Empirical study
Implicit intelligence beliefs of computer science students: Exploring change across the semester

https://doi.org/10.1016/j.cedpsych.2016.10.003Get rights and content

Highlights

  • Investigated change in computer science students’ implicit intelligence theories.

  • Incremental theories decreased and entity theories increased for all students.

  • Magnitude of change differed across motivated engagement profiles.

  • Relationship between change and achievement outcomes varied across courses.

  • Changes in implicit theories weakly predicted achievement outcomes.

Abstract

This study investigated introductory computer science (CS1) students’ implicit beliefs of intelligence. Referencing Dweck and Leggett’s (1988) framework for implicit beliefs of intelligence, we examined how (1) students’ implicit beliefs changed over the course of a semester, (2) these changes differed as a function of course enrollment and students’ motivated self-regulated engagement profile, and (3) implicit beliefs predicted student learning based on standardized course grades and performance on a computational thinking knowledge test. For all students, there were significant increases in entity beliefs and significant decreases in incremental beliefs across the semester. However, examination of effect sizes suggests that significant findings for change across time were driven by changes in specific subpopulations of students. Moreover, results showed that students endorsed incremental belief more strongly than entity belief at both the beginning and end of the semester. Furthermore, the magnitude of changes differed based on students’ motivated self-regulated engagement profiles. Additionally, students’ achievement outcomes were weakly predicted by their implicit beliefs of intelligence. Finally, results showed that the relationship between changes in implicit intelligence beliefs and student achievement varied across different CS1 courses. Theoretical implications for implicit intelligence beliefs and recommendations for STEM educators are discussed.

Introduction

Student achievement and retention in science, technology, engineering, and mathematics (STEM) courses and majors have been the focal point of much research over the past two decades (e.g., Gasiewski et al., 2012, Seymour and Hewitt, 1997, Watkins and Mazur, 2013). In today’s rapidly advancing technological environment, the need to attract and retain students in STEM majors is possibly greater than ever before. In fact, it has been proposed that STEM-related fields will grow at nearly twice the rate as non-STEM fields between 2008 and 2018 (Langdon, McKittrick, Beede, Khan, & Doms, 2011). Additionally, the number of jobs in STEM fields grew at nearly three times the rate as the number of jobs in non-STEM fields during the first decade of the 21st century (Langdon et al., 2011). However, research has demonstrated that a large percentage of students pursuing a STEM major switch to a non-STEM major before graduation (Seymour and Hewitt, 1997, Watkins and Mazur, 2013). As a result, increasing the number of students who declare a STEM-related major and retaining those students until they graduate and enter the workforce is a concern for post-secondary administrators.

These issues facing the STEM community call for more exploration of the factors that influence success and retention in STEM-related majors and courses. Because academic achievement predicts retention in courses and majors in and out of the STEM domain (DeBerard et al., 2004, Stinebrickner and Stinebrickner, 2013), understanding the factors that impact achievement is a major priority for educational researchers. Literature has shown how factors such as motivation (Allen, Robbins, Casillas, & Oh, 2008), self-regulation (see Duckworth & Carlson, 2013 for a review), self-efficacy (Komarraju, Swanson, & Nadler, 2014), goal orientation (Hsieh, Sullivan, & Guerra, 2007), and students’ implicit beliefs about the nature of intelligence (Blackwell et al., 2007, Dupeyrat and Mariné, 2005) are positively associated with achievement in and out of STEM. However, although research has associated implicit intelligence beliefs with student success and retention in STEM domains (Dai & Cromley, 2014), less is known about the nature of change in STEM students’ implicit intelligence beliefs across time. The purpose of our study is to further the educational community’s understanding of the relationships between implicit intelligence beliefs and achievement in STEM by examining (a) how implicit intelligence beliefs change across time for students in a suite of introductory-level computer science courses, (b) how motivational and environmental factors impact implicit intelligence beliefs, and (c) how implicit intelligence beliefs predict achievement and learning outcomes in introductory-computer science.

Studies related to implicit intelligence beliefs have highlighted how students’ beliefs about their ability to master skills and learn course material have important implications for their learning, achievement, and persistence in STEM courses (e.g., Dai and Cromley, 2014, Shively and Ryan, 2013). According to Dweck (2000), implicit intelligence beliefs refer to people’s general beliefs about whether their intelligence is a fixed trait (entity belief) or a malleable quality that can be enhanced through learning and effort (incremental belief). Commonly known as fixed (entity belief) and growth (incremental belief) mindsets, Dweck and Leggett’s (1988) framework for implicit beliefs of intelligence has received widespread attention in research.

