated in 0.0016 seconds