Abstract
Background
Climate change poses significant risks to student well-being, yet sustainability education remains limited in higher education curricula. University program directors play a key role in providing sustainability education. This research examined program directors’ beliefs about having a mandatory sustainability course and identified key drivers and barriers to inform intervention strategies aimed at increasing their intention to have such a course.
Focus of the Article
This research used the Reasoned Action Approach to understand what motivates program directors to have a mandatory sustainability course in their programs.
Research Question
What beliefs underlie the intention of university program directors to have a mandatory course on sustainability in their program, and which beliefs are most influential in shaping this intention?
Importance to the Social Marketing Field
This research contributes to social marketing by applying the Reasoned Action Approach to inform the development of interventions to promote sustainability education. The findings provide actionable insights into belief-based strategies that can be linked to social marketing principles for motivating university program directors to have a mandatory sustainability course.
Methods
The research used a mixed-methods approach with two studies. Study 1 used a semi-structured belief elicitation procedure with open-ended questions to identify program directors’ beliefs. In Study 2, the most salient beliefs were transformed into closed-ended questions to quantitatively assess their strength and valuation to determine which beliefs drive program directors’ intention to have such a course.
Results
The results revealed several behavioral beliefs (e.g., “With a sustainability course I can encourage students to think independently and critically”), key referent groups (e.g., advisory boards, teachers, students), and control beliefs (e.g., “Knowing how to ensure program alignment”) as key candidates to target in interventions aimed at increasing program directors’ intention. Such interventions are needed to increase the supply of sustainability education in higher education, as program directors sampled in our research reported a weak intention to have a mandatory sustainability course.
Recommendations for Research or Practice
The findings allow for the development of various social marketing-based intervention initiatives aimed at strengthening the intention of university program directors to have a mandatory sustainability course. Interventions should, for example, target key behavioral beliefs, reinforce normative pressure from advisory boards, teachers, and students, and address control beliefs by, for example, demonstrating how to ensure that a sustainability course fits the program end-terms.
Keywords
Introduction
While climate change poses significant risks to university student’s prospects, their limited ability to influence its course leaves them vulnerable to a range of emotional and psychological impacts, including anxiety, distress, and uncertainty about their future (Pihkala, 2020). A comprehensive global survey found that 59% of young people aged 16–25 are very or extremely worried about climate change, with 84% reporting at least moderate concern (Hickman et al., 2021). From a perspective of market demand, it is crucial for universities to provide education on the climate crisis and equip students with the knowledge and skills needed to understand, confront, and adapt to these complex challenges (Gitsham & Clark, 2014). A recent survey shows that while a majority (53%) of students express a desire to learn about sustainability in their curriculum, only a small percentage (20%) have actually encountered it in their education (Student Minds, 2023). In a context where higher education institutions are often required to gain competitive advantage (Liu et al., 2025; Mahdi et al., 2019; Raposo Junior et al., 2025), offering sustainability courses in their programs can help meet the expectations of stakeholders, particularly prospective students.
Universities typically aspire to provide responsible education and often endorse the UN Sustainable Development Goals (Leal Filho et al., 2019), which include calls to improve education on climate change and sustainability (United Nations, n.d.). Climate change and sustainability issues can be embedded within existing education by introducing the theme into current courses and programs by broadening or deepening the content. Additionally, new elective or mandatory courses can be developed, reaching all students within a program. In line with the latter option, scientists in the Netherlands wrote an open letter calling on higher education institutions “to include mandatory climate education in all curricula in order to familiarize students with the systematic nature of the climate and ecological crisis and empower them to take action” (Alenda et al., 2023). This course should be “tailor-made for each study program, including at least the basic science on global heating and its consequences, and discussing feasible mitigation pathways, and questions of global equity and justice” (Alenda et al., 2023). Including such a mandatory sustainability course in all programs of Dutch Higher Education Institutions would align with the universities’ ambition to provide responsible education. Equipping students with climate change and sustainability knowledge not only raises their awareness but also provides knowledge and skills to participate in and contribute to the development of solutions.
In the Netherlands, university program directors are responsible for the content and structure of their programs and thus can decide whether to include a mandatory sustainability course. In terms of looking at higher education as a service (Quinn et al., 2009), program directors act as product managers, their programs being the offered service. As such, they can play an influential role in increasing the number of programs that include a mandatory sustainability course.
To understand how this demand from stakeholders can be adopted, the present research aims to investigate what motivates program directors to adopt a sustainability course in their programs. We aim to inform future intervention initiatives that seek to strengthen the intention of university program directors to have a mandatory sustainability course. Ultimately, this would result in more students at Dutch universities being prepared to face the challenges of the climate and ecological crisis while facilitating higher education institutions’ competitive advantage as well as societal impact. To do this, we used the Reasoned Action Approach to predicting and changing behavior (Fishbein & Ajzen, 2010), a proven approach to explaining behavior and informing intervention development.
