Although computer modelling and simulation (M&S) methodologies have been popular among physical scientists for 50 years, taking advantage of such methodologies by social scientists began only after the 1990s (Alvarez 2016; Gilbert and Conte 1995; Lane 2013; Shults 2019). The M&S methods offered social scientists the opportunity to run social experiments by creating a virtual world. They manipulate parameters to produce different conditions that match their experiences in the real world and monitor the situations as they evolve in the virtual world. The M&S methodologies transform their research practices, which take the static representation of information to dynamic virtual worlds where the consequences of different imaginary conditions of societal dynamics are experienced as if alive. As Wesley Wildman and colleagues (2017) argue, M&S methods in university curricula could raise awareness that might encourage students to consider applying such methods in their future careers.
Advocates of M&S methods in higher education curricula have applauded the methods, saying they increase student motivation by enabling them to take a scientific stance in examining social issues, facilitate understanding of complex systems (e.g., social inclusion and exclusion, migration dynamics), stimulate students in critical conversation and begin to ask their questions (Demerath et al. 2020; Holter and Schwesinger 2020; Poudel et al. 2020; Szczepanska et al. 2020). A variety of M&S-based software has been created to help with learning, teaching and research about complex societal dynamics, for example, GAMA, an agent-based simulation platform that enables users to build their own multi-agent simulation platform considering their specific needs (Taillandier et al. 2012), Air Pollution (APoME) (Wu 2010), ViMAP Program (Sengupta et al. 2015), and Anylogic Software (The AnyLogic Company 2015). Researchers and university students can interact with these tools or software based on their knowledge, skill and experience (Hsu et al. 2012; Lehrer and Schauble 2012).
I offer two examples of university curricula that introduced the M&S methods into methodology courses alongside the more familiar research techniques, such as questionnaire surveys and interviews. First, Erica Holter and Sebastian Schwesinger (2020) report on implementing the M&S methods as an element of the archaeological curriculum for MA programmes in digital archaeology at Humboldt University of Berlin. They argued that the use of M&S methods transformed their academic practices. This transformation included shifting from unidirectional knowledge transfer and traditional museum displays to dynamic forms of archaeological communication. These dynamic forms involve multiple visualisations, adaptive or interactive games, and immersive experiences to engage a broader audience. Second, emphasising the importance of the M&S methods, Timo Szczepanska and colleagues (2020) documented their study of M&S methods within their research-oriented master's programme, called ‘Urban Futures’ at the Potsdam University of Applied Sciences, Germany. A three ECTS1 point course within the newly launched master's programme focuses on teaching technological and methodological skills for future leaders of urban change through a well-founded analysis of complex relationships between the physical, socio-cultural, and informational fields of urban systems.
This article reports on a study of how undergraduate students utilised an opportunity to learn about M&S-based research methods. The small-scale study was initiated with the Social Simulation Research Group2 to design an M&S-based research methods module for students following a bachelor-level religious studies programme. It was hypothesised that the students would develop a sense of how researchers in their field utilised M&S-based research methods. The study also observed how students utilised the opportunities to explore M&S-based research methods with the M&S-based professional practitioners in their field.
The article is organised as follows: The next section discusses why and how the M&S-based research methods course was introduced and revised. I then describe the empirical methodology used and what students gained from the round-table discussion with M&S-based researchers. I conclude the article by summarising the results and reflecting on the implications for research methodology courses in religious study programmes.
Why and how was the M&S-based research methods course introduced and revised?
Research methods courses within university programmes play a critical role in students’ interest in the course and their learning of research skills and competencies (Kilburn et al. 2014; Lovell 2019). Research methodology courses offer tools for students to apply theoretical and procedural knowledge as well as practical skills. Daniel Kilburn and colleagues (2014) emphasise the importance of research methods courses in three different ways: (1) actively engaging students in research processes, (2) creating opportunities for students to learn by doing, and (3) encouraging students’ reflection on research practices.
