Self-efficacy and its role in computing education

Wednesday 23rd June 2021 | Blog entry: 2

A person’s belief in their ability to influence events that affect their lives is known as “self-efficacy” (e.g., Bandura, 1977). These beliefs are fundamental to our well-being, motivations and accomplishments, because unless people believe that their actions can have their preferred outcome, then there is little incentive to persevere in the face of obstacles. It is now accepted that self-efficacy affects almost every area of our lives (Pajares, 2002).

It is the interpretation of one’s own ability and what you expect to happen, that is crucial when considering self-efficacy, rather than actual ability. There are many examples of successful people who have achieved great things, but who despite this, have interpreted their abilities differently (e.g., Pajares, 2002).

"I have written 11 books, but each time I think, 'uh oh, they’re going to find out now. I’ve run a game on everybody, and they’re going to find me out.'”

Maya Angelou

According to Bandura (1977), self-efficacy is influenced in four ways; 1) considering a person’s own performance. For example, having successfully written some lines of code, a young person may have greater self-efficacy to write some more. Pajares (2002) gives the wonderful example, adapted here for a computing context, of two students with the same exam result, perhaps a grade 6 (broadly equivalent to a grade B prior to 2020) in their GCSE computer science exam. One of the students who normally gets a grade 9 (equivalent to an A*) and who worked very hard for their exam, may view their grade differently to another student who worked equally as hard but normally gets a grade 4 (equivalent to a grade C). The student who normally gets the higher grades will perhaps feel more disheartened and the student who normally achieves a grade 4 may feel boosted. 2) The “if they can do it, then I can too” attitude. For example, seeing peers also write lines of code successfully may increase the self-efficacy of the observer. 3) A learner may be motivated by “social persuasion”. This could be from a teacher, parent or peer saying they believe the learner in question can do the task may also increase self-efficacy. 4) The emotional response to a task or event, for example, a young person sitting an exam which involves writing some lines of code, may have reduced levels of self-efficacy due to high levels of anxiety around the outcome of the task. They may “learn to interpret their physiological arousal as an indicator or personal competence…” (Usher and Pajares, 2009).

A person’s self-efficacy may be very different for different tasks—how a person feels about their skills in a particular area is not necessarily how they feel about their skills in all areas. A learner may therefore feel high self-efficacy in one area of computing, say information security, but lower self-efficacy in another area, say computer systems. They may also have a lower self-efficacy about getting their desired grade in a computer science exam but high self-efficacy in the delivery of a computer science project or coursework (Betz and Hackett, 2006; Lent and Brown, 2006).

It is important to note the difference between self-efficacy, and “growth mindset” and “self-esteem”. Self-esteem refers to a person’s sense of self-worth, a high self-efficacy could increase an individual’s sense of self-worth. On the other hand, a “growth mindset” is the belief that the ability to do something is a malleable quality that can be developed through effort, practice and resilience and is not a fixed trait. It differs from self-efficacy in that the belief is on the individual’s capacity to change and “grow” rather than whether an individual believes they have the capability to achieve a specific task, such as completing a computer science course (e.g., Dweck, 2017).

Self-efficacy can also sometimes be confused with two other very similar ideas: “outcome expectancy” and “self-concept”. Outcome expectancy is the degree to which someone believes that a particular outcome will occur, whereas self-efficacy is the belief someone has in their capacity to influence an outcome (e.g., Landry 2003). For example, consider the following scenario: A school student, Sam, did well in her computing mock exam. Once home, Sam’s parents asked how the exam went. Sam said she felt confident and believed she had all the knowledge and skills to go into the actual computing exam (she has high self- efficacy). She thinks that she will do well in her exam (she has high outcome-expectancy).

Self-concept is an individual’s evaluation of themselves. Using the example above, if Sam considers herself to be good at computing, then she has a high self-concept in computing.