The Reasoned Action Approach (RAA; Fishbein & Ajzen, 2010) is the most recent version of a model from a family of sequentially improved models of behavioral prediction whose predecessors include the Theory of Reasoned Action (Fishbein & Ajzen, 1975), the Theory of Planned Behavior (Ajzen, 1985), and the Integrative Model of Behavioral Prediction (Fishbein, 2000). According to the RAA, intention (how likely program directors think it is that they will have a mandatory sustainability course) is the most proximal and best predictor of the target behavior, given that program directors have the skills and there are no environmental barriers to performing the behavior. Intention, in turn, is predicted by attitude, perceived normative pressure, and perceived behavioral control regarding having a mandatory sustainability course. These determinants each have two components. Attitude includes an instrumental attitude (an evaluation of positive or negative attributes regarding performing the behavior, such as bad or good) and an experiential attitude (an evaluation of positive or negative affective experiences, such as boring or interesting). Perceived norm includes an injunctive norm (perceptions of whether important others think the program director should perform the behavior) and a descriptive norm (the extent to which a program director thinks important others have a sustainability course in their program). Perceived behavioral control includes perceived autonomy (the extent to which a program director believes the decision is up to them) and capacity (the extent to which a program director feels capable of having a mandatory sustainability course).
Attitude, perceived norm, and perceived behavioral control are “cognitive summaries of the full set of highly specific behavioral, normative, and control beliefs” (Yzer & Zhu, 2025, p. 14) of program directors about performing the behavior of having a mandatory sustainability course in their programs (Fishbein & Ajzen, 2010). Attitude is shaped by behavioral beliefs, and these reflect thoughts and feelings about the likelihood and desirability of the consequences or outcomes of having a mandatory sustainability course. Perceived norm is shaped by normative beliefs, that is, thoughts about what important others think and do regarding a sustainability course (and the extent to which program directors want to conform to and be like those important others). Finally, perceived behavioral control is shaped by control beliefs, which are thoughts about the likelihood and power of facilitating or hindering factors for having a sustainability course. What is crucial is that it is these specific beliefs that ultimately guide the intentions and behaviors of people.
According to the RAA (Fishbein & Ajzen, 2010), in order to change intentions and behaviors, the underlying beliefs must be changed. Interventions can serve this purpose by introducing, confirming or reinforcing, or changing relevant beliefs. Importantly, in order to inform the design of such interventions, one must first identify the beliefs that program directors hold about this behavior, and then determine which of these beliefs are most important and thus might be good candidates to target in interventions (Fishbein & Ajzen, 2010; Hornik & Woolf, 1999; Yzer & Zhu, 2025). This is the purpose of the present research.
The Present Research
As recommended by Fishbein and Ajzen (2010), we first conducted a belief elicitation study to identify the salient beliefs (Study 1), followed by a quantitative survey to determine which of these beliefs actually guide intention (Study 2). The target population was defined as university program directors, and we conducted our research with a sample of program directors from the Vrije Universiteit Amsterdam in the Netherlands. The target behavior was defined as “having a mandatory course addressing sustainability in the program(s) in 2026/2027”. Based on the open letter from Dutch scientists (Alenda et al., 2023), we explained to our participants that by a mandatory course, we meant a mandatory course that is made for and fits their study program, including at least the basic science on global heating and its consequences, and discussing feasible mitigation pathways, and questions of global equity and justice. We also informed participants that the course should prepare students to face the challenges posed by the climate and ecological crisis and equip them to lead and effect change. We used “have” in our definition so that both program directors who did not yet have a course like the one described and those who did could answer the questions. In addition, the time element of 2026/2027 was chosen to provide a realistic timeline for implementing a new course (if needed). By specifying the time element we ensured that all program directors were thinking about the same timeline, as time frame has been shown to influence the type of beliefs that become salient (Lutchyn & Yzer, 2011).
Studies 1 (Belief Elicitation Survey) and 2 (Quantitative Survey)
The research was approved by the Research Ethics Review Committee of the Vrije Universiteit Amsterdam. We preregistered the sampling plan and surveys of both studies on the Open Science Framework (OSF). Materials and data can be found at the OSF (https://osf.io/y5mr2/overview). The instructions and survey questions of both studies followed established procedures (i.e., based on the appendix “Constructing a reasoned action questionnaire” in Fishbein & Ajzen, 2010; materials of Hagger et al., 2019). Data were analyzed using Jamovi (2022) and JASP (2024).
Study 1: Belief Elicitation Survey
Methods
Design and Aim
Study 1 employed a semi-structured belief elicitation procedure that used open-ended questions to identify the behavioral, normative, and control beliefs of program directors about having a mandatory course addressing sustainability in their program(s) in 2026/2027.
Participants
Summary Background Variables and Sample Demographics for Studies 1 and 2
aOne participant selected ‘other’, did not indicate a faculty of school, but identified as a program coordinator. We decided to include this participant because program coordinators and directors often work closely together, and the participant’s responses were comparable to the nature of the other participants’ responses, and therefore were considered valuable input to retain.
bOne participant indicated to be coordinator of a bachelor program in Study 1 and in Study 2 one participant indicated to be MBA-program director.