On the other hand, Darrell Lovell (2019) argues that introducing research methods courses at the undergraduate level requires an emphasis on mentoring students to increase their knowledge of empirical design and methodology. Research methods curricula are often limited to traditional coursework, such as survey analysis, interviews, participant observation and ethnography. Based on these curricula, coursework focuses on superficial learning about the research methods without actively engaging in the research processes. Contrary to this conventional curriculum, Ron Griffiths supports a research-based curriculum in which students learn as future professionals, with the curriculum specifically designed to actively involve students in research processes, thus minimising the division of roles between teacher and student (2004: 722). This curriculum can, for instance, provide students with a deeper understanding of the scope, opportunities and challenges inherent in applying the research methodology in their respective fields. Furthermore, a research-based course module that allows students to design and implement a small-scale field-oriented research project can greatly enhance their understanding of the research process, as opposed to a mere focus on subject content acquisition. In this fashion, the research methodology curriculum could effectively bridge teaching and research in undergraduate education programmes, as suggested by Griffiths (2004) and Mick Healey (2005).
In Norway, it is generally assumed that a quality indicator for a study programme in higher education is that it should be founded on, informed by, and infused with research and development (R&D) that is current within the disciplinary core of the programme (Meld. St. 16 2016–2017). A solid foundation in the discipline's R&D ensures that the curriculum is cutting-edge, future-oriented, relevant, illuminating, and above all, exciting! A university education holds the possibility of undergraduate students engaging with practitioner researchers working on the cutting edge of their field, and these opportunities should be exploited.
According to David Gibson and Dirk Ifenthaler (2017) and Fabian Held and Ian Wilkinson (2018), undergraduate students should have the opportunity to experience emerging research approaches, such as M&S, within a research methodology curriculum that fosters new horizons of knowledge and new types of skills. I was aware that a group of researchers at my university utilised M&S-based research methods to study social dynamics, such as religious and social conflicts (Gore et al. 2018; Lane and Shults 2020; Shults et al. 2018a). Students are legitimate peripheral participants in the community of university researchers and professionals who utilise various research methods in their workplace environment (Lave and Wenger 1991). Participation can be both through formal and informal activities. In this vein, Lave and Wenger (1991: 98) defined a community of practice as ‘a system of relationships between people, activities, and the world, developing with time, and in relation to other tangential and overlapping communities of practice’. The university community shares a common practice of research methods, as well as a repertoire of concepts, tools and artefacts relevant to the practice.
In this environment, students can utilise the opportunity to learn about methodology through a research methods course, at formal meetings (i.e., seminars, workshops, course work) and informal meetings (i.e., lunchtime gatherings), and by tangential participation, such as ‘circling around’ (as used in Engeström 2005). I wanted to explore how students would react to being exposed to this emerging M&S-based research methodology.
Modelling and simulation as a research method
Modelling and simulation-based methods are emerging as a tool to represent and study societal problems. ‘Many social problems are difficult to address adequately with traditional analytical and statistical techniques due to the diversity and a great number of factors involved (e.g., evolution of culture), complicated dynamics (e.g., social networks) and hard-to-measure social processes (e.g., psychological processes, world size phenomena)’ (Hassan et al. 2010: 136). M&S-based methods enable social scientists to model and simulate social processes (Gilbert 1999), bringing them closer to the real world (e.g., Alvarez 2016; Gore et al. 2018; Lane 2013; Shults 2019).
Researchers’ motivation to choose M&S-based research methods include the following: (i) traditional research methods are not sufficient to investigate the dynamics of complex social systems, as social phenomena cannot be represented by static and simple interaction rules (Hmelo et al. 2000), (ii) the ability to alter computer simulation parameters, which allows researchers to create ‘what-if’ experiments for factors that might affect complex social problems, such as religious conflicts, refugees and tolerance (Gore et al. 2018; Lane and Shults 2020; Shults et al. 2018a), (iii) modelling a social phenomenon is a simplified representation of real-world phenomena, therefore the new approach can be a helpful tool for testing theories against empirical reality, for evaluating assumptions of theories, and for understanding the interaction between theories (Lave and March 1993; Shults et al. 2018b), and (iv) M&S methods help researchers to avoid hindrances of cost, ethical issues and time factors by creating virtual societies as reasonable substitute for live experiments (Poudel 2021; Poudel et al. 2020).
An early example of such an artificial society was Schelling's (1971) checkerboard simulation of residential segregation, which showed that almost total segregation could result from slight individual bias towards their neighbours. Like Schelling's model, artificial societies imitate a real-world phenomenon; setting up a series of simulations allow researchers to run virtual experiments through changing social or community characteristics.