A significant body of literature has explored how students’ implicit beliefs about the nature of intelligence influence their achievement and motivation. Based largely on the work of Stanford psychologist Carol Dweck, this literature affirms that students of all ages possess varying levels of entity and incremental intelligence beliefs or mindsets (Carr and Dweck, 2011, Dweck, 2000, Dweck and Leggett, 1988, Yeager and Dweck, 2012). Students who score highly on the entity belief of intelligence believe that intelligence is a fixed entity (i.e., either you are born intelligent or you are not). Students with a strong entity belief believe that no matter how much time and effort they put into learning, they are bounded by their natural level of intelligence and their intellectual ability cannot be increased through their own efforts (Dweck & Leggett, 1988). Conversely, students who score highly on the incremental belief of intelligence believe intelligence is a malleable trait that can be enhanced through learning, time, and effort. Students with a strong incremental belief believe their intellectual ability can be cultivated and increased through their own efforts (Dweck & Leggett, 1988).

Differential patterns in motivation and reactions to failure have been associated with each of these implicit intelligence beliefs. Students who have a strong entity belief of intelligence tend to focus more on their performance outcomes (e.g., getting a passing grade or appearing smart to one’s classmates), attribute failure to a lack of ability, and believe that working hard reflects a lack of ability rather than a commitment to improvement (Dweck, 2000, Dweck and Molden, 2005). Students with an entity belief of intelligence believe poor performance reflects inadequacies in their intelligence; in their eyes, giving a purposeful effort to improve would only confirm their inadequacies to their classmates. Alternatively, students with an incremental belief of intelligence are more likely to set mastery goals geared toward gaining a complete understanding of the course material, believe that effort is a means to becoming more intelligent, and view failure as an opportunity for improvement (Dweck, 2000, Dweck and Molden, 2005). Students who have a strong incremental belief of intelligence do not believe that poor performance indicates an unchangeable deficit in their intelligence. Instead, they believe that with sufficient effort, intelligence can be increased and performance can be enhanced. Differential outcomes in academic achievement are also associated with scoring highly on either the incremental or entity beliefs of intelligence. Specifically, it has been found that students who strongly endorse an incremental belief of intelligence tend to receive higher grades than those who endorse an entity belief (Blackwell et al., 2007, Dupeyrat and Mariné, 2005). Relatedly, Stump, Husman, and Corby (2014) found that undergraduate engineering students who endorse an incremental belief of intelligence are more likely to use active learning strategies and collaborate with their peers, both of which are activities that can promote meaningful learning of course content, and thus, academic achievement.

Many researchers have treated implicit intelligence beliefs as a trait-like characteristic that remains stable over time (e.g., Robins & Pals, 2002) and focused their efforts on exploring how implicit beliefs of intelligence relate to motivation (Chen & Pajares, 2010) and achievement (Blackwell et al., 2007, Gonida et al., 2006). However, other research has demonstrated a more dynamic, changing nature of implicit intelligence beliefs across time (e.g., Aronson et al., 2002, Gonida et al., 2006). Rather than representing two opposite ends of a spectrum (e.g., you are either an entity theorist or you are an incremental theorist), research has begun to show that students’ hold varying levels of both incremental and entity intelligence beliefs and that these are individually dynamic and changing over time. Research involving STEM students in fields such as engineering (Reid & Ferguson, 2014), biology (Dai & Cromley, 2014), mathematics (Shively & Ryan, 2013), and computer science (Simon et al., 2008) has demonstrated decreases in incremental belief and increases in entity belief across time. These trends are associated with negative outcomes such as lower course performance (Shively & Ryan, 2013) and higher attrition rates in STEM courses and majors (Dai & Cromley, 2014).

Research has explored changes in students’ implicit beliefs of intelligence with and without the presence of interventions designed to influence students’ implicit beliefs. Intervention studies leading to change in STEM students’ implicit intelligence beliefs are described next, followed by descriptions of studies that have detected changes in implicit intelligence beliefs without an intervention.

