Using The General Self-Efficacy Scale to Measure Engineering Students' Self-Efficacy in a Project Management Course

Research Paper Recommended citation: Markman, A. O., Christiansen, S. H., Fornø, A. A., & Al-Subaihi, M. (2025). Using The General Self-Efficacy Scale to Measure Engineering Students’ Self-Efficacy in a Project Management Course. In Kangaslampi, R., Langie, G., Järvinen, H.-M., & Nagy, B. (Eds.), SEFI 53rd Annual Conference. European Society for Engineering Education (SEFI), Tampere, Finland. DOI: 10.5281/zenodo.17631567. This Conference Paper is brought to you for open access by the 53rd Annual Conference of the European Society for Engineering Education (SEFI) at Tampere University in Tampere, Finland. This work is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 4.0 International License. USING THE GENERAL SELF-EFFICACY SCALE TO MEASURE ENGINEERING STUDENTS’ SELF-EFFICACY IN A PROJECT MANAGEMENT COURSE A. O. Markmana, 1 , S. H. Christiansenb, A. A. Fornøc, M. Al-Subaihid a Aalborg University, Aalborg, Denmark, https://orcid.org/0009-0002-5183-7241 b Aalborg University, Aalborg, Denmark, https://orcid.org/0000-0002-1329-9836 c Technical University of Denmark, Ballerup, Denmark, https://orcid.org/0009-00041445-3622 d Technical University of Denmark, Ballerup, Denmark, https://orcid.org/0000-00021941-5201 Conference Key Areas: 8. Continuing education and life-long learning in engineering & 10. Engineering skills, professional skills, and transversal skills Keywords: General self-efficacy scale, self-efficacy, project management, engineering education ABSTRACT An important aspect of engineering education is the crucial role it plays in preparing engineering students for the complexities of professional life, especially in relation to the development of self-efficacy. This study examines engineering students’ selfefficacy development over the timespan of a semester in a project management course with focus on the impact of study-related employment and gender differences. Using the General Self-Efficacy (GSE) scale, data was collected from 103 students at the beginning of the course and 46 students at the end. While the results indicate no significant overall change in self-efficacy throughout the semester, students with study-related jobs reported consistently higher self-efficacy scores, particularly in handling unexpected challenges. Additionally, male students scored higher than their female counterparts. These results suggest that real-world work experience play a crucial role in self-efficacy development for engineering students. Future research should incorporate qualitative methods to further explore the nuances of self-efficacy growth in engineering education. 1 Corresponding Author A. O. Markman [email protected] 1 INTRODUCTION Motivating a diverse group of students academically can be challenging, and it is often hindered by the use of ineffective strategies or approaches that only resonate with a subset of the group (Anyichie & Butler, 2023). Students who lack confidence in their abilities are less likely to engage with or apply a given skill (Carberry et al., 2018). Therefore, understanding the principles that enhance self-efficacy is crucial for educators (Ponton et al., 2001), as self-efficacy has been linked to improved learning outcomes, higher academic performance, increased motivation, persistence, resilience, help-seeking behaviour, self-management, and retention in higher education programs (Mamaril et al., 2016; Power et al., 2024). Self-efficacy refers to an individual’s belief in their ability to successfully perform a task in a given context, particularly in relation to their perceived strengths and limitations (Waddington, 2023). It is generally seen as a stronger predictor of an individual’s task performance than other concepts such as self-confidence or self-esteem (Heslin & Klehe, 2006). Previous research show that self-efficacy beliefs could significantly affect academic achievement and the persistence in the field of engineering (e.g., Basith et al., 2020; Fitzpatrick et al., 2023). Bandura (1982; 1994) presents four main sources of influence on individual’s self-efficacy. These include: 1) mastery experience (successfully accomplishing tasks), 2) vicarious experiences (observing similar others succeed), 3) social persuasions (receiving encouragement to believe in one’s abilities), and 4) physiological and emotional states (interpreting one’s capabilities based on emotional and physical responses). Research has shown that these sources significantly influence students’ self-efficacy beliefs and interest in engineering (e.g., Concannon et al., 2018). The ways in which self-efficacy has been measured in engineering have varied considerably, e.g., general