Procedure
Participants were given information about the study and, after providing consent, answered demographic and background questions that were used to describe the sample. Next, the target behavior was described. Participants were told that the questions were about their views on having a mandatory course addressing sustainability in their program(s) in 2026/2027. We explained what we meant with this mandatory course (see section 1.1) and asked participants to list thoughts that immediately came to mind in response to the questions. After answering the open-ended questions, participants were thanked for their participation and directed to a separate survey regarding the gift voucher.
Measures
Behavioral Beliefs
To elicit behavioral beliefs, we asked “What do you see as the advantages of having a mandatory course addressing sustainability in your program(s) in 2026/2027?”, “What do you see as the disadvantages of having a mandatory course addressing sustainability in your program(s) in 2026/2027?”, and “What else comes to mind when you think about having a mandatory course addressing sustainability in your program(s) in 2026/2027?”
Injunctive Normative Beliefs
We told participants “When it comes to having a mandatory course addressing sustainability in your program(s) in 2026/2027, there might be individuals or groups who would think you should or should not have this in your program” and then asked “Please list the individuals or groups who would approve or think you should have a mandatory course addressing sustainability in your program(s) in 2026/2027” and “Please list the individuals or groups who would disapprove or think you should not have a mandatory course addressing sustainability in your program(s) in 2026/2027.”
Descriptive Normative Beliefs
We told participants “Sometimes, when we are not sure what to do, we look to see what others are doing” and then asked “Please list the individuals or groups that you know, and are important to you, that are most likely to have a mandatory course addressing sustainability in their program(s) in 2026/2027” and “Please list the individuals or groups that you know, and are important to you, that are least likely to have a mandatory course addressing sustainability in their program(s) in 2026/2027.”
Control Beliefs
We asked “Please list any factors or circumstances that would make it easy or enable you to have a mandatory course addressing sustainability in your program(s) in 2026/2027” and “Please list any factors or circumstances that would make it difficult or prevent you from having a mandatory course addressing sustainability in your program(s) in 2026/2027.”
Results
Description and Frequency of the Identified Beliefs
Note. + Selected for the quantitative survey of Study 2. * Included in Study 2 but added by the researchers.
Discussion
Study 1 identified many different beliefs that represent how participating program directors think about having a mandatory course addressing sustainability in their program(s) in 2026/2027. Behavioral beliefs revealed thoughts about the topic of sustainability and its place in the program. Teachers and students in the program were mentioned as the main groups of injunctive referents, while other programs and program directors were mentioned as the main groups of descriptive referents. Beliefs about facilitating or hindering control factors were related to budget, space and alignment in the program, and top-down decision making. Study 2 investigated which of the identified beliefs actually guide program directors’ intention to have a mandatory sustainability course.
Study 2: Quantitative Survey
Methods
Design and Aim
Study 2 included a survey with close-ended questions to examine which beliefs drive program directors’ intention to have a mandatory sustainability course.
Participants
In late May 2024, we emailed a link to the survey to all program directors. According to a Vrije Universiteit Amsterdam website (visited on March 28, 2024), there were 109 program directors at that moment. We spread out the survey by mailing flyers, emailing a reminder (early June), and contacting by phone (late June). Participation was voluntary and participants were offered a €10 gift voucher. We decided to allow future program directors to participate as well to increase the response rate. The survey closed on July 30. A total of 61 participants started the survey, of which 45 answered all questions and thus made up our final sample (see Table 1 for sample characteristics).
Procedure
Similar to Study 1, participants were given information about the study and, after giving consent, answered demographic and background questions that were used to describe the sample. Next, the target behavior was described as in Study 1. After answering the survey questions, participants could leave a comment if they wished, were thanked for their participation, and then directed to a separate survey where they could leave their email address if they wished to receive the gift voucher.
Measures
Intention
Intention was measured with three items: “It is likely that I will have a mandatory course addressing sustainability in my program(s) in 2026/2027”, “I expect I will have a mandatory course addressing sustainability in my program(s) in 2026/2027”, and “I want to have a mandatory course addressing sustainability in my program(s) in 2026/2027” (1 = strongly disagree, 7 = strongly agree). Scores were averaged to form an intention scale (α = .90).
Behavioral Beliefs
The 12 beliefs selected from Study 1 were translated into survey questions that measured both outcome strength and outcome evaluation for each belief. To measure outcome strength, we used the stem “If I have a mandatory course addressing sustainability in my program(s) in 2026/2027, I would …” (e.g., need to remove a topic from my current program(s), address a topic that students find appealing, impart values or opinions to my students) and the response scale of 1 = very unlikely to 7 = very likely. To measure outcome evaluation, we used the stem “When it comes to having a mandatory course addressing sustainability in my program(s) in 2026/2027, …” (e.g., my needing to remove a topic from my current program(s) is, my addressing a topic that students find appealing is, my imparting values or opinions to my students is) and the response scale of 1 = bad to 7 = good. We computed the product of outcome strength and evaluation for each belief and calculated a behavioral beliefs composite score by summing the belief-specific product scores (cf. Fishbein & Ajzen, 2010).