First, although research has found that interventions can enhance adolescents’ incremental belief of mathematics intelligence (Blackwell et al., 2007, Good et al., 2003) and undergraduate African-American students’ incremental belief of general intelligence (Aronson et al., 2002), studies involving undergraduate STEM students have been mixed. For example, Reid and Ferguson (2014) implemented a team-based project designed to enhance the incremental intelligence beliefs of 84 first-year engineering students. In their study, groups of students in an introductory-level engineering course completed an assignment that required them to design a device that would alleviate the impact of poverty in select countries or populations. The task was designed for students to initially struggle before succeeding to develop a suitable device. These researchers were interested in examining how such a transformative experience would impact students’ beliefs about the nature of intelligence. Participants’ implicit intelligence beliefs were measured at the beginning of course during the fall semester (pretest) and the end of the spring semester (posttest), one semester after completing the team-based project. Results indicated that the transformative, team-based project did not lead to significant changes in incremental or entity beliefs across the school year for these first-year engineering students.

Similarly, a study by Simon et al. (2008) found that training engineering students to adopt an incremental belief of intelligence did not significantly increase students’ incremental belief. In their study, Simon et al. (2008) drew participants from two sections of a one-semester introductory computer-science course designed for engineering students. One section acted as the control group and the second acted as the experimental group. Participants in the experimental section (n = 84) received a lecture about incremental and entity beliefs of intelligence and took part in two writing exercises to apply the information learned in the lecture. In the first exercise, experimental participants wrote about their real-life experiences of learning something new that they initially struggled to understand. In the second exercise, experimental participants wrote a letter to a beginning programmer emphasizing how their programming intelligence develops across time. The control group (n = 97) received no such lecture and did not participate in the writing exercises. Participants’ incremental and entity beliefs were measured at the beginning of the semester and again at the end of the semester, eight weeks after the experimental group received the lecture and completed both exercises. Results showed that incremental beliefs of experimental group participants did not differ significantly from the control group at either the beginning or end of the semester.

Although Reid and Ferguson, 2014, Simon et al., 2008 were unable to induce change in STEM students’ implicit beliefs, Cutts, Cutts, Draper, O’Donnell, and Saffrey (2010) found that a more intensive intervention may be effective. In this study, the researchers targeted the incremental intelligence beliefs of students in an undergraduate programming course. They designed and implemented three different intervention components that could stand alone or used in combination. In the first intervention component, participants attended a series of four 15-min lectures about various aspects of Dweck and Leggett (1988) framework for implicit theories of intelligence. In the second intervention component, participants were given a crib sheet containing questions, hints, and strategies to use when they encountered difficult course material. The rationale for the crib sheet was that showing students they could solve difficult problems by adjusting learning strategies should increase students’ incremental beliefs. Finally, the third intervention component involved providing students with incremental belief-oriented feedback on assignments. This study was designed using a 2 (lecture/no lecture) × 2 (crib sheet/no crib sheet) × 2 (feedback/no feedback) factorial design, resulting in eight treatment conditions, ranging from no intervention to receiving all three intervention components. Results indicated that, across the academic year, students who received instruction about implicit intelligence beliefs displayed a significant increase in their incremental belief scores from the beginning until the end of the 26-week course. Furthermore, participants in groups that received incremental belief training and participants in groups that received incremental belief-oriented feedback performed higher on the final course exam than students in the control group and students in the crib sheet-only group. Findings indicated no relationship between the use of crib sheets and students’ incremental beliefs of intelligence.

These findings demonstrate that interventions designed to impact STEM students’ incremental beliefs have achieved mixed results. Reid and Ferguson’s (2014) attempt to increase engineering students’ incremental intelligence beliefs through a transformative group project and Simon et al.’s (2008) use of instruction and writing exercises both failed to alter engineering students’ incremental and entity beliefs about the nature of intelligence. However, contrary to Simon et al., 2008, Cutts et al., 2010 found that, when paired with incremental belief-oriented feedback from an instructor on assignments, teaching programming students that intelligence is a malleable trait was associated with increases in students’ incremental intelligence beliefs across a two-semester, 26-week course and performance on a final course exam.

Although research in this area for STEM students is limited, two important findings emerged from these studies. First, Reid and Ferguson, 2014, Simon et al., 2008 both investigated the intelligence beliefs of engineering students, while Cutts et al. (2010) studied programming students. The findings of these studies suggest that interventions designed to influence STEM students’ intelligence beliefs may be more effective for CS-oriented courses than engineering. Second, more rigorous, prolonged interventions may be needed to elicit changes in STEM students’ implicit intelligence beliefs. While Reid and Ferguson (2014) administered a smaller scaled intervention (i.e., a brief lecture and two writing exercises) than Simon et al. (2008) (i.e., a semester long group project) and Cutts et al. (2010) (i.e., various combinations of three interventions), only Cutts and associates spread their intervention across multiple semesters. As a result, the significant findings of their study may reflect the need for more rigorous, time-intensive interventions to increase STEM students’ implicit intelligence beliefs.