More recently, M&S methods have made inroads into social science, generating new insights into religious conflicts, refugees, tolerance, religious belief and affiliation in secularising contexts and other fields (Boshuijzen-van Burken et al. 2020; Shults 2019; Gore et al. 2018; Shults et al. 2018a). In this way, the agent-based modelling approach is helpful in exploring the dynamics of social networks by modelling complex systems and observing the effects of individual actions by autonomous agents and their interaction with the system as a whole (Gilbert and Troitzsch 2005; Wilensky and Rand 2015). In this way, the M&S method offer advantages over traditional methods in social sciences such as statistical analysis by creating an artificial society to study trends and theories of complex social dynamics.
Empirical methodology
To achieve the study goals, I used a design-based approach. The methodology supports the design of new educational practices in which a new curriculum module is developed and implemented (Bakker 2018; Plomp 2013). The design-based methodology was helpful to utilise the iterative, cycle method of curriculum practice – ideate, develop, implement, evaluate, re-design and start all over again – as an empirical context to gain new scientific knowledge about how students used the opportunity to learn about new methodology in their fields. The first iteration of the module ‘M&S-based research methods’ was piloted in the fall semester of 2017 with a group of undergraduate students registered in a religious studies programme. The second iteration was conducted with development studies students at the same university. With these interventions, the study's goal was to explore and make sense of how students engage with the opportunities to learn about M&S-based research methods. I also followed documentation (i.e., an account of design, implementation, and evaluation processes) of iterative cycles of the M&S-based research methods module, which allows for the understanding of design processes, challenges and failure of the design that could be improved in successive iterations (McKenney and Reeves 2012; O'Neill 2016; Van den Akker et al. 2013; Van den Akker et al. 2006). It was observed that a three-hour seminar was insufficient to learn about M&S-based research methods for novices.
I redesigned the second cycle of the module to include an hour-long tutorial session for a small group to help students clarify the concepts associated with these methods. Further, the tutoring sessions were intended to help students who chose to write an essay about M&S-based research methods. During these sessions, students raised some critical questions that demanded an expert (i.e., M&S-based researcher) opinion. At this point, I realised that the learning and knowing processes required first-hand interaction with M&S-based researchers so that students could ask questions regarding the usefulness, opportunities, challenges and limitations of utilising M&S-based research methods. Therefore, student meetings with experts in M&S-based research methods were organised as part of the concluding session of the third cycle of the M&S-based research method module. Progressing through the module, students examine how researchers use M&S-based research methods for the scientific study of religion to generate new insights and new tools that could inform the formulation of more effective policies for reducing religious radicalisation, violence, and extremism (i.e., Lane 2013; Shults 2018). Further, the module was intended to enable students to consider using M&S-based research methods in their future studies or even in their future careers. All participants were informed beforehand of the aim of the seminar, tutoring session, and meet-the-expert events. Ethical approval for this study was received from Sikt – Norwegian Agency for Shared Services in Education and Research.3
In this article, I provide an evaluation, based on the observation, of the interaction between the students and experts in the ‘meet-the-expert’ session, which was a significant part of the third iteration of the design cycle. This evaluation sought to answer a primary question: ‘How do students in religious studies utilise the opportunity to learn about M&S-based research methods?’ To answer this primary question, the following two questions were articulated: To what extent and how did students utilise the opportunity to (i) develop a sense of researchers’ motivation for using M&S-based research methods? (ii) expose their understanding of the usefulness, opportunities, limitations, and challenges of utilising M&S-based research methods?
The context and content of the M&S-based research method module
The University of Agder is a publicly funded institution located in southern Norway. A comprehensive portfolio of undergraduate and post-graduate programmes includes a three-year undergraduate religious studies programme that enrols approximately 30 students in each year cohort. The participants of this study were students in their second year in the religious studies undergraduate programme, and they were voluntarily invited to participate in the M&S-based research methods, which was an optional part of the curriculum to supplement the core course, ‘Religious radicalisation, extremism, and violence’ (REL-206). The programme's syllabus also incorporates modules to introduce recent theoretical and empirical studies of religious radicalisation, extremism and violence, which intends to develop students’ understanding of religious radicalisation, extremism and violence within Islam, Hinduism, and Christianity. The programme graduates find employment opportunities in community development as municipal consultants and social workers. In the fall semester of 2018, I implemented the third design iteration of the module to test the central hypothesis that students will take up the opportunity to learn about M&S-based research methods. The M&S-based research methods module was further developed to include a seminar and tutoring and meet-the-expert sessions to provide students with an experience of the research methods.