Instrumental and Experiential Attitude
Four seven-point semantic differential items measured attitude. We used the stem “My having a mandatory course addressing sustainability in my program(s) in 2026/2027 would be:” followed by the items bad-good and unimportant-important (for instrumental attitude), and unsatisfying-satisfying and boring-interesting (for experiential attitude). Scores were averaged to form an instrumental attitude (τb = .678, 95% CI [.492, .834], p < .001; note that we used Kendall’s tau-b correlation, see data analysis plan below) and an experiential attitude scale (τb = .771, 95% CI [.603, .906], p < .001).
Injunctive Normative Beliefs
The seven beliefs selected from Study 1 were translated into survey questions that measured both injunctive belief strength and motivation to comply for each belief. To measure injunctive belief strength, we used the stem “The following people or groups are likely to think I should have a mandatory course addressing sustainability in my program(s) in 2026/2027 …” (e.g., teachers in my program(s), students in my program(s), my faculty board) and the response scale of 1 = very unlikely to 7 = very likely. To measure motivation to comply, we used the stem “When it comes to matters of sustainability education, …” (e.g., I want to do what teachers in my program(s) think I should do, I want to do what students in my program(s) think I should do, I want to do what my faculty board thinks I should do) and the response scale of 1 = strongly disagree to 7 = strongly agree. We computed the product of injunctive belief strength and motivation to comply for each belief and calculated an injunctive normative beliefs composite score by summing the belief-specific product scores (cf. Fishbein & Ajzen, 2010).
Descriptive Normative Beliefs
The four beliefs selected from Study 1 were translated into survey questions that measured both descriptive belief strength and identification with referents for each belief. To measure descriptive belief strength, we used the stem “The following entities are likely to have a mandatory course addressing sustainability in their program(s) in 2026/2027 …” (e.g., programs that have a natural relationship with sustainability, programs similar to mine at other universities (national or international), other program directors) and the response scale of 1 = very unlikely to 7 = very likely. To measure identification with referents, we used the stem “When it comes to matters of sustainability education, …” (e.g., I want my program to be like programs that have a natural relationship with sustainability, I want my program to be like similar programs at other universities (national of international), I want to be like other program directors) and the response scale of 1 = strongly disagree to 7 = strongly agree. We computed the product of descriptive belief strength and identification with referents for each belief and calculated a descriptive normative beliefs composite score by summing the belief-specific product scores (cf. Fishbein & Ajzen, 2010).
Perceived Injunctive and Descriptive Norm
Two items measured perceived injunctive norm: “Most people whose opinions I value would approve of my having a mandatory course addressing sustainability in my program(s) in 2026/2027” and “Most people who are important to me think that I should have a mandatory course addressing sustainability in my program(s) in 2026/2027”. Two other items measured perceived descriptive norm: “Most people I respect and admire will have a mandatory course addressing sustainability in their program(s) in 2026/2027” and “Most people like me will have a mandatory course addressing sustainability in their program(s) in 2026/2027”. All items used a response scale of 1 = strongly disagree to 7 = strongly agree. Scores were averaged to form a perceived injunctive norm (τb = .697, 95% CI [.539, .834], p < .001) and a perceived descriptive norm scale (τb = .834, 95% CI [.685, .946], p < .001).
Control Beliefs
The nine beliefs selected from Study 1 were translated into survey questions that measured both control belief strength and power of control factors for each belief. We first measured power of control factors, using the stem “Do you agree that …?” (e.g., Having available budget would enable me to have a mandatory course addressing sustainability in my program(s) in 2026/2027, Being able to ensure a good fit with my program and end-terms would enable me to have a mandatory course addressing sustainability in my program(s) in 2026/2027, A top-down decision to make as sustainability course obligatory would enable me to have a mandatory course addressing sustainability in my program(s) in 2026/2027) and the response scale of 1 = strongly disagree to 7 = strongly agree. To measure control belief strength, we used the stem “When it comes to matters of sustainability education, …” (e.g., I will have budget available, I will be able to ensure a good fit with my program and end-terms, I will experience a top-down decision to make a sustainability course obligatory) and the response scale of 1 = very unlikely to 7 = very likely. We computed the product of control belief strength and power of control factors for each belief and calculated a control beliefs composite score by summing the belief-specific product scores (cf. Fishbein & Ajzen, 2010).
Perceived Behavioral Control (Autonomy and Capacity)
Two items measured perceived autonomy: “It is mostly up to me whether I have a mandatory course addressing sustainability in my program(s) in 2026/2027” and “I have complete control over whether I have a mandatory course addressing sustainability in my program(s) in 2026/2027”. Two other items measured perceived capacity: “I am confident I can have a mandatory course addressing sustainability in my program(s) in 2026/2027” and “It would be easy for me to have a mandatory course addressing sustainability in my program(s) in 2026/2027”. All items used a response scale of 1 = strongly disagree to 7 = strongly agree. Scores were averaged to form a perceived autonomy (τb = .744, 95% CI [.626, .838], p < .001) and a perceived capacity scale (τb = .651, 95% CI [.457, .803], p < .001).
