Fig 1.
The proposed one-dimensional learner- to teacher-centered framework.
Examples of student behaviors and instructor practices at the learner-centered end (in gray) juxtapose those that are more teacher-centered (black) at the other end of the framework. Learner-centered descriptors (gray) were expected to positively correlate with each other, while teacher-centered descriptors (black) were expected to positively correlate with each other. Negative correlations (dashed line) were expected between two related but contrasting descriptors, as both would fall on opposite ends of the learner- to teacher-centered framework. For example, deep approaches are more learner-centered, while surface approaches are more teacher-centered; a student that engaged in deeper learning approaches would not be expected to engage in as many surface approaches, or vice versa.
Table 1.
Five instruments for comparing perceptions of learner-centeredness.
Within each student and instructor instrument exists primary and secondary subscales that we used in our study; we indicate the possible score ranges for each subscale and at which end of the learner-centered (LC) gradient a high score on that subscale would capture.
Table 2.
Pearson correlations between primary instructor subscales, primary student subscales, and RTOP scores across all sections.
Table 3.
Summary statistics for student, instructor, and expert observer instruments.
Fig 2.
Twelve course sections are shown as open circles in student primary subscale space using NMS.
(a) Several components of the ETL and KLA positively correlate with Axis 1, the strategy axis. Conceptual change of the ATI also correlated at the positive end of axis one, though was not included in the ordination figure. (b) The Deep and Surface approaches of the R-SPQ-2F associate with the positive and negative ends of Axis 2, the motive axis, respectively. In this panel, the relative symbol size of the 12 course sections are coded by RTOP score; high RTOP scores (i.e., larger circles) correlate with the positive end of Axis 2.
Table 4.
Mantel tests between primary and secondary subscale scores.
Correlation coefficients and p-values in upper corner compare primary subscale scores, while correlation coefficients in the lower corner compare secondary subscale scores.
Fig 3.
Twelve introductory biology course sections independently clustered by student and instructor primary subscales.
In the dendrogram, information remaining (%) is indicative of the strength of the relationship between class sections; clusters joined with greater information remaining are more closely related. Sections are clustered by student perceptions in the dendrogram to the left (a), while the same sections are clustered by instructor perceptions in the right dendrogram (b). Identical course sections are connected in the center to aid in visualization of similarities; connector lines patterns denote enrollment size (dashed line ≤70 students, solid line = 71–150 students, bolded double line >150 students [one section, n = 391]). In the instructor dendrogram, Cluster A is the true learner-centered cluster; Cluster B is characterized by internal confusion within individual faculty; Cluster C is epitomized by the conflict in perspectives among groups; and Cluster D is the non-learner-centered cluster based on instructor and student perceptions.