• 期刊

Weighted Penalty Model for Content Balancing in CAT

This paper presents the Weighted Penalty Model (WPM) for content balancing in computer adaptive testing. The WPM balances content properties as well as other non-statistical constraints, while simultaneously considering item information and the scarcity of items relative to content constraints, by assigning a penalty value to each eligible item in the item pool. Items with smaller penalty values are more desirable for selection. The WPM can be applied to both fixed/variable-length CAT once proper constraint codes and weights have been assigned. In this paper, applications of the WPM are presented using both empirical and simulated data. These analyses demonstrate the success of the approach in content balancing and in achieving targeted item exposure rates in the administration of CAT.

• 期刊

Conditional Randomesque Method for Item Exposure Control in CAT

One of the biggest concerns in the implementation of computerized adaptive tests (CAT) is that some items in the item pool are overexposed while some items are underexposed or never administered. The purpose of this study is to present the conditional randomesque (CR) method and assess its performance using simulation based on real item pool data under various study conditions. The CR method is compared to the other two item selection methods: the none_IEC method and the Stocking and Lewis Conditional Multinomial method. The results show that there is some precision loss. However, the conditional and overall maximum item exposure rates are much smaller. In addition, the pool usage is greatly improved.

• 期刊

A Study of E+C Blended Learning Courses for School Candidate Principals: Application of Fuzzy IPA Method

The main purpose of the study is to explore perceived importance and performance of E+C courses for school candidate principals on the 4 instructional objectives. Importance-performance analysis (IPA), fuzzy statistics and nonparametric statistics are employed in the study and totally 179 subjects of Taiwan school candidate principals are included. Briefly, the main findings are as follows: 1. most of the candidate principals who attended the E+C courses perceived high importance and performance of CFP, FSD, PPLCs, CCM courses on the 4 instructional objectives. 2. school candidate principals with different personal backgrounds (gender, age, years of experience, and educational stage) perceive no different importance and performance of CFP, FSD, PPLCs, CCM courses respectively on the 4 instructional objectives. 3. re-evaluate the feasibility of ＂online learning and HBS face-to-face learning＂ in planning campus & facilities and managing campus crisis.

• 期刊

The Impact of Learning Styles on Academic Achievement

This study investigated the impact of learning styles on academic performance of 9th graders in middle school. 140 students in two cities of Guangdong Province, China, were recruited. Learning Style Inventory and Multi-dimensional Group Intelligence Test for Adolescence were administered. Also, scores of High School Entrance Examination was collected about one month after the examination date. Results showed that students who prefer to study through discussion with peers are more likely to achieve better academic achievement when controlling for other covariates. Two types of learning styles, discussion and programming, showed additional contribution (ΔR^2 = .10) to predict later academic achievement beyond gender and intelligence.

• 期刊

Using Fixed Effect Model to Adjust WIOA Performance Employment Rate Measure for Rehabilitation Services Administration

The purpose of this study was to use a fixed effect model applying rehabilitation data to establish each state's ＂adjusted levels of performance＂ for one of the primary performance indicators-employment rate. Since the Workforce Innovation and Opportunity Act (WIOA) requires the use of a statistical adjustment model for setting targets for core performance indicators, the Departments of Labor and Education have collaborated to develop a Statistical Adjustment Model per Sec.116 (b)(3)(A)(viii) of WIOA. Adjusted performance measures can be used to evaluate the relative performance across states that serve individuals with disabilities. State-level actual performance outcomes are a function of the characteristics of the participants being served and the labor market conditions where those participants are being served. The study results show that using fixed-effect prediction models based on vocational rehabilitation (VR) participants' characteristics and local economic conditions to adjust performance measures can allow policy makers or state VR administrators to set more accurate and realistic goals and standards for state performance measures and separate them from those factors that VR program administrators are able to control.