academic self-efficacy measures, domain-general selfefficacy measures, and self-efficacy measures for specific engineering tasks or skills (Mamaril et al., 2016; Venugopal et al., 2020). General academic self-efficacy measures broadly assess engineering students’ beliefs in their capabilities to perform academically (Mamaril et al., 2016). General self-efficacy (GSE) involves a more trait-like generality dimension of self-efficacy, defined as one’s belief in one’s overall competence to effect requisite performances across a wide variety of achievement situations, capturing differences among individuals in their tendency to view themselves as capable of meeting task demands in a broad array of contexts (Chen et al., 2001). GSE reflects a generalization across various domains of functioning in which people judge how efficacious they are (Luszczynska & Schwarzer, 2005). The GSE scale has previously been used to measure gender differences, where females have consistently reported a lower general self-efficacy level than males (Nielsen et al., 2018; Wang et al., 2019). However, studies considering the role of engineering students’ employment in shaping self-efficacy beliefs are limited. This paper examines survey results from a project management course for engineering students at a Scandinavian technical university, using the GSE scale. During the course, the students were exposed to various activities that were based on Bandura’s (1982; 1994) main sources of influence on self-efficacy. The GSE scale was used in the beginning of the course (n = 103) and end of the course (n = 46). The study explores the following research questions: How can the GSE scale be used to assess students’ self-efficacy in a course? Are there differences in students’ self-efficacy based on factors such as gender or students’ having a student job? 2 METHODOLOGY To establish a baseline and measure the impact of the course, students responded to the survey twice, once at the beginning of the semester and again at the end. Respondents were required to indicate the extent to which each statement applies to them on a Likert scale (Schwarzer & Jerusalem, 2010). The GSE scale includes 10 items, or statements (Table 1), e.g., “I can always manage to solve difficult problems if I try hard enough”, with the possible responses: 1) Not at all true, 2) Hardly true, 3) Moderately true, and 4) Exactly true (Luszczynska & Schwarzer, 2005). The GSE scale assesses the strength of an individual’s belief in his or her own ability to respond to novel or difficult situations and to deal with any associated obstacles or setbacks (Schwarzer & Jerusalem, 2010). The GSE scale has been used for numerous research projects, where it typically yielded internal consistencies between α (alpha) = 0.75-0.91 (Scholz et al., 2002). To get a reliable score for comparison, Romppel et al.’s (2013) study was used. They assessed the item characteristics of the original scale with participants (n = 19,717) from 26 different countries (Table 1). The survey distributed to the students in this study incorporated the 10 items from the GSE scale alongside three context-specific questions regarding age, gender, and student employment. The study was conducted in language of the country the study was conducted in; the translation was found on the official GSE scale website (Schwarzer, 2012). The software IBM SPSS was used for the data analysis. 2.1 Sample The study participants were sixth and seventh semester bachelor’s of engineering students, with 103 students completing the initial survey (men = 54%, women = 46%) and 46 students responding to the survey at the end of the semester (men = 52%, women = 48%). The sample represented students from nine different engineering programs. The study was conducted within a project management course designed for bachelor’s of engineering students. The course focused on designing and managing projects while equipping students with skills to identify and address challenges as they arise. The course consisted of 13 lessons focused on developing students’ project management competencies, with additional emphasis on time management and the concept of self-efficacy. In particular, the curriculum was closely aligned with Bandura’s (1982; 1994) framework on the sources of selfefficacy. For example, vicarious experiences were fostered through the inclusion of role models and guest lectures by alumni, offering students relatable examples of success. Time management instruction was designed to equip students with practical strategies for balancing multiple tasks, thereby creating opportunities for mastery experiences and reinforcing their sense of self-efficacy. It was mandatory for one study program (13% of participants), while 87% enrolled as an elective. 