Testing a Measure of Growth Mindset–Relevant Behavior: The Persistence, Effort Investment, Resilience, Challenge-Seeking (PERC) Task

• In Event: 29.079 - Investigating Performance Tasks as Measures of Social and Emotional Learning Skills

Abstract

Theoretical framework
Research has shown that students’ beliefs about intelligence can affect achievement in school. Believing intelligence is a malleable trait that can be developed (growth mindset) can increase achievement, whereas believing intelligence is immutable (fixed mindset) can blunt academic success (Dweck, 1999).

Students’ behavior is a mechanism through which these beliefs are expected to operate. Students with more of a growth mindset are thought to display more adaptive self-management behaviors (challenge-seeking, effort investment, and persistence). By contrast, those with more of a fixed mindset are expected to display more helpless behaviors (avoiding challenge, not persisting).

Although some laboratory research supports these predictions, more often researchers have used self-report measures to tap mindset-relevant behaviors (Burnette et al., 2013). These measures, though beneficial for their efficiency, are vulnerable to limitations including reference and recall bias, which can compromise their accuracy (Duckworth & Yeager, 2015). Thus, a behavioral assessment could provide a more sound measure of students’ mindset-relevant behaviors.

We introduce PERC, a computer-based task that taps Persistence, Effort, Resilience, and Challenge-seeking behaviors. The task follows the procedure developed by Mueller & Dweck (1998), but can be administered online in approximately 20 minutes. Here, we investigate the properties and validity of the task.

Methods
1394 middle school students (Mage = 11.43, SD = .79; 52% girls) attending an urban school in the northeastern United States participated in this research.

Students completed the PERC task in fall, 2015 (see Figure 3 for a task description, and Figure 4 for descriptive information). One month earlier, students completed a self-report survey assessing relevant constructs (see Table 3).

Objectives
We investigated the following research questions:
1. Were task constructs (Persistence, Effort, Challenge-seeking and Resilience) statistically independent from each other,
2. Did the task constructs relate to relevant self-report measures,
3. Did the task constructs predict achievement test scores,
4. Did the task constructs predict achievement over and above self-report measures?

Results
Measures of Persistence, Effort, Challenge-seeking and Resilience were minimally correlated, if at all (Table 4), suggesting that each measure was statistically independent.

Self-reports. The task predicted students’ Mindset, Mastery Goals, Positive Views of Effort, and Grit (Table 5). As would be expected, different task components related differently to self-report constructs. Effort Investment (practicing) was most related to students’ mastery goals, while Persistence (time on difficult problems) was most related to effort beliefs and mindset. All correlations greater than .08 remained significant in partial correlations controlling for all other self-report and behavioral constructs.

Achievement. Challenge-seeking and Persistence were positively associated with students’ Math and English achievement (Table 5). These associations remained significant in partial correlations controlling for other self-report and behavioral constructs. Challenge-seeking and persistence also explained unique variance in achievement, evincing their independence.

Significance
The PERC task assesses four independent self-management behaviors. The task largely predicts self-report measures and achievement as expected. Thus, the PERC task offers a promising tool for those interested in assessing mindset-relevant behavior. In future analyses we will examine whether increases in growth mindset are reflected in task measures over time.

Authors

• Tenelle Porter, University of California - Davis

• Lisa Blackwell, Mindset Works

• Sylvia Rodriguez, Mindset Works

• Diego Catalan Molina

• Kali Trzesniewski, University of California - Davis