Data Analysis Plan
Following the recommended approach (Crutzen & Peters, 2023; Fishbein & Ajzen, 2010), we focused on examining bivariate associations and univariate distributions. We first examined descriptive statistics of the variables and then moved to analyses at the level of the individual beliefs, which is the level most informative for providing input for the development of an intervention to motivate program directors to have a sustainability course in their program. Given the relatively small sample size, we used Kendall’s tau-b as a measure of correlation rather than Pearson’s correlation. Kendall’s tau-b is a non-parametric statistic based on ranked data and is less sensitive to extreme values and potential violations of assumptions (Field, 2013), ensuring robustness, also reporting the associated confidence intervals based on 1,000 bootstraps. The strength of the correlation coefficients was interpreted as follows: around .00 as negligible, around .06 as weak, around .26 as moderate, around .49 as strong, and around .71 as very strong (Wicklin, 2023).
Results
Descriptive Statistics of Reflective Variables: Means, Standard Deviations, and Kendall’s tau-b Correlations (With Confidence Intervals Based on 1,000 Bootstrap Replicates in Brackets)
Note. Means are relative to scales ranging from 1 to 7 with higher scores reflecting more positive or stronger reflective variables. * p < .05, ** p < .01, *** p < .001.
We also correlated the variables with their respective beliefs composite scores. A moderate to high correlation suggests that the measured beliefs underlie the RAA variables (cf. the theoretical assumption); a low or no correlation could indicate that the measured beliefs do not underlie the corresponding RAA variables (Fishbein & Ajzen, 2010). The behavioral beliefs composite score correlated strongly with both the instrumental (τb = .584, 95% CI [.412, .735], p < .001) and experiential attitude (τb = .597, 95% CI [.406, .747], p < .001). The injunctive normative beliefs composite score showed a moderate correlation with the injunctive norm (τb = .484, 95% CI [.294, .651], p < .001), and the descriptive normative beliefs composite score was moderately correlated with the descriptive norm (τb = .302, 95% CI [.064, .525], p = .006). Finally, the control beliefs composite score was moderately correlated with perceived capacity (τb = .418, 95% CI [.214, .612], p < .001) but unrelated to perceived autonomy (τb = .027, 95% CI [-.231, .289], p = .807). The results suggest that the beliefs measured in the current study do indeed underlie their corresponding RAA variables, with the exception that control beliefs appear to reflect only capacity-related, but not autonomy-related, perceived behavioral control. This may be because the questions used in Study 1 to identify control beliefs primarily triggered capacity-related beliefs.
Behavioral Beliefs About Having a Mandatory Sustainability Course: Means and Standard Deviations of Outcome Strength, Outcome Evaluation, Belief-Specific Product Scores, and Kendall’s tau-b Correlations (With p-Value, Confidence Intervals Based on 1,000 Bootstrap Replicates, and Classification) Between the Product Score and the Intention to Have Mandatory Course Addressing Sustainability
Note. The means of outcome strength and evaluation are relative to scales ranging from 1 (very unlikely; bad) to 7 (very likely; good). The mean of the product score is relative to a scale from 1 to 49 with higher scores reflecting a higher impact of the belief on the overall attitude. * p < .05, ** p < .01, *** p < .001.
Injunctive Normative Beliefs About Having a Mandatory Sustainability Course: Means and Standard Deviations of Injunctive Beliefs Strength, Motivation to Comply, Belief-Specific Product Scores, and Kendall’s tau-b Correlations (With p-Value, Confidence Intervals Based on 1,000 Bootstrap Replicates, and Classification)
Note. The means of injunctive belief strength and motivation to comply are relative to scales ranging from 1 (very unlikely; strongly disagree) to 7 (very likely; strongly agree). The mean of the product score is relative to a scale from 1 to 49 with higher scores reflecting a higher impact of the belief on the overall injunctive norm. * p < .05, ** p < .01, *** p < .001.
Descriptive Normative Beliefs About Having a Mandatory Sustainability Course: Means and Standard Deviations of Descriptive Beliefs Strength, Identification With Referents, Belief-Specific Product Scores, and Kendall’s tau-b Correlations (With p-Value, Confidence Intervals Based on 1,000 Bootstrap Replicates, and Classification)
Note. The means of descriptive belief strength and identification with referents are relative to scales ranging from 1 (very unlikely; strongly disagree) to 7 (very likely; strongly agree). The mean of the product score is relative to a scale from 1 to 49 with higher scores reflecting a higher impact of the belief on the overall descriptive norm. * p < .05, ** p < .01, *** p < .001.