53% of students in the initial survey reported having or having had a part-time job while studying. By the end of the semester, this proportion of students was 65% in the second survey. 3 RESULTS The internal consistency for the total sample in September 2024 (n = 103) was α = 0.81, indicating a strong level of reliability for the GSE scale. Similarly, the internal consistency for the total sample in December 2024 (n = 46) was 0.85. Table 2 presents a comparison of the means of the GSE scale in lecture 1 (September 2024) and lecture 13 (December 2024), as well as a third sample (n = 19,719) for reference (Romppel et al., 2013). Table 1. Comparison of mean results. Items (Q) Mean of the GSE scale in Romppel et al., 2013 (n = 19,719) Mean of the GSE scale in September (n = 103) Mean of the GSE scale in December (n = 46) Q1. I can always manage to solve difficult problems if I try hard enough 3.14 3.3 3.3 Q2. If someone opposes me, I can find means and ways to get what I want 2.96 3.0 3.0 Q3. It is easy for me to stick to my aims and accomplish my goals 2.80 3.1 3.1 Q4. I am confident that I could deal efficiently with unexpected events 2.85 3.1 3.0 Q5. Thanks to my resourcefulness I know how to handle unforeseen situations 2.90 3.1 3.2 Q6. I can solve most problems if I invest the necessary effort 3.03 3.5 3.4 Q7. I can remain calm when facing difficulties because I can rely on my coping abilities 2.95 3.2 3.1 Q8. When I am confronted with a problem, I can usually find several solutions 2.98 3.1 2.9 Q9. If I am in trouble I can usually think of something to do 3.01 3.3 3.2 Q10. No matter what comes my way I’m usually able to handle it 2.97 3.4 3.2 As shown in Table 2, while differences between the two time points (September and December) were minimal, students generally scored higher on the GSE scale compared to the reference sample of 19,719 individuals. A paired-samples t-test was conducted to assess changes over time, and the results showed no statistically significant differences (p > 0.05 for all items). The analysis instead focuses on two factors: gender and study-related job status. First, the results examining gender differences in students’ responses are presented. Second, the analysis explores differences based on whether students had a study-related job or not. Due to the limited number of participants, a t-test combining both gender and job status could not be conducted. 3.1 Students’ gender The self-efficacy scores for male and female students were compared across all 10 survey items (see Table 2 for descriptive statistics). Table 2. Gender – descriptive statistics for the two surveys (September = Sep, n = 103; December = Dec, n = 46). Men Women Out of the 103 students who responded to the first survey in September, 46% identified as women and 54% as men. In general, male students reported slightly higher self-efficacy than female students across the majority of the items, with the largest differences observed in Q4 (Men: M = 3.29, SD = 0.594; Women: M = 2.82, SD = 0.747) and Q7 (Men: M = 3.36, SD = 0.70; Women: M = 2.96, SD = 0.85). On the other hand, for Q1, Q2, and Q9, the mean scores were almost identical for both genders, suggesting minimal gender differences in these areas. Notably, Q2 showed identical mean values (M = 3.02) for both genders, though women exhibited slightly higher variability (SD = 0.81) compared to men (SD = 0.75). The standard deviations (SD) for women’s responses were generally higher, indicating a greater variation in self-efficacy perceptions among female students. This was particularly evident in Q7 (SD = 0.85) and Q8 (SD = 0.81). An independent samples t-test was performed to compare the self-efficacy scores between male and female students. The results indicated that while most survey items showed no significant difference (p > 0.05), two items did exhibit statistically significant differences. Q4 showed a highly significant result (t(99) = 3.472, p. < 0.001), indicating that male students reported significantly higher self-efficacy in this area. Similarly, Q7 also showed a significant difference (t(99) = 2.604, p = 0.011), again with male students reporting higher selfefficacy. Finally, Q8 approached significance (t(99) = 1.741, p = 0.085), suggesting a marginally higher self-efficacy score for male students. Mean Std. deviation Mean Std. deviation Independent Samples t-test Significance (p) Q1 Sep 3.25 0.667 3.31 0.596 0.633 Dec 3.13 0.680 3.45 0.596 0.089 Q2 Sep 3.02 0.751 3.02 0.812 0.978 Dec 2.88 0.537 3.09 0.610 0.208 Q3 Sep 3.02 0.726 3.16 0.737 0.349 Dec 3.08 0.717 3.14 0.774 0.811 Q4 Sep 3.29 0.594 2.82 0.747 <0.001 Dec 3.17 0.565 2.91 0.868 0.235 Q5 Sep 3.20 0.616 3.04 0.673 0.240 Dec 3.21 0.509 3.18 0.733 0.887 Q6 Sep 3.54 0.571 3.40 0.780 0.316 