Validation of Measures

4 Analysis and Result 1 Verification of Measures In this survey, we improved the measures and evaluated reliability and effectiveness. First of all, alphas values ​​for Cronbach's market orientation, gradual and radical service innovation, market and financial performance, market turmoil, technical disruption, and competition intensity range from 0.70 to 0.92, both recommended levels It is over 0.70 Nunnally, 1978). Based on the first cut, internal consistency and reliability in 4 stages has been established.

Psychometricists acknowledge that the validity of the measurement scale has been established from the outset. As Nunnally (1978) pointed out, "Although the effectiveness of policies should not be tested after policies are established, the effectiveness of measures should be secured through planning and construction procedures" (p. 258). By carefully selecting the initial scale item, the scale will be defined as "relevance of content" and "the score or scale used will represent the concept of generalization" (Bohrnstedt, 1970 Year, page 91). When discussing the relevance of content, psychometricists often rely on the "domain sampling model" (Bohrnstedt, 1970; Nunnally, 1978). And it assumes that there is a domain corresponding to the content of interest for each variable. You can select candidate content areas

Remember that the effectiveness evaluation effectiveness refers to the accuracy of the metric. The measurement standard is effective to accurately measure the concept that you want to measure. There are several ways to evaluate effectiveness. We will focus on three aspects of content validity, composition validity, standards validity. Content Validity Content validity is to subjectively or logically evaluate the extent to which metrics fully reflect the complete domain or entire content of the measured concept. For example, if you are interested in evaluating the skills of elementary school math skills and if only the questions about additions are covered in the test, the validity of the test is lost. In other words, the test does not measure the complete domain or the entire contents of the mathematical ability concept by not including the problem of evaluation of subtraction, multiplication, division.

Effectiveness refers to the extent to which the tool measures the amount that the tool should be measured and implemented in the designed way. If it is almost impossible, 100% effective instrumentation is rare, so effectiveness is measured in normalness. As a process, verification includes the collection and analysis of data to assess the accuracy of the instrument. There are many statistical tests and measures to evaluate the effectiveness of quantitative measures. And it usually includes a pilot test. In the remainder of the discussion, we will focus on external validity and content validity.