Research has also demonstrated shifts in the implicit intelligence beliefs of students in STEM-related courses in the absence of interventions. For example, Dai and Cromley (2014) conducted a year-long study of 335 undergraduate biology students to explore how their implicit intelligence beliefs about biology ability changed across time and how these changes related to retention in the biology major. They found entity beliefs to increase and incremental belief to decrease across the year. In addition, increase in entity belief did not significantly predict retention in the biology major, but decline of incremental belief predicted lower retention. Additionally, achievement in an introductory biology course was found to positively predict retention in the biology major. Relatedly, Shively and Ryan (2013) investigated the stability of undergraduate mathematics students’ implicit intelligence beliefs across one semester. In this study, students in a college algebra course completed self-report questionnaires targeting implicit intelligence beliefs at the beginning and end of the semester. There was a significant decrease in incremental belief and an increase in entity belief across the semester, increase in entity belief associated with lower course grades. Participants also reported a greater incremental belief for general intelligence than for mathematics intelligence, a finding which suggests that the participants held domain specific beliefs about mathematics intelligence and viewed mathematics intelligence to be less malleable over time than general intelligence. In a similar year-long longitudinal study of fifth and sixth grade students, Gonida et al. (2006) found that students who experienced high levels of achievement (as measured by course grades) decreased in their belief that intelligence is a fixed, unalterable entity, while students who experienced moderate and low levels of achievement did not experience a significant decrease in their entity belief across the school year. Similarly, students who had experienced high levels of achievement by the first data collection point more strongly endorsed the incremental belief at the end of the school-year.

Collectively, the reviewed studies suggest that students’ implicit beliefs of intelligence are not stable entities. Rather, implicit beliefs can change with and without the facilitation of an intervention. In contrast to the mixed findings of intervention-based studies, investigations of change across time without the presence of an intervention present a more unified picture of shifts in implicit intelligence beliefs over time. This suggests that factors other than classroom interventions (e.g., development, motivation and self-regulation, classroom experiences) might impact students’ implicit beliefs of intelligence and cause them to shift.

Section snippets

The present study

Cumulatively, the findings related to changes in students’ implicit intelligence beliefs with or without the presence of interventions hold two aspects of significance. First, they contrast with the traditional implicit intelligence belief literature which posits that implicit beliefs are stable across time (e.g., Dweck and Molden, 2005, Robins and Pals, 2002) by establishing the dynamic, changing nature of implicit beliefs. Second, they have documented that changes in implicit intelligence

Participants

Participants in this study were 443 undergraduate students (380 males; 63 females) from a suite of CS1 courses at a large Midwestern university. Participants included 218 freshmen, 132 sophomores, 48 juniors, 36 seniors, and 9 who identified as “other.” Data were collected in the fall 2012 semester and the spring and fall semesters of 2013 from the suite of CS1 courses described in Section 2.2: CS1-Computer Science: (n = 68) taught in fall 2012 and spring 2013; CS1-Engineering (n = 205) taught in

Results and discussion

In order to address our research questions, we explored (a) how CS1 students’ implicit intelligence beliefs changed across the course of the semester, (b) the factors that potentially influence changes in students’ implicit intelligence beliefs (i.e., student profile and course enrollment), and (c) how changes in implicit intelligence beliefs could be used to predict course achievement, as measured by standardized course grades and computational thinking knowledge-test scores. Changes in sample

Scholarly significance

Results of the present study provide additional evidence for the dynamic, changing nature of implicit intelligence beliefs across time found in recent studies (Dai and Cromley, 2014, Reid and Ferguson, 2014, Shively and Ryan, 2013). Like prior studies, we found that students’ incremental beliefs about intelligence decreased and their entity intelligence beliefs increased across the semester. Although change for the entire student sample was significant, the effect sizes for these shifts were

Acknowledgments

This material is based upon work supported by the National Science Foundation under Grants No. 1122956 and No. 1431874. This work is based on an earlier work: “Exploring changes in computer science students’ implicit beliefs of intelligence across the semester,” in Proceedings of the 11th ACM International Computing Education Research Conference (ICER’15) (pp. 161-168), August 9-13, 2015, Omaha, NE ©ACM, 2015. DOI: http://dx.doi.org/10.1145/2787622.2787722.