The introduction seminar
A seminar entitled ‘Research approaches to study social phenomena: Will simulations give insights?’ was prepared. It was purposefully designed around the theme of inclusion or exclusion (i.e., social segregation) because students in the religious studies programme are required to take a course in ‘Religious radicalisation, extremism, and violence’ (REL-206). The three-hour seminar was conducted in three parts. The first part was mainly a lecture-style presentation in which background information about conventional research methods, such as surveys, historical approaches and ethnography, was given. M&S-based research methods were introduced as an alternative research approach to studying social dynamics in which subjects are not actual people but virtual entities. In this way, M&S-based research methods allow for conducting studies using imaginary scenarios and thus eliminate any risks that can result from human participation.
The second part of the seminar sought to answer the questions, ‘What is a simulation?’ and ‘What is a model?’ In this part, students were given the opportunity to obtain hands-on experience with a social simulation applet (i.e., Schelling Applet) related to a theme of religious studies: the social inclusion and exclusion of people. The Schelling Applet was developed based on Schelling's Segregation Model (Schelling 1971); the animated applet is freely available at Stanford University's webpage, archived under Nifty Assignments.4 The representation by the computer simulation of Schelling's model (i.e., the virtual world) imitates a real-world phenomenon. It allows the exploration of complex social dynamics through changing social or community characteristics.
The third part of the seminar included a discussion guided by probing questions such as: ‘Why do researchers use simulation-based research methods?’, ‘What questions could be answered by creating a virtual Norway?’, ‘Will the ageing population affect tolerance in the community?’, ‘What are the assumptions and limitations of these methods?’ and ‘Are you interested in using M&S-based research methods in the future?’
After the seminar, students were asked to register suitable dates for the tutoring session, which was planned accordingly. The author assigned students different roles, depending on where they were in the sessions.
Tutor session
The tutoring session was designed as a small group session with a discussion of the opportunities and challenges related to M&S-based research methods and the possibility of helping students clarify the concepts associated with these methods. The session was also intended to help students who chose to write a short essay on M&S-based research methods. Each tutor session lasted about an hour, and these are some of the questions that guided the students’ reflection:
What do you see as the most promising aspects of these methods?
What do you see as the most challenging aspects of these methods?
How might this approach be applied to other contemporary social issues related to religious and social behaviour?
If mentorship were available to help you learn these methods, would you be interested in using it for your future research?
‘Meet-the-Expert’ event
The experts were members of the Social Simulation Research Group at the University of Agder. Expert 1 is a post-doctoral researcher, and expert 2 also taught the same group of students the course ‘Religious radicalisation, extremism, and violence’ (REL-206). Both utilise M&S methods to analyse social behaviour in their research practices. The setting of the expert meeting was a round-table discussion. Taking turns, all the students posed questions to the experts (M&S-based researchers) regarding the usefulness, opportunities, limitations and challenges of utilising M&S-based research methods. The researchers’ role was to take notes regarding the students’ questions, coordinate the meeting and respond to the students’ questions. The duration of the expert meeting was one hour.
A few examples of the students’ questions were: ‘Do you design the model? How do you collect data? How do you validate your model? Who can take advantage of the model you created? What does your model predict about religious and social peace?’ The experts’ strategies were to listen to the students’ questions and help them to understand how social science researchers utilise M&S methods in their practices. In so doing, they used examples from their daily lives. For example, a map is a model of geographical landscapes that is helpful in navigating places, and the map includes critical characteristics of the landscape's dynamics.
Essay about the M&S-based research methods module
The students were asked to write a short (three hundred word) essay, which was to be submitted along with the end-of-semester essay. The task was voluntary. They were encouraged to write the essay based on the knowledge they developed at the seminar, tutoring session and meet-the-expert session. The primary goal of the essay task was to assess how students utilised the opportunity to learn about M&S-based research methods.
