Alana Unfried (California State University Monterey Bay), Douglas Whitaker (Mount Saint Vincent University), Leyla Batakci (Elizabethtown College), Marjorie Bond (Pennsylvania State University), April Kerby-Helm (Winona State University)
Abstract
Background. Attitudes matter in education: student attitudes are known to correlate with content retention and future work in the discipline; instructor attitudes relate to chosen teaching methods and impact student attitudes and outcomes. However, there has been a lack of modern, validated instruments for cohesively measuring student attitudes, instructor attitudes, and the learning environment in introductory statistics and data science courses. The Motivational Attitudes in Statistics and Data Science Education (MASDER) NSF project has developed a family of six instruments that allow statistics and data science educators and researchers to identify evidence-based best practices in our disciplines. The Student Survey of Motivational Attitudes toward Statistics (S-SOMAS) measures motivational attitudes towards statistics in introductory undergraduate courses. I-SOMAS measures instructors’ attitudes towards teaching statistics, and EPIC-S (Environment Pedagogy Instructor and Course for Statistics) collects characteristics of the course, the institutions, pedagogical methods that are used in the course, and teacher-student relationships. Parallel instruments were developed for the discipline of data science. We will discuss the structure of these six instruments, and describe their rigorous development process, using S-SOMAS as an example. We will also discuss how to access and implement the instruments.
Methods. Each of the MASDER instruments was developed using a staggered, iterative design process. Because the literature about student attitudes in statistics is most developed, student instruments were developed before instructor instruments, and statistics instruments were developed before data science instruments. The four attitudinal instruments are grounded in Eccles’s Expectancy Value Theory, a psychological theory of motivation used across many contexts. Validity for the four psychometric instruments is established by the design process and psychometric analyses: confirmatory factor analysis and multidimensional item response theory using the Graded Response Model were employed in multiple pilot studies to result in robust final instruments.
Findings. The focus of the results will be on the S-SOMAS: more robust psychometric analyses have been performed on this instrument because of the large data collected using it (15,000+ responses across several pilot studies). Psychometric analyses support measuring 11 constructs using a 38-item instrument with good internal consistency reliability. Documents have been created to support researchers’ use of the instruments, and a description of the score interpretations for each construct will be provided as well as descriptions of the intended (and not intended) uses.
Implications For Teaching and For Research. Implementing this family of instruments can inform evidence-based best practices in statistics and data science education. Researchers can administer all surveys via the MASDER website (portal.sdsattitudes.com) with new data automatically added to the national sample. Researchers can access their own data and receive customized, automated reports comparing their sample to the national sample. Qualtrics and PDF files of the instruments will also be freely available for educational researchers to use in their specific context, allowing for rigorous study of connections between attitudes, the learning environment, and student outcomes in introductory statistics and data science courses.