Control Beliefs About Having a Mandatory Sustainability Course: Means and Standard Deviations of Control Belief Strength, Power of Control Factors, Belief-Specific Product Scores, and Kendall’s tau-b Correlations (With p-Value, Confidence Intervals Based on 1,000 Bootstrap Replicates, and Classification)
Note. The means of control belief strength and power of control factors are relative to scales ranging from 1 (very unlikely; strongly disagree) to 7 (very likely; strongly agree). The mean of the product score is relative to a scale from 1 to 49 with higher scores reflecting a higher impact of the belief on the overall perceived behavioral control. * p < .05, ** p < .01, *** p < .001.
Five of the 12 behavioral beliefs showed a moderate to strong relationship with intention (Table 4). The beliefs that having a sustainability course makes it possible to link discipline-specific topics to sustainability (τb = .511) and encourages independent and critical thinking in students (τb = .410) were related to a stronger intention to have such a course. The beliefs that a sustainability course would provide an appealing topic for students (τb = .337), prepare them for their professional life (τb = .327), and make them more aware of sustainability (τb = .284) were also related to a stronger intention, although to a lesser extent. Furthermore, perceived normative pressure from all injunctive referents was associated with intention (Table 5), especially perceived normative pressure from advisory boards (τb = .517; e.g., program committee, student council, works council) and teachers (τb = .475) and students (τb = .415) within the program. Beliefs about what similar programs at other universities (τb = .414) and programs that have a natural relationship with sustainability do (τb = .301), and the extent to which participants want to be like them, were also associated with intentions to have a sustainability course in the program (Table 6). Finally, six of the nine control beliefs showed a relationship with intention (Table 7). Specifically, beliefs about having space in the curriculum (τb = .474), being able to ensure a good fit with the program and the end-terms (τb = .450), and having staff knowledgeable about sustainability (τb = .358) shaped program directors’ intentions to have a sustainability course in their program.
Discussion
Study 2 showed that program directors had a weak intention to have a mandatory sustainability course in their program in 2026/2027. Thus, it makes sense for an intervention to focus on changing intentions to try to increase the number of programs with a sustainability course. According to the RAA (Fishbein & Ajzen, 2010), this could be achieved by changing program directors’ attitude, perceived norm, or perceived behavioral control. Consistent with this reasoning, the results showed that all variables (except the perceived autonomy component of perceived behavioral control) were related to intention, and all scores were mostly around the neutral midpoint of the scales, indicating that there is sufficient room for change and that, on average, program directors are not too averse to the possibility of having a mandatory sustainability course. Beliefs provide the foundation for these variables, and the specific behavioral, normative, or control beliefs that are most strongly related to the intention are those that should be targeted in an intervention to make it most likely to change intentions (Fishbein & Ajzen, 2010). Study 2 provided this information by pinpointing which of the specific beliefs from Study 1 guided program directors’ intentions to have a mandatory sustainability course.
General Discussion
We used the RAA to investigate what motivates university program directors to have a mandatory sustainability course in their program in 2026/2027. Study 1 identified the beliefs held by program directors about having a mandatory sustainability course. Study 2 showed that program directors had a weak intention to have a mandatory sustainability course, highlighting that interventions are needed to increase the supply of sustainability education in higher education. Study 2 identified the specific beliefs from Study 1 that most strongly guided program directors’ intention, providing key targets for interventions aimed at increasing these intentions.
To identify key beliefs, we followed three guidelines established in the literature (Crutzen et al., 2017; Crutzen & Peters, 2023; Fishbein & Ajzen, 2010; Fishbein & Yzer, 2003; Hornik & Woolf, 1999; Yzer & Zhu, 2025). First, the belief should be strongly related to intention. A stronger relationship increases the likelihood that changing the belief will result in a change in intention. Second, there should be room for changing the belief (i.e., potential ceiling or floor effects should be considered). If a belief is associated with intention, but most program directors already strongly (or weakly) hold or value that belief, there may not be much to gain by further strengthening (or weakening) that belief through an intervention. Instead of trying to change these beliefs, an intervention might focus on reinforcing them to strengthen the association between belief and intention (Fishbein & Yzer, 2003). Third, it should be possible to change or influence the belief through an intervention.
Several behavioral beliefs from our study may be good target candidates for interventions. The beliefs that a sustainability course facilitates linking discipline-specific topics to sustainability and encourages students to think independently and critically were most strongly related to the intention of having a sustainability course. Beliefs that a sustainability course addresses a topic that students find appealing, prepares students for their professional life, and increases student awareness of sustainability and its consequences, including links to the discipline were also associated with a stronger intention. While these beliefs show positive associations with intention, there may not be much room for change. This is because most program directors already seem to agree that these are likely and positive outcomes of having a sustainability course. However, if an intervention targets program directors who are not yet convinced, these beliefs may be highly relevant. Furthermore, an intervention to confirm and reinforce these five beliefs could increase program directors’ intention to have a sustainability course in their program. Among these beliefs, the idea that sustainability is an appealing topic for students is especially significant. Students are particularly drawn to sustainability education that equips them with actionable perspectives to navigate the planetary crisis. This shows a demand Dutch higher education institutions are not yet meeting. Students value courses that are current, applied, and solution-oriented, empowering them to address real-world challenges and contribute meaningfully within their disciplines (Youngworks, 2023). Such courses resonate with students’ interests and align with their goals for professional preparation. Program directors can design courses that meet both academic and career-oriented expectations by emphasizing how sustainability education enhances readiness for professional life, equipping students with relevant skills and insights and increasing awareness of sustainability’s consequences and its intersections with the fields of study. Recognizing and integrating these preferences into course design could strengthen the appeal of sustainability courses. This would, in turn, reinforce program directors’ intentions to include them. Previous research also shows that there is a demand for professionals to be knowledgeable in sustainability-related topics (e.g., Gitsham & Clark, 2014), making this study also relevant for understanding how higher education institutions respond to (internal and external) stakeholder pressure (Christou et al., 2024).
