Reliability of measurement

Abstract

In quantitative educational research, instruments for data collection should lead to the validity and reliability of results. A question that researchers are frequently referred to is how high or how large must the reliability coefficient be. This depends on the purpose and importance of the decisions to be made, based on the scores obtained by the subjects. Reliability is a necessary attribute in all quantitative tests used to assess someone. The present paper aims to address the reliability of the measurement. To achieve the following chapters are reviewed: theory of measurement error, domain-sample model, temporal stability or stability coefficient, parallel forms or stability coefficient and equivalence and some of the most used internal consistency coefficients.