AN ELECTRONIC ASSESSMENT OF SUB-METER SCALAR SKILL: SPATIAL CONCEPTIONS OF RELATIONSHIP AND MAGNITUDE TASK (SCRAMT)

Spatial thinking is a unifying theme in the sciences and critically important in the geosciences. The ability to conceptualize spatial size and scale is critical to understanding many scientific concepts and may play a role in whether students persist in science, technology, engineering, and math (STEM) fields.

To assess spatial scalar skill, we designed an online, nationally relevant, valid, and reliable survey that measures undergraduate student familiarity with sub-meter metric scale and common sub-meter size science objects, as well as sub-meter scalar skill informed by Cheek, LaDue, and Shipley’s (2017) conceptual framework of scale. The Scalar Conceptions of Relationship and Magnitude Task (SCRAMT) was modified from the scale of objects questionnaire (SOQ) created by Tretter, Jones, Andre, Negishi, & Minogue (2006). Tasks in the SCRAMT were informed by Cheek et al. (2017) to assess scalar relationship and scalar magnitude skill in undergraduate students. Scalar relationship was assessed by ordering metric measurements and objects by size. Spatial magnitude was assessed by accurately placing each object in the correct magnitude “bin.”. Additionally, two tasks captured participant familiarity with metric units and the objects featured on the SCRAMT survey. The overall SCRAMT design followed the tailored design methodological framework of Dillman, Smyth, and Christian (2014).

Amazon’s Mechanical Turk was used to obtain a national sample of 181 STEM and 88 Non-STEM undergraduates. The metric measurement and referent object familiarity tasks are reliable (5 items; α = .848 and 7 items; α = .782, respectively) and factor in logical factor loading indicating construct reliability. Additionally, the metric measurement and referent object rank order tasks are reliable (5 items; α = .847 and 7 items; α = .757, respectively). The magnitude binning task, when isolated, was not reliable (7 items; α = .353); however, when used as a part of the entire SCRAMT, the instrument is reliable (14 items; α = .759). Findings may impact how educators use reference frames and scientific notation in the classroom.