The Benefits of Experimentally Manipulated Student Agentic Engagement

• In Event: Agentic Engagement: A New, Important, and Progress-Enabling Educational Construct

Abstract

Purpose
The purpose of this paper was to present the first experimentally manipulated study in agentic engagement to assess its causal capacity to produce positive student outcomes, including students’ motivation, performance, and relationship with the teacher.

Theoretical Framework
Agentically-engaged learners are those who take the initiative, express their preferences, and ask questions to help them learn. Such proactively-engaged students generally make progress, achieve highly, and recruit greater autonomy support from their teachers (Matos et al., 2018; Reeve, 2013). These findings, however, are from correlational (albeit longitudinal) studies, so the causal effect of high agentic engagement on these outcomes remains undocumented. Accordingly, we conducted a laboratory study to experimentally manipulate the presence vs. absence of students’ agentic engagement. The hypothesis was that students trained to be agentically engaged (vs. untrained controls) would receive greater autonomy support and lesser teacher control and show greater academic progress, need satisfaction, performance, and intrinsic motivation and lesser need frustration.

Methods
In a laboratory setting, 121 teacher-student pairs of college-aged students engaged in an unfamiliar learning activity. The research design employed random assignment to conditions and was a 2 x 2 ANOVA. One independent variable was the presence vs. absence of a 12-minute training session for the student in how to be agentically engaged during the learning activity; the second independent variable was the presence vs. absence of a 12-minute training session for the teacher in how to be autonomy supportive toward the student. The ensuing 12-minute teaching-learning session was videotaped (with participants’ awareness and consent). After the teaching-learning session, the teacher and student completed a questionnaire, the student completed a 5-minute free-choice period (to assess intrinsically motivated behavior), and the videotapes were scored by objective raters.

Results
Dependent measures broken down by the two experimental manipulations appear in Table 1. Here we report the main effects for agentic engagement. The agentic engagement manipulation increased students’ perceived autonomy support, F(1, 117) = 4.38, p = .039, rater-scored autonomy support, F(1, 117) = 5.38, p = .023, and free-choice time, F(1, 117) = 3.89, p = .048, while it decreased perceived teacher control F(1, 117) = 7.34, p = .008, rater-scored teacher control, F(1, 117) = 4.30, p = .040, and need frustration, F(1, 117) = 4.22, p = .042, but it did not affect need satisfaction, F(1, 117) = 1.78, p = .185, perceived academic progress, F < 1, or actual performance, F < 1. [The autonomy support manipulation increased perceived autonomy support, rater-scored autonomy support, and perceived academic progress, while it decreased perceived teacher control and rater-scored teacher control.
A few two-way interactions also emerged.]

Mediation analyses (using bootstrapping) showed that the reason why agentic engagement produced its educational benefits (i.e., perceived progress, intrinsically motivated behavior) was because it recruited greater autonomy support and lesser control from the teacher.

Discussion and Scholarly Significance
Manipulated agentic engagement recruited greater autonomy support (and lesser control) from the teacher, which explained the benefits from high agentic engagement.

Authors

• Jiseul Sophia Ahn, Laval University

• Hye-Ryen Jang, Australian Catholic University

• Stephanie Hyewon Shin, Michigan State University