Data collection and analysis
The audio recording of the students’ interactions with M&S-based researchers (i.e., experts) was transcribed. A thematic analysis approach (Braun et al. 2019) was used for the data analysis. By reading the transcripts thoroughly, I familiarised myself with the data, identified and grouped codes, developed themes, and revised and defined themes. These well-defined themes were used to present results, along with interpretations and relevant literature. Moreover, to maintain the trustworthiness of the analysis, I worked independently on the transcripts and developed tentative themes, then asked a critical friend to check them. This approach helped me to compare the outcomes and defined the themes through iterative cycles of construction of meaning and establishing a connection between the themes. Several themes were identified, including the usefulness of M&S-based research methods in understanding religious and social behaviour, the opportunities of utilising M&S-based tools, the lack of coding and programming knowledge limiting students’ understanding, and specific language or jargon used by experts being difficult for novices.
What did students gain from the round-table discussion with M&S-based researchers?
The students’ opportunity to learn about M&S-based research methods commenced with the teacher-centred seminar and moved towards student-centred learning opportunities. For example, the seminar was a mixture of a lecture and workshop that used teacher-designed materials, while the tutoring sessions comprised semi-guided activities that were intended to lead students towards knowing and understanding more about M&S-based research methods. In contrast to the previous iterations, the third cycle of the M&S-based research method module comprised specific design innovations, such as the meet-the-expert session. In this session, students interacted with experts (e.g., M&S-based researchers were positioned as knowledgeable others). The conversation allowed students to explore the relevance of M&S-based research and enquire about the practices of M&S-based professionals. Therefore, among these events, I chose to examine the students’ interaction with the experts, which provided a collaborative learning opportunity for both the experts and novices to explore a cutting-edge, future-oriented, relevant, and illuminating research methodology. The analysis of this event seeks to expose evidence of the extent to which students engaged meaningfully and with an understanding of the M&S-based research methods in the context of M&S-based professionals in their field.
In the meet-the-expert event, five students (pseudonyms: S3, S5, S9, S10, and S16) and two experts (pseudonyms: Expert 1 and Expert 2) participated in the discussion. The author structured the interaction between the students and experts to respond to the two research questions identified previously:
To what extent and how did students utilise the opportunity to:
develop a sense of the researchers’ motivation for using M&S-based research methods?
expose their understanding of the usefulness, opportunities, limitations, and challenges of utilising M&S-based research methods?
The motivation behind M&S-based methods
The students asked how experts used M&S-based research methods in their professional lives. The students’ questions and comments included, for example, the following:
So, when you make a model, what do you do after it's finished? Do you use it… or other people use it to write articles about it, then [write] books about it? [S5]
So, what's the coolest thing you learned… from the models? [S3]
The experts explained that every model they have built has had a specific purpose. Experts will publish a model to reach a broader audience when it is ready. Expert 1 responded by explaining that by using the models, he could manipulate the behaviour of individual agents to see the output: ‘You can come up with some…, any kind of behaviour if you can, so you can manipulate, so to say, individuals, and then just put some, some kind of behaviours and see what comes out. So that's, that for me is nice, so I really like that, I really like these models’. The students were trying to determine the experts’ motivation in adopting M&S-based research methods as well as obtaining an insider's perspective of practitioners of the methods. They learned through questioning that the M&S-based research methods required collaborative actions among the researchers, such as gathering data and publishing articles.
What strategies do researchers’ employ to make their models better?
The students were curious about the validity of the methods, specifically their validation process and what strategies the experts employed to tackle issues of validity. The students’ questions and comments from the interaction with the experts led to an understanding of M&S-based research methods. For example:
So, how well or is it will be like how well your model will work depends on how much empirical you have?… Yeah, so if we just do tons of empirical data could result in better research, no, your models will be… [S3]
Can I ask how you would make that… what approach would you take… for making the model closer to the reality? [S10]
Students showed their awareness regarding the quality of data sources as well as the efficiency and accuracy of models developed by experts. For example, student S10 asked what approaches the researchers used to reduce the gap between reality and their model. Her question indicated that when a model is remote from reality, then the model could misrepresent the social issues either by providing false recommendations or false predictions about future scenarios. Here is evidence that the student showed an understanding that developing models required some measure of trustworthiness. Imagining a situation in which M&S methods were applied in their field, they enquired how researchers would deal with the issue of making a model close to real-life societal problems. Expert 1 explained that models were never entirely representative of reality. According to Expert 2, researchers could implement some processes to validate models, such as observing real-life situations and revising the model accordingly.