Dec 3.29 0.464 3.45 0.596 0.305 Q7 Sep 3.36 0.699 2.96 0.852 0.011 Dec 3.13 0.680 3.00 0.756 0.558 Q8 Sep 3.18 0.606 2.93 0.809 0.085 Dec 2.96 0.624 2.91 0.811 0.818 Q9 Sep 3.27 0.587 3.24 0.645 0.849 Dec 3.21 0.588 3.27 0.631 0.722 Q10 Sep 3.46 0.571 3.27 0.688 0.118 Dec 3.29 0.550 3.18 0.853 0.603 In the second survey conducted in December, 52% were male students, while 48% were female students. An observation is that male students’ self-efficacy appears to slightly decline over the course of the semester, whereas female students’ selfefficacy remains relatively stable or increases in some areas. For instance, in Q1, men’s mean self-efficacy score decreases from 3.25 in September to 3.13 in December, while women’s score increases from 3.31 to 3.45. A paired t-test was conducted to assess gender differences over time. For male students in Q7, there was a statistically significant increase (t(11) = 2.244, p = 0.046). Additionally, several borderline significant changes were observed for Q6 (t(11) = 2.159, p = 0.054), Q4 (t(11) = 1.915, p = 0.082), and Q10 (t(11) = 1.820, p = 0.096). However, the t-test for female students showed no significant differences from September to December. 3.2 Students with or without student jobs The self-efficacy scores for students with or without a student job were compared across all 10 survey questions (see Table 3 for descriptive statistics). Table 3. Students with or without a student job – descriptive statistics for the two surveys (September = Sep, n = 103; December = Dec, n = 46). Student job No student job Mean Std. deviation Mean Std. deviation Independent Samples ttest Significance (p) Q1 Sep 3.30 0.638 3.25 0.636 0.684 Dec 3.29 0.693 3.27 0.594 0.910 Q2 Sep 3.09 0.766 2.94 0.783 0.312 Dec 3.03 0.547 2.87 0.640 0.367 Q3 Sep 3.11 0.725 3.04 0.743 0.625 Dec 3.06 0.727 3.20 0.775 0.565 Q4 Sep 3.19 0.652 2.96 0.743 0.100 Dec 3.06 0.772 3.00 0.655 0.782 Q5 Sep 3.25 0.617 3.00 0.652 0.055 Dec 3.26 0.514 3.07 0.799 0.331 Q6 Sep 3.57 0.636 3.38 0.703 0.155 Dec 3.42 0.502 3.27 0.594 0.367 Q7 Sep 3.23 0.800 3.13 0.789 0.523 Dec 3.03 0.752 3.13 0.640 0.657 Q8 Sep 3.06 0.745 3.08 0.679 0.851 Dec 2.84 0.638 3.13 0.834 0.191 Q9 Sep 3.38 0.596 3.13 0.606 0.037 Dec 3.19 0.601 3.33 0.617 0.467 Q10 Sep 3.45 0.574 3.29 0.683 0.201 Dec 3.26 0.729 3.20 0.676 0.797 In the survey conducted in September, a total of 51% of students reported having a student job, while 49% reported not having one. Across the majority of the items (Q1, Q2, Q3, Q4, Q5, Q6, Q7, Q9, & Q10), students with a study job reported higher selfefficacy mean scores compared to those without a student job. For example, in Q5, the mean score for those with a student job is 3.25, compared to 3.00 for those without. Similarly, in Q6, individuals with a student job score 3.57, while those without score 3.38 on average. The variability in responses (as reflected by the standard deviations) is similar for both groups across most items, indicating that the degree of spread in responses is comparable between individuals with and without a student job. The largest standard deviations are observed in Q7, where both groups show some variation in their responses, but no significant difference between the groups in terms of variability. In general, the results revealed that most items did not show significant differences between individuals with and without a student job. However, one item demonstrated statistically significant differences. Q9 showed a significant result (t(99) = 2.109, p = 0.037), suggesting that students with a job scored higher than those without, while Q5 approached significance (t(99) = 1.941, p = 0.055). In the survey conducted in December, a total of 31 students reported having a student job, while 17 reported not having one. Compared to the survey conducted in September, the mean differences are generally small across all items, with some items showing slightly higher means for students with student jobs. Similarly, there are no statistically significant differences between students with or without student jobs in the survey conducted in December. A paired t-test were applied to assess these differences over time. All the paired comparisons between September and December show no statistically significant differences, indicating that, on average, there were no changes in the responses between the two months across the 10 items. 