Diferencias entre coeficientes alfa, con muestras y partes pequeñas; adenda

This paper presenting a computer program written in VB 6.0, to calculate the difference between internal consistency coefficients (Cronbach’s alpha) obtained in independent small samples and with instruments have a small number of parts or items. The comparison of reliability coefficients allow to identify possible differences in amount of measurement error in instruments; this methodology use the hypothesis testing approach for test the null hypothesis of equally reliability coefficients. This situation is tending to be common in clinical practice between psychologists or allied career, and even in the construction phases of instruments of measurement, for example in pilot samples. The proposed technique is from the work of Feldt and Kim (2006), and offers a viable and interesting methodological proposal that expands the analysis of the reliability of instruments of psychological and educational measurement.

Determining Factors of Community Commitment in Automotive and Sneaker Communities

This study to examine the effect of brand community integration and perceived benefits on community commitment as moderated by type of community. This study uses a quantitative approach with a hypothesis testing approach. The sample of this research is the automotive and sneaker communities. The data analysis used was SEM analysis using PLS software. Variable Brand Community Integration (x1) does not have the effect of significantly at variable community commitment (Y) and the perceived benefit (X2 ) effect is significant to the variable community commitment (Y). The study explains how importance of brand community integration and perceived benefits have a contribution to building brand community commitment. Companies can pay attention to community involvement in increasing their commitment to continue using the product. Besides, Members of communities will tend to pay attention to the benefits that they get to follow the community, It is can be used by companies to improve the provision of benefits to the members, and will give benefits for the company to increase the commitment of the customers.

Five common pitfalls in eye-tracking research

Eye tracking has become an increasingly popular research methodology among language researchers to examine online cognitive processing of second-language (L2) speakers and bilinguals. As the scope of eye-tracking research expands, there is a need to ensure that the methodology is used properly, so as to safeguard the validity of research findings and the empirical basis upon which theories are built. We present five pitfalls in eye-tracking research that may threaten the internal or external validity of the study. We offer concrete recommendations to avoid each pitfall and demonstrate how these recommendations can be implemented so researchers may take full advantage of the opportunities of eye-tracking research. The five pitfalls occur at all stages of the research cycle. We emphasize (1) adopting a confirmatory (theory-driven, hypothesis-testing) approach to research, (2) selecting and handling eye-movement measures in a principled manner, (3) aligning the research design with the eye tracker properties (spatial accuracy, precision), (4) inspecting raw, sample-level eye-tracking data and their preprocessing by the eye-tracking software. We conclude by inviting eye-tracking researchers (5) to embrace open science practices to enhance the replication and reproducibility of their work.

Possibility Measure of Accepting Statistical Hypothesis

Taking advantage of the possibility of fuzzy test statistic falling in the rejection region, a statistical hypothesis testing approach for fuzzy data is proposed in this study. In contrast to classical statistical testing, which yields a binary decision to reject or to accept a null hypothesis, the proposed approach is to determine the possibility of accepting a null hypothesis (or alternative hypothesis). When data are crisp, the proposed approach reduces to the classical hypothesis testing approach.