Literature

Most conceptualizations of CT converge on the idea that it involves “an individual’s capability to […] identify central issues and assumptions in an argument, recognize important relationships, make correct references from the data, deduce conclusions from information or data provided, interpret whether conclusions are warranted based on given data, evaluate evidence of authority, make self-corrections, and solve problems” (Pascarella and Terenzini Reference Pascarella and Terenzini2005, 156).

One commonly advanced strategy to develop student CT is known as “scaffolding.” This entails using targeted assignments to help students break down complex judgments into a series of simpler ones punctuated by guided feedback before asking them to tackle more complex types of reflection (Sharma and Hannafin Reference Sharma and Hannafin2004; van de Pol, Volman, and Beishuizen Reference van de Pol, Volman and Beishuizen2010).

Adding assignments online (scaffolded or otherwise) also has been shown to motivate students to complete assigned readings, increase participation in class discussion, and improve performance on exams (Brothen and Wambach Reference Brothen and Wambach2004; Johnson and Kiviniemi Reference Johnson and Kiviniemi2009). In brief, more is better.

This view is not universally endorsed, however. Some advocate a “less-is-more” approach. The idea is to “shift from a broader focus on ‘coverage’ of a variety of types of document and concepts to deeper focus on a more narrow range of topics and/or assignments” (Skurat Harris, Nier-Weber, and Borgman Reference Harris, Heidi, Borgman, Ruefman and Scheg2016, 19).

The rationale for adopting a more minimalist approach stems from the unique characteristics of the online medium. Communication is more uncertain in online courses. Students may not choose to click on all the available course materials. The more materials there are, the higher the likelihood that something important will be missed. Furthermore, because the online environment is usually text based, if students do not read as well as they should (or professors do not write as clearly as they think they do), the potential for miscommunication may be greater. Finally, waiting for an email response to a query takes time; students may not seek clarification if they do not believe an answer will be forthcoming in a convenient time frame.

Advocates of the minimalist approach stress the need to declutter online courses, extend the period between deadlines, and focus scarce student attention on a limited quantity of materials and assignments.

Method and Data

I teach an introductory course in political theory required for political science majors at a midsize American public university.Footnote ¹ For the past eight years, I have collected a dataset consisting of essays scored for their CT using a rubric adapted from the Washington State University Critical Thinking Initiative (Condon and Kelly-Riley Reference Condon and Kelly-Riley2004).Footnote ²

During this time, both minimalist and more scaffolded approaches were sometimes adopted. Scaffolded semesters were identified by the presence of specifically designed scaffolding exercises. Minimalist semesters were identified by the small number of required assignments (i.e., fewer than seven). Anything else was put into a residual category. Although the selection criteria are simple, table 1 shows that they align well with other dimensions of the concepts.

Table 1

Summary Statistics: Three Types of Semesters (Standard Deviation)

Notes: Means sharing the same superscript are not significantly different from one another (Tukey-Kramer Honest Significance Test; p<0.05).

* Nominal variables are analyzed using a paired one-tailed chi-square test. Means sharing the same superscript are not significantly different from one another (p<0.05).

The scaffolded semesters have more assignments, more specific scaffolding assignments, more scaffolded CT quiz questions, more quiz questions overall, a greater variety of reading assignments, more required discussion, and more assignments with individualized feedback. The semesters with a minimalist approach have less of all of these elements. The residual category usually falls in between these two approaches. Space does not permit a discussion of the specific types of scaffolding exercises used, but several are discussed elsewhere (see Phillips Reference Phillips2018). Table 1 also shows that sections using the different approaches are not statistically different in terms of academic qualifications of students.

Summary of Findings

Overall, there were no statistically significant differences in mean CT scores across the three different types of semesters (table 2). The extra work that went into scaffolding online classes yielded no aggregate dividends in terms of measured CT. Table 2 also lists secondary learning outcomes for context.Footnote ³ Overall, there were few differences between the minimalist and the scaffolded semesters. The only statistically significant difference is that students withdraw less often from the minimalist semesters. Despite the author’s best efforts, alternate methods of analyzing the data (e.g., regression analysis) did not uncover further differences.

Table 2

Dependent Variables of Interest: Three Semester Types (Standard Deviation)

Notes: Means sharing the same superscript are not significantly different from one another (Tukey-Kramer Honest Significance Test; p<0.05).

* Dummy variables are analyzed using a paired one-tailed chi-square test. Means sharing the same superscript are not significantly different from one another (p<0.05).