Another student posed a question to understand how the researchers’ models were better than other models. Expert 2 explained that their models were better than other models in two ways. First, the researchers use multi-agent artificial intelligence models, which are well-validated. Multi-agent artificial intelligence systems are composed of multiple interacting agents in which they share an environment to attain common or conflicting goals. Second, they generate cognitively realistic agents, which represent human behaviour better. The multi-agent artificial intelligence systems architecture has features such as the capacity to enable the coupling of natural systems and human behaviour, utilise diverse data in the validation and verification process, and inform policy planning and analysis through forecasting scenario experiments (Lane 2013; Shults and Wildman 2020). Therefore they ‘are the most appropriate computational tool for the study of complex social phenomena’ (Lane 2013: 161). Expert 2 further clarified that ‘humans are not that simple. They have biases, attitudes, and networks that shape them. So,… our agents are much more cognitively realistic’. The students probed the experts about approaches that could improve their models. They posed questions to the experts while imagining themselves as future researchers because the students imagined problem situations in which the M&S-based research methods could be applied in their field. They advanced their knowledge of M&S-based research methods by developing insights about the accuracy of the models and strategies to ensure the model represented reality and provided better explanations of social dynamics.
How are M&S-based research methods useful in understanding religious social behaviour?
Students were motivated to understand how M&S-based research methods enabled experts to study social behaviour. While the experts’ explanation was connected to students’ compulsory course REL 206 (Religious radicalisation, extremism, and violence), they were comfortable asking follow-up questions. The following are a few selected questions and comments that indicated their curiosity:
There are also, erm, er, did the model show that erm, if that happens, er, we'll erm, will it [religion] in decrease quicker? Or like, more naturalism5 [Expert: ‘Right’.] leads to less relatedness, ah, I don't know… (so the model… ) [S3]
How many years did you say? [S3]
The students’ questions were consistent regarding understanding the usefulness of the research methods to understand religious and social behaviour. Student S3 was interested if the experts had developed a model that could explore many different scenarios related to religious and social behaviour, such as what happens if people behave less religiously or follow naturalism. She demonstrated her understanding of M&S-based research methods that allowed researchers to run virtual experiments asking various hypothetical questions. Her use of phrases, such as ‘if that happens’ and ‘if religion… increases or decreases’, are evidence of her engagement in ‘what-if thinking’ that characterises the use of M&S-based research methods.
Students identified that M&S-based research methods could be helpful to them in understanding changing religious and social traditions in Norway. The following excerpt depicts the exchange of ideas between the novices and experts in realising a need for more research to develop models that describe under which conditions traditional religious beliefs would decrease or new age religion would increase:
S9: OK. So, the model doesn't show like new religious tendencies, and stuff like that?
Expert 2: Or, or, or less attendance in churches and anything to do with that, it only has to do with whether people believe in, say spirits or the Holy Spirit or angels or stuff like that. That's going down in the population, and it should continue to go down if those conditions hold.
S9: So, you will get more new age people? Correct?
Expert 2: Well, no, new age, new age is included in that.
S9: I was just thinking that naturalism is growing in Norway and so is the sort of new spirituality in (indistinct) with traditional religions losing their following.
Student S9 showed her concern about the new trends in religion and spirituality as signs of disappearing traditional beliefs. ‘Naturalists typically resist… supernatural agents as part of their explanations of the causal structure of the world’ (Shults et al. 2018a: 221). Further, student S5 posed questions about a newly published model by the team of experts on people's behaviour regarding supernaturalism and naturalism. The student's enquiry related to peace. The student asked, ‘What… what will happen down there… I mean more peace or not?’ She was curious about whether expert-developed models could explain future scenarios regarding religious and social peace. Expert 2 explained that their models did not make any claim about peace but did determine that there would be more naturalism. The excerpts indicate that the student was wondering whether the models developed by the experts could run experiments under several imaginary conditions to help understand the relationships between religiosity, naturalism and supernaturalism. More so, their curiosity led to ‘what-if thinking’, which, in turn, led to hypothetical questions to understand causal reasoning while dealing with religious and social behaviour. These examples revealed that the students imagined themselves as future professionals in their field; they imagined a sense of the problem in their field in which M&S methods could be applied.
Opportunities for utilising M&S-based tools
The students identified not only the research potential of the M&S-based research methods but also stated the power of the methods in developing individualised approaches to tackle social issues. The following excerpt encapsulates the exchange of ideas between the novices and experts in co-constructing as well as advancing meaning, which led the novices and experts to envisage a unique feature of M&S-based research methods:
S16: I wonder, er, is it possible to make a model, er that's, where it takes a specific person's, er, I don't know, data to put into a simulation, see how, figure out the way, for example, to rehab… rehabil… rehabilitate the person from, for example, drug addiction, and then you know exactly what the best way to, to help this person, this specific person?
