การพัฒนาแบบทดสอบเชิงวินิจฉัยทางพุทธิปัญญา เรื่อง ความรู้สึกเชิงจำนวน สำหรับนักเรียนชั้นประถมศึกษาตอนต้น: การประยุกต์ใช้โมเดลจีดีไอเอ็นเอ Development of a Cognitive Diagnostic Test on Number Sense: An Application of GDINA Model

Main Article Content

วรางคณา ก้อมน้อย

Abstract

The objectives of this research were to develop and validate the cognitive diagnostic test on number sense for lower-elementary school students. Research resources consist of three sources which are 5 experts for validating research instrument, 152 samples for phase 1 tryout group who are 3rd-grade students in 2nd semester of 2018 academic year from a large, medium, and small school selected by multi-stage sampling, and 348 samples for phase 2 tryout group who are 3rd-grade students in 1st semester of 2019 academic year from a large school. The research instruments consisted of the evaluation form of the structural relationship between attributes of number sense and the evaluation form of item-objective congruence for multidimensional items.


               The research findings can be summarized as follow:


               The cognitive diagnostic test on number sense for lower-elementary school students has 20 items in the form of four multiple choices, aim to diagnose six attributes of number sense.


               And psychometric properties of the test revealed as follow; content validity of the test had and IOC value between .50 - .82, Q-matrix validation is found fit to data, criteria-related validity is found that number sense scores have been correlated to numeracy scores from NT at .01 level of statistical significance,  discriminant found that item discrimination index is between .204 - 1.000 and discrimination index at test level is .606, classification accuracy found that test-level classification accuracy is .734 and attribute-level classification accuracy is between .901 - .964, classification consistency found that test-level classification consistency is .652 and attribute-level classification consistency is between .842 - .949, and model fit statistics found that RMSEA2 is .021.

Article Details

Section
Dissertations