多元智能兴趣偏好量表在四川地区的初次修订——以美国的阿姆斯特朗量表为例

First Revision of Multiple Intelligence Interest Inventory in Sichuan Province:With Thomas Armstrong Scale as an Example

研究和应用与测评学生多元智能兴趣、偏好相关的量表,对学生智能的多元化评价方案的实施具有一定的理论与现实意义。以该类量表中具有代表性的阿姆斯特朗量表为测量工具,采取随机分层抽样的方法,对其在四川地区的适用性进行了考查。经过对该表的修订,共保留了33个题项,包括8个分量表。对修订后量表的信度和效度进行检验,结果:各分量表和总量表的内部一致性系数(a系数)0.61-0.83、分半信度(Split-half)0.57—0.74;两周后再测信度0.68-0.72;各题目与分量表总分相关0.521-0.719,均在0.01水平达到了显著相关;各分量表与总量表之间的相关(0.621-0.753)都大于各分量表之间的相关(0.241-0.564);因素分析中提取的8个因子,能解释总方差的52.807%,且各项目在相应因子上有较满意的因子载荷量(一般≥0.5)。结论:阿姆斯特朗量表修订后的信度和效度指标基本达到了测量学指标要求。并且简化了的量表更适合学生自测。

Research and application of the inventory for students＇multiple intelligence interest is of some theoretical and realistic significance in evaluating students＇ multiple intelligence. With Thomas Armstrong Scale as a representative measure tool, this paper adopts random stratified sampling method in the study of its applicability in Sichuan Province. The revised Scale reserves 33 - items, including 8 sub - scales. The reliability and validity of the revised Scale has been tested. The result shows that of the whole scale and subscales, the interior consistency coefficient ranges from 0.61 to o. 83 and the split - half reliability ranges from 0.57 -0.74. Two weeks later, the retest tells that the reliability is between 0.68 and 0.72. The correlation of each item score with the total score in sub -scales is 0.521 to 0.719 （ P 〈 0.01 ）. The correlation between each subscale and the total scale is greater than those within the 8 subscales （0. 621 -0. 753 〉0. 241 -0. 564）. The 8 factors obtained from factor analysis make clear 52. 807% of the total variance. The factor load capacity of each item is satisfactory （≥0.5）. Thus the paper has drawn such a conclusion that both the reliability and the validity of the revised Thomas Armstrong Scale are up to the basic requirements of surveying index. And the revised edition is more suitable for students＇serf - test.