應用概化理論估計運動技能測驗多層面變異來源的信度－壘球打擊成績表現測驗為例

Applications of Generalizability Theory to Estimate the Reliability of Multi-facet Source of Variation in Psychomo-tor Test: the Performance of Hitting in Softball Test

姚漢禱(Han-Dau Yau)

《體育學報》 23期 (1997/09) Pp. 109-120

https://doi.org/10.6222/pej.0023.199709.3110

摘要

本研究的目的是：應用概化理論估計多層面變異來源的運動技能測驗。五十九位受試者實施壘球打擊成績表現測驗。採受試者、試做次數和評分者等三個層面的重複設計，以電腦程式GENOVA進行估計。概化研究的結果是：
一、測驗的眞實變異佔總變異成份的22.15%。
二、「受試者和試做」交互作用的變異成份佔總變異的70.55%，爲最主要的變異來源。
三、其餘的變異來源所佔的比率很小。決斷研究的結果發現：增加試做次數可以有效的提高概化係數，而增加評分者人數對測驗的概化係數增加不多。
根據研究的結果提出兩點結論：1.多層面變異來源的運動技能測驗應用概化理論，可以得到整體測驗的概化信度係數。2.概化研究可以了解各種變異來源佔總變異成份的比率，依測驗使用者的需求，決定最適當的測驗條件配置。

關鍵字

概化理論 ； 運動技能測驗 ； 概化研究 ； 決斷研究

並列摘要

The purpose of this study was to apply generalizability(G)theory deciding thecriterion for trials-to-criterion tests in the psychomotor test.The subjects were threetrials-to-criterion tests.The object of measurement is subjects variance.Thedesigns of this study are a one-facet crossed-design.We approach different sourcesof score variation(subjecs,criterion and interaction)and the phi coefficient of multiplecriterion by using a GENeral purpose analysis Of VAriance system(GENOVA).The conclusion were:1.The true variance of tests are acceptable and the percentage of total variance arefrom 12.5 to 16.1%.2.The criterion variance of tests are small.They are from 0.0 to 2.2% in thepercentage of total variance.Therefore,it is very similar between G coefficientand phi coefficient.3.The main soure of variation is residual error of test.they are from 81.7 to 87.5%.4.The trials-to-criterion tests of psychomotor test are:(1)From 6 to 7 attempts ofcriterion yield mediocre levels of estimated phi coefficient.(2)From 21 to 28attempts of criterion yield reasonable levels of estimated phi coefficient.

並列關鍵字

Generalizability Theory ； Psychomotor Test ； Generalizability Study ； Decision Study