Self-efficacy in computing

In computing, self-efficacy is a person’s belief about their ability to use a computer proficiently. The level of self-efficacy expectations, or the degree to which a person believes they can do something, will influence whether that person tries to do something or not. In secondary education, this may, in part, determine whether a young person chooses a particular subject at GCSE or post-16, and which could ultimately lead to imbalances in the skills of young people entering the workforce (e.g., Bandura, 1999; Betz, 2000). Several studies have demonstrated that a higher computing self-efficacy is associated with greater enjoyment and more frequent use of computers (e.g., Compeau and Higgins, 1995; Hill, Smith and Mann, 1987). And previous positive computing experience, through quality teaching and training, rather than duration of the training, is associated with increased computing self-efficacy (e.g., Torkzadeh and Koufteros, 1994; Ertmer et al., 1994; Cassidy and Eachus, 2000). The Raspberry Pi Foundation have also recently explored self-efficacy in a pilot study of 15 young people and found a mismatch between computing and interviewees’ own identities, and low self-indicated self-efficacy (Sentance, 2021).

Self-efficacy and gender in computing

Lips and Temple (1990) suggest that girls' computing and mathematical self-efficacy is correlated to participation in computer science. It has also been recognised that gender stereotypes negatively influence self-efficacy of girls in STEM, of which there are significant stereotypes surrounding computing (e.g., McGuire et al., 2020). And as discussed above, computing self-efficacy is affected by previous computing experience and performance. Therefore, as women often have less programming experience (e.g., He and Freeman 2010) and may show interest in computer science at an older age than men (Lang, 2010), it is perhaps unsurprising that many studies, but by no means all, have demonstrated a gender difference in self-efficacy relating to computing, in particular those tasks involving more advanced computing skills (e.g., Torkzadeh and Koufteros, 1994; Cassidy and Eachus, 2002; Huang, 2013).

Increasing self-efficacy in the computing classroom

As discussed above, self-efficacy can be influenced by four factors: a person’s own performance, the performance of peers, social persuasion and the emotional response to the activity. By addressing these four factors it may be possible to increase levels of self-efficacy amongst learners in the classroom. The following strategies are adapted from Margolis and McCabe (2006).

Developing a sense of achievement

Tasks should not be overly simple but should provide “achievable challenge” that doesn’t provoke fear or embarrassment. There should be a clear roadmap through the computing curriculum, with frequent, small, clear and achievable goals, where learners can confidently complete one goal before moving on to the next. This may require flexibility in terms of the time allocated for the task. Learners should be able to track their own progress and see evidence of their own success in the subject.

Capitalising on the influence of peers

Providing opportunities for learners to observe others who are working through a similar task or activity or who are demonstrating an effective strategy, may prove beneficial to developing self-efficacy. Draw attention to the learner’s good practice, particularly when the learner attributes their strategy to something that they can control, for example, by trying out a new method of doing something, or practicing in their own time. Ideally, peers should be like learners in ways that the learners deem important – this could be age, ability, gender, race, interests etc. It is also worth ensuring that the contexts given are directly relevant and relatable to the learner. The Science Capital Teaching Approach gives some examples of how this can be achieved (Godec et al., 2017).

Communicating persuasively

Provide frequent and specific feedback that reflects the factors determining the learner’s performance (e.g., trying something new, working collaboratively etc.). An example could be “You coded that correctly: You read the instructions carefully, asked yourself what it was about, and selected the correct code and inserted it in the right location. Because you did this, you made it work. Well done.”. Educators, for instance, can remind students of their recent, similar successes in related subject areas or tasks.

Managing emotion

Creating a classroom environment where getting things wrong is accepted as part of the learning process is ideal. Strategies such as the use of mini-white boards that can quickly erase errors or anonymous online question boards may also serve as a way of tackling anxiety and building confidence. Have discussions with learners about coping in “high stakes” exam situations, acknowledge the worry and continuously support learners in their development of coping strategies and provide ongoing encouragement to navigate setbacks. Setting a challenging test at the start of the course where learners perform badly is likely to increase anxiety and reduce self-efficacy later in the course.

Self-efficacy and the SCARI Computing Research Project

Examining the self-efficacy of learners in relation to computing in school, especially attitudes towards coding, will be a considerable focus of the SCARI Computing research project. Learner self-efficacy in computing will be obtained through a comprehensive survey of secondary-age students in more than 20 co-educational state schools across England during the Summer and Autumn terms of 2021. Self-efficacy will be measured through a series of questions, validated by the previous work of Vandenberg et al., (2021). The student survey is not limited to self-efficacy in computing, but also the influence of family, the computing curriculum, ability sets, and stereotypes, as well as students’ thoughts around subject choice, jobs and the relevance of computing in their lives, amongst other things. It is hoped that through such a comprehensive survey, we will identify the barriers that exist for young people in computing, as well as begin to unpick the characteristics and motivations of those that do choose to pursue computing as a subject option.