Expert 2: Great idea!
S16: Picking up, specifically about, for example, criminals, that instead of punishment we can have like everyone to do go through a system and when they come out, they will be good citizens instead of wasting time just locking, locking them up.
Student S16 was wondering if M&S-based research methods could create opportunities to develop an individualised treatment to rehabilitate a person with drug addiction. Expert 2 appreciated his ideas and described ‘many statistical and empirical studies that describe: the different types of people, acceptability of criminality or drug addiction, and lots of factor analysis correlationally that connects to those things’. The student showed his agency in imagining that M&S-based tools could enable him to develop an individualised treatment plan determined by a range of personal characteristics of the individual in need.
Limitations and challenges in utilising M&S-based tools
Lack of coding and programming knowledge limits students’ understanding
The students showed their interest in utilising models published by the experts. Student S5 mentioned, ‘But they are not for us to use because we do not understand them. I don't get it’. The models developed by the experts are difficult for them to understand. As users, students expect a user-friendly model that they can change into handy visualisations to help them understand the abstract concepts (i.e., radicalisation, extremism) that the teachers were explaining in their regular course REL 206 (Religious radicalisation, extremism, and violence). Further, student S5 said, ‘I wanna see when it's done, you know, not the programming, that does not interest me. Aah, it's what I can do with the model after that interests me’. She considered herself a layperson who cannot utilise models developed by experts due to her lack of knowledge of coding and programming. It seemed the students wanted a visualisation tool, such as the Schelling Applet.
Student S5 cannot see herself as part of M&S-based research practice. That is why she could not experience the expert-designed models that might lead them to become part of the realisation. It showed that the M&S-based tools were as not useful to her as to her colleague student S9. Hence, a lack of coding knowledge prevented them from understanding the expert-designed models and exploring them further. Student S5 mentioned that the use of codes in the case of models was somewhat equivalent to the coding or programming skills to develop a web page. In the past, while creating a webpage, someone had to know how to write (HTML) codes. However, nowadays, no one needs to know about coding because user-friendly versions of webpage programming are easily accessible. She believed both the researchers and her colleagues who understood and introduced the webpage analogy as a reification to come to the realisation that the issue of codes is part of simulations and models.
Researchers utilise codes to describe the assumptions behind the models, and they use specially designed software (i.e., AnyLogic) to run simulations of their model (The AnyLogic Company 2015). For instance, in the agent-based model, researchers analyse the behaviour of individual entities, referred to as agents, who engage with each other within a simulated environment (Shults et al. 2018b). In this environment, these agents can interact with not only themselves but also with variables in simulation. In their article, Shults et al. (2018b) also used AnyLogic software to run the simulation file. In the technical details, they recommend downloading the AnyLogic software, which requires basic knowledge of the function of the given codes to interpret the simulation and visualisation. For them, coding in the model illustrates how behaviour might change and can influence behaviour. They could also adjust the code to the parameters of the simulation models and explore the possible consequences. In contrast, students’ activities prepare them for work as municipal consultants or social workers. For student S5, simulation was instrumental in the visualisation of complex societal issues, such as segregation, inclusion or exclusion. In her understanding, simulation existed in computer codes that were not useful for her to see visualisations, and laypeople could not engage in the coding behind the models.
Simulation is a researcher's tool to understand hidden causal mechanisms and develop theories, but all students could not establish such relevance in relation to these purposes. Hence, the author identified two interacting activities: the activity of university students and the activity of researchers. The author perceived tension between the two activities because the students do not have the knowledge of coding and programming language to understand the expert-designed models thus hindering them from exploring them further. In this respect, the lack of coding and programming knowledge delayed their participation in future practices, such as becoming M&S-based professionals in their field.
Specific language or jargon used by experts is difficult for novices
The students experienced an obstacle when the experts described the M&S-based research processes utilising their tools, routines and jargon or language. The following are the few selected questions and comments that indicated the breakdown of the communication between the students and experts:
Okay… Okay. [S5]
I am sorry I don't understand all the words… aah… you are saying. [S5]
In one