References (78)

  • J. Allen et al.

    Third-year college retention and transfer: Effects of academic performance, motivation, and social connectedness

    Research in Higher Education

    (2008)
  • A. Bandura

    Social foundations of thought and action: A social cognitive theory

    (1986)
  • L.S. Blackwell et al.

    Implicit theories of intelligence predict achievement across an adolescent transition: A longitudinal study and an intervention

    Child Development

    (2007)
  • M. Boekaerts et al.

    How far have we moved toward the integration of theory and practice in self-regulation?

    Educational Psychology Review

    (2006)
  • P.B. Carr et al.

    Intelligence and motivation

  • A.M.M. Conley

    Patterns of motivation beliefs: Combining achievement goal and expectancy-value perspectives

    Journal of Educational Psychology

    (2012)
  • M.V. Covington

    Self-worth theory goes to college or do our motivation theories motivate

  • M. Covington

    Self-worth theory: Reflections and prospects

  • J.W. Creswell et al.

    Designing and conducting mixed methods research

    (2011)
  • Q. Cutts et al.

    Manipulating mindset to positively influence introductory programming performance

  • M.S. DeBerard et al.

    Predictors of academic achievement and retention among college freshmen: A longitudinal study

    College Student Journal

    (2004)
  • A.L. Duckworth et al.

    Self-regulation and school success

  • C.S. Dweck

    Self-theories: Their role in motivation, personality, and development

    (2000)
  • C.S. Dweck

    Boosting achievement with messages that motivate

    Education Canada

    (2007)
  • C.S. Dweck et al.

    A social-cognitive approach to motivation and personality

    Psychological Review

    (1988)
  • C.S. Dweck et al.

    Self-theories: Their impact on competence motivation and acquisition

  • A.J. Elliot et al.

    A 3×2 achievement goal model

    Journal of Educational Psychology

    (2011)
  • N. Entwistle et al.

    The conceptual bases of study strategy inventories

    Educational Psychology Review

    (2004)
  • J.A. Gasiewski et al.

    From gatekeeping to engagement: A multicontextual, mixed method study of student academic engagement in introductory STEM courses

    Research in Higher Education

    (2012)
  • A. Ghasemi et al.

    Normality tests for statistical analysis: A guide for non-statisticians

    International Journal of Endocrinology and Metabolism

    (2012)
  • E. Gonida et al.

    Implicit theories of intelligence, perceived academic competence, and school achievement: Testing alternative models

    The American Journal of Psychology

    (2006)
  • J.T. Guthrie et al.

    Profiles of reading motivation among African American and caucasian students

    Journal of Literacy Research

    (2009)
  • M.P. Hazley et al.

    Changes in student goal orientation across the semester in undergraduate computer science courses

  • Y.Y. Hong et al.

    Implicit theories, attributions, and coping: A meaning system approach

    Journal of Personality and Social Psychology

    (1999)
  • P. Hsieh et al.

    A closer look at college students: Self-efficacy and goal orientation

    Journal of Advanced Academics

    (2007)
  • J. Husman et al.

    The intersection of students’ perceptions of instrumentality, self-efficacy, and goal orientations in an online mathematics course

    Zeitschrift für Pädagogische Psychologie

    (2007)
  • J. Husman et al.

    Validating measures of future time perspective for engineering students: Steps towards improving education

  • G. Keppel et al.

    Design and analysis: A researcher’s handbook

    (2004)
  • M. Komarraju et al.

    Increased career self-efficacy predicts college students’ motivation, and course and major satisfaction

    Journal of Career Assessment

    (2014)
  • Cited by (43)

    • Trajectories of motivation and their academic correlates over the first year of college

      2020, Contemporary Educational Psychology
      Citation Excerpt :

      However, such distal measurements can miss more nuanced fluctuations that likely occur within semesters as students navigate coursework in new and often unfamiliar domains. Recent studies across different theoretical traditions have documented motivational change over the smaller time scale of a single semester and provided critical insights into motivational malleability (Brahm, Jenert, & Wagner, 2017; Dai & Cromley, 2014; Flanigan et al., 2017; Kosovich et al., 2017; Robinson, Lee, et al., 2019; Young, Wendel, Esson, & Plank, 2018; Zusho & Pintrich, 2003). For example, Dai and Cromley (2014) found that students increasingly endorsed a fixed view of ability while declining in their endorsement of a malleable view of ability over the course of an introductory biology class.

    View all citing articles on Scopus
    View full text