Files in this item
|(no description provided)|
|Title:||Two Interaction Systems for Analyzing Supervisor/teacher Behavior in the Conference Phase of Clinical Supervision|
|Author(s):||Dyasi, Rebecca E.|
|Department / Program:||Education|
|Degree Granting Institution:||University of Illinois at Urbana-Champaign|
|Abstract:||Purpose. This dissertation describes the development and utility of two interaction instruments for self- and peer assessment of a clinical supervisor/teacher behavior during a supervisory session. The self assessment instrument developed was a category system, called Low Inference Supervisor/Teacher Observation System (LIOS) to provide diagnostic data focussing attention on the presence or absence of particular supervisory behaviors or techniques that characterize supervision as being clinical. The peer assessment instrument was a scale system called High Inference Observation Criteria (HIOC). HIOC determines the degree to which the observed supervisory activities were clinical.
Procedure and Analysis. Key items of clinical supervision that were practical, and would enhance the quality and quantity of teacher participation, and which were also sensitive to supervisors' professional growth were extracted from literature and video recordings of clinical supervisory sessions. These key items were assembled into two pools appropriate for each type of instrument in clinical supervision as universal, practical, sensitive to growth, and suitable for quantification. The items of the two pools formed the categories for LIOS and the sub-scales for HIOC. Both instruments were used by three groups of observers--experts, novices (no knowledge of clinical supervision), and supervisors-in-training to analyze nineteen supervisory videotapes. The data generated by the analysis were used to establish the instruments reliability, usability (ease of use), validity, and the efficacy of LIOS by examining its sensitivity to changes in the individual supervisory style.
Scott's method for calculating inter-observer reliability was used to analyze data for LIOS's reliability and usability. Changes in the percentages of LIOS's categories and the behavior pattern the categories represent were used to illustrate sensitivity of the instrument. Analysis of variance was used to analyze data for HIOC's reliability and usability. Besides content validity, the study also established criterion-related validity by comparing data generated when both instruments were used to analyze supervisory videotapes with results of an independent classification of the same supervisory videotapes.
Selected Findings and Conclusions. LIOS's inter-observer reliability ranged from .58 to .93 across observers. HIOC inter-rater reliability was .98 for the expert raters and .95 for the novice raters. The novice observers used both instruments as effectively as the experts with fifteen minutes training and explanation of the instruments' categories and subscales.
Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 1986.
|Date Available in IDEALS:||2014-12-15|