Despite the popularity of the Likert scale response format for data collection in the social sciences including education, its continued use has raised some serious methodological questions. It was therefore against this background that the current study established the psychometric properties of the Likert scale in relation to an alternative response format called the expanded scale format. Using the instrumentation research design, the study compared the extended scale format and Likert scale format using the graded response model of the Item Response Theory (IRT). The study was guided by four research questions, while a sample of 2742 undergraduate students were used for the study. The sample was drawn using the convenience sampling technique. The instrument used for data collection was the Rosenberg Self-Esteem Scale, which was converted from the original Likert scale format to the expanded format. The instrument was assessed for validity and reliability and was shown to possess preliminary psychometric tools for further investigation. Data collection was done using an online tool, Question Pro® to administer the two version of the instrument to programmed respondents through social media platforms. Data analysis was done using exploratory factor analysis, confirmatory factor analysis, difficulty and discrimination parameters where applicable. The result showed that across all psychometric indicators, the expanded scale format was more suitable than the Likert version. On the basis of the result obtained, it was recommended that psychometricians develop future scales using more of the expanded format.