Perceived normative pressure from all groups (injunctive referents) in this research was associated with intention and could therefore be used in interventions. However, normative pressure from advisory boards (e.g., program committee, student and works council), teachers and students within their own program appear to have the most influence. Program directors are also most likely to listen to these three groups, as indicated by their reported levels of motivation to comply with these groups. Interventions that successfully get advisory boards, teachers, and students to communicate to program directors that they want a sustainability course in the program (e.g., by reinforcing the relevant behavioral beliefs identified in this research) may increase directors’ intention to have such a course. Descriptive normative beliefs (about similar programs at other universities and programs with a natural connection to sustainability) were related to intention but to a lesser extent than injunctive normative beliefs. Unsurprisingly, program directors believed that programs with a natural connection to sustainability were more likely to offer a mandatory sustainability course. It is possible, then, that an intervention increasing program directors’ identification with such programs may boost their intention to have a course. Interventions that make program directors aware of similar programs at other universities that offer a mandatory sustainability course may positively influence intention by strengthening this descriptive belief. Moreover, product and service developments often follow the path of their competitors, and the higher education sector is not immune to this sort of market logic (Marginson, 2006).
Several control beliefs may be good candidates for interventions. Specifically, beliefs about space in the curriculum, fit with the program and the end-terms and staff knowledge about sustainability showed the strongest correlations with intention and thus appear to be the most influential. In addition, the scores were around the scales’ midpoint, suggesting that there is room for change. Naturally, to be a candidate for intervention, beliefs should be amenable to change. While a typical intervention may not be able to create space in the curriculum (unless a top-down decision is made to enforce it), it could teach program directors how to incorporate the topics of the proposed mandatory sustainability course (e.g., the basic science on global heating and its consequences) into the current curriculum. Interventions might also demonstrate how to ensure a good fit with the program and its end-terms, and how to increase sustainability knowledge among staff. More indirectly, interventions focusing on educating teachers about sustainability could both address this control belief and perhaps fuel the normative pressure that teachers have on their program directors to have a sustainability course.
In conclusion, the current findings allow for the development of interventions aimed at strengthening the intention of university program directors to implement a mandatory sustainability course. Ultimately this could increase the number of programs with a sustainability course, help universities achieve their green ambitions, and equip more students with sustainability literacy. This provides students with the necessary awareness, knowledge and skills to participate in and contribute to the development of solutions. These are assets needed in this time of widespread, rapid, and intensifying climate change (Budihardjo et al., 2021; Gitsham & Clark, 2014). To further assist interventionists, Figure 1 visually summarizes our key findings, highlighting the desired focus (labeled “Activity” in Figure 1) and the most relevant beliefs to target (labeled “Thoughts of the program director” in Figure 1). Inspired by Avram (2025), Table 8 provides examples of how the identified beliefs translate into concrete social marketing strategies. Visual summary of the results Examples of How to Translate Identified Beliefs Into Concrete Social Marketing Strategies That Align With the 4 Ps of Social Marketing
Moreover, besides what is mentioned in Table 8, there are other intervention approaches that have proved to be successful. First, following the example of Arizona State University (Wiek et al., 2014), program directors can incorporate problem-based learning modules that address real-world sustainability challenges into existing courses. This approach covers attributes that program directors deem important, such as fostering critical thinking, linking disciplinary content to sustainability, and preparing students for professional practice (behavioral beliefs). Second, since advisory boards, teachers, and students are influential referents (normative beliefs), co-creation mechanisms can reinforce social norms that support sustainability integration. The University of Exeter’s Green Futures initiative and Utrecht University’s Sustainability Education & Engagement community demonstrate how participatory networks encourage shared ownership of sustainability education (University of Exeter, n.d; Utrecht University, n.d.). Finally, to address constraints such as limited curriculum space and staff expertise (control beliefs), universities can support faculty through professional development programs, curriculum mapping initiatives, and shared modular courses. For example, the 4TU Responsible Sustainability Challenge is a collaborative, challenge-based honors program offered by Dutch technical universities (Fuentes Bongenaar et al., 2024). Together, these practices demonstrate that belief-based interventions can be translated into real courses through collaboration, resource sharing, and institutional support.