References

Bandura, A., 1977. Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191–215.

Bandura, A., 1999. Social cognitive theory: An agentic perspective. Asian Journal of Social Psychology, 2(1), 21-41.

Betz, N. E., and Hackett, G., 1981. The relationship of career-related self-efficacy expectations to perceived career options in college men and women. Journal of Counselling Psychology, 28(5), 399-410.

Betz N.E., 2000. Self-efficacy theory as a basis for career assessment. Journal of Career Assessment, 8 pp. 205-222.

Cassidy, S., and Eachus, P., 2000. The Role of Computer Training in Determining Levels of Computer Self-Efficacy in Students. Proceedings of Computers in Psychology (CiP), University of York.

Cassidy S., and Eachus, p., 2002. Developing the Computer User Self-Efficacy (Cuse) Scale: Investigating the Relationship between Computer Self-Efficacy, Gender and Experience with Computers. Journal of Computing Research. Vol 26 (2).

Compeau, D. R., and Higgins, C. A., 1995. Computer self-efficacy: Development of a measure and initial test. MIS Quarterly, 95(19), 189-211.

Dweck, C.S. 2017. Mindset. Robinson, 6^th Ed.

Esher E.L., Pajares F., 2009. Sources of self-efficacy in mathematics: a validation study. Contemporary Educational Psychology, 34, pp. 89-101.

Godec, S., King, H. and Archer, L., 2017. The Science Capital Teaching Approach: engaging students with science, promoting social justice. London: University College London

He, J., and Freeman, L., A., 2010. Are men more technology-oriented than women? The role of gender on the development of general computer self-efficacy of college students. Journal of Information Systems Education, 21, 203–212

Hill, T., Smith, N.D., and Mann, M.F., 1987. Role of Efficacy Expectations in Predicting the Decision to Use Advanced Technologies: The Case of Computers. Journal of Applied Psychology, 72:2, pp. 307-313.

Huang, C., 2012. Gender differences in academic self-efficacy: a meta-analysis. European Journal of Psychology of Education volume 28, pages1–35.

Landry, C. C., 2003. Self-efficacy, motivation, and outcome expectation correlates of college students' intention certainty. LSU Doctoral Dissertations. 1254

Lang, C., 2010. Happenstance and compromise: A gendered analysis of students’ computing degree course selection. Computer Science Education, 20, 317–345.

Lent, R.W., and Brown, S.D., 2006. On conceptualising and assessing social cognitive constructs in career research: A measurement guide. Journal of Career Assessment. Vol. 14 (1).

Lips, H. M., and Temple, L., 1990. Majoring in computer science: Causal models for women and men. Research in Higher Education, 31, 99–113.

Margolis, H., and McCabe, P., P., 2006. Improving self-efficacy and motivation what to do, what to say. Intervention in school and clinic, 41(4), 218-227.

McGuire, L., Mulvey, K.L., Goff, E., Irvin, M. J., Winterbottom, M., Fields, G. E., Hartstone-Rose, A., and Rutland, A., 2020. STEM gender stereotypes from early childhood through adolescence at informal science centers. Journal of Applied Development Psychology, 67:101109.

Pajares, F., 2002. Overview of social cognitive theory and of self-efficacy. Retrieved 1/6/2021 from http://www.emory.edu/EDUCATION/mfp/eff.html

Sentance, S., 2021. The digital divide: interactions between socioeconomic disadvantage and computing education. Raspberry Pi Foundation. The digital divide: interactions between socioeconomic disadvantage and computing education - Raspberry Pi. Accessed 28/06/21.

Torkzadeh, G., and Koufteros, X., 1994. Factorial Validity of a Computer Self-Efficacy Scale and the Impact of Computer Training. Education and Psychological Measurement, 54:3, pp. 813-821.

Vandenberg, J., Rachmatullah, A., Lynch, C., Boyer, K. E., and Wiebe, E., 2021. Interaction effects of race and gender in elementary CS attitudes: A validation and cross- sectional study. International Journal of Child-Computer Interaction. 29.