4 DISCUSSION AND CONCLUSIONS The study has several indications of the potentials of applying the GSE scale as part of assessments and evaluations in engineering courses with a project management focus. This includes providing targeted support as lower GSE scores might highlight a need for additional mentoring or scaffolding for certain student groups. Additionally, changes in GSE scores across a course could indicate growth in students’ perceived ability to e.g., handle challenges, emphasizing the possibility of tracking students’ development, These findings must be interpreted with caution, acknowledging both the limited participant pool and the context-specific nature of engineering-related study programs. However, results derived from applying the GSE scale are relevant to the requirements and conditions of the specific situation (Debapriyo et al., 2018). It is important to recognize that these students do not represent a universal sample, nor can their experiences be easily generalized to other disciplines. Although the study suggests the potential of using the GSE scale in engineering education, the results did not indicate a significant shift in students’ self-efficacy over the course. This raises questions about which factors, such as course content, personal experiences, or individual motivation levels, may contribute to changes in self-efficacy, even if they were not explicitly reflected in the data. Further research is needed to explore how these elements interact and whether specific aspects of the course contribute to shifts in self-efficacy. Elements such as vicarious experiences, where students observing teammates solving problems or managing tasks effectively, could model similar beliefs in their own abilities in combination with an intended element of social persuasion (Loo and Choy, 2013). Since the GSE scale is widely used across various contexts (Chen et al., 2001), it is relevant to consider how its application aligns with the specific setting of this study. Had we developed a context-specific self-efficacy scale tailored to the project management course (Bandura, 2005), it would have allowed for a more precise measurement of not only students’ general outlook but also their confidence in executing course-specific engineering tasks. This study’s context, a Scandinavian technical university’s engineering course, raises important considerations regarding generalizability and measurement. While focusing on a specific project management course provided a well-defined setting, it also limits the extent to which findings can be applied to other programs or institutions. Cultural and institutional factors (e.g., the university in question’s systemic emphasis on group work and egalitarian participation) may shape self-efficacy development differently than in other educational environments. For instance, the relatively high baseline GSE scores observed could reflect a self-selecting group of already confident students or a cultural environment that inherently supports self-efficacy (Hendrickson, 2021). This could have resulted in a ceiling effect, making it difficult to detect significant growth over time. Male students reported higher self-efficacy than their female counterparts. This aligns with prior research on self-efficacy in STEM (Chan, 2022). Given methodological limitations, it is uncertain whether the project management course provided female students with such experiences. Future research incorporating qualitative methods could offer deeper insights into gender differences in selfefficacy among engineering students. The findings also indicate that students with a study-related job generally reported higher self-efficacy scores than those without, particularly on items related to handling unexpected challenges. This suggests that real-world exposure through employment may play a significant role in shaping students’ perception of their abilities to navigate unforeseen situations. Having a student job likely provides students with practical problem-solving experiences, requiring them to adapt to new challenges, manage tasks under pressure, and develop a sense of resourcefulness. Students with a student job may have greater exposure to professional environments where they observe and engage with experienced colleagues, which could contribute to their sense of vicarious learning (Bandura, 1982; 1994). However, other factors could also influence these differences. For instance, students who seek out student jobs might already possess higher initial confidence levels, making them more inclined to take on responsibilities that further enhance their self-efficacy. Future research could explore whether specific job characteristics, such as task complexity, level of responsibility, or mentorship opportunities, contribute to self-efficacy growth. Additionally, investigating how other experiential learning opportunities could provide further insights into how engineering students develop self-efficacy. In conclusion, by pursuing these future directions and attending to the limitations, the field can advance toward a more nuanced and powerful understanding of how to nurture