Strengths and Limitations
Most research using the RAA to inform interventions assesses the predictive strength of attitude, perceived norm, and perceived behavioral control on intentions. However, these studies typically do not investigate specific beliefs underlying these determinants (Fishbein & Ajzen, 2010; Yzer, 2023). While such research can provide information about which determinant an intervention might best target (e.g., attitude because it was the strongest predictor of intention), it does not specify the content of that intervention (e.g., which belief to change or reinforce). Thus, the current research, which included a belief elicitation study and belief-based measures in the quantitative survey, provides a detailed understanding of program directors’ beliefs about mandatory sustainability education and thus direct guidance that interventionists can use in developing their interventions. In addition, studies that assess beliefs often assess only one aspect, such as belief strength, whereas the current study–consistent with the RAA (Fishbein & Ajzen, 2010)–also assessed their evaluations (or motivation to comply, identification with referents, or power of control factors). This provides useful information for interventionists.
A limitation of the present research is that we cannot generalize the findings to populations of program directors other than those from the Vrije Universiteit Amsterdam. Dutch universities generally operate similarly, with program directors primarily responsible for designing and developing their programs. Thus, they are key decision-makers in the curriculum. However, broader organizational and contextual factors also influence implementation. According to Weiss and colleagues (2021), institutional leadership, staff development, cross-campus integration, and stakeholder engagement are main drivers, while limited resources, coordination, and incentives are major barriers. These factors shape the organizational climate and decision-making autonomy of program directors, which may then influence the relative importance of the beliefs identified in this study. Similarly, UNESCO (2018) emphasizes that national policies and institutional commitment are decisive for embedding sustainability education. Therefore, these contextual differences may affect how program directors in other settings form and prioritize their beliefs about implementing a mandatory sustainability course. To test the robustness of the identified belief set, we recommend replicating our belief elicitation and quantitative approach in other countries and institution types.
In addition, the findings are specific to the concrete behavior examined in our research: having a mandatory course addressing sustainability in the program(s) in 2026/2027. While these aspects are limitations to generalizability, they also reflect a key aspect of the RAA, which argues that behaviors have unique underlying belief systems that may differ between populations and types of behaviors. That is, changing one of the elements in the definition of the behavior of interest (target, action, context, or time) changes the behavior, which may affect the nature or importance of the beliefs that underlie the behavior (Fishbein & Ajzen, 2010). Thus, it may well be that the beliefs identified in the current research are also relevant for other populations of program directors or for related target behaviors (e.g., a non-mandatory sustainability course or sustainability education in general), but we cannot be sure without conducting a new belief elicitation study and quantitative survey. However, the current research demonstrates how this can be done relatively easily, providing high-quality input for intervention development. It is advisable though to pilot test the questionnaires to improve clarity and accessibility, as our participants mentioned that the survey questions were sometimes difficult to answer.
Conclusion
We employed the RAA to provide belief-based guidance for developing interventions to increase sustainability education at the Vrije Universiteit Amsterdam. Consistent with prior views (e.g., Yzer & Zhu, 2025) and empirical evidence (e.g., Hornik et al., 2019), developing interventions using the beliefs identified in the current research will be far superior to developing interventions based on intuition. Ultimately, we hope this will equip more students with essential knowledge about climate change and sustainability, raising their awareness and providing them with the knowledge and skills to participate in and contribute to developing solutions. Furthermore, this research can inspire other higher education institutions to identify and appeal to the beliefs that motivate their program directors to incorporate sustainability education into their programs.
Footnotes
Acknowledgements
We thank Sanchayan Banerjee, Nicky Bosman, Simone Burger, Jessica Hill, Clemens Kaupa, Meike Morren, and Karen Verduijn for their feedback during this research project. We also thank Lucia de Rosa-Hansen for her help with the data collection.
Ethical Considerations
The research was approved by the Research Ethics Review Committee of the Vrije Universiteit Amsterdam (approval numbers 2024-3-1-310/2024-3-27-960 (Study 1) and 2024-4-24-375/2024-4-24-615 (Study 2)).
Consent to Participate
Participants provided consent to participate by selecting this option in the online surveys (i.e., written consent). Informed consent was required by Research Ethics Review Committee of the Vrije Universiteit Amsterdam.
Consent for Publication
Informed consent for publication of the findings was provided by the participants (i.e., written consent).
Author Contributions
GK: Conceptualization, data curation, formal analysis, funding acquisition, methodology, writing – original draft, writing – review & editing.
AL: Conceptualization, funding acquisition, methodology, writing – review & editing.
JA: Formal analysis, methodology, writing – review & editing.
JT: Funding acquisition, writing – review & editing.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Amsterdam Sustainability Institute Seed Money Program 2024 of the Vrije Universiteit Amsterdam.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
Declaration of Generative AI and AI-Assisted Technologies in The Writing Process
During the preparation of this work, the authors used DeepL Write to improve the readability and language, as well as NotebookLM to create an initial draft of Table 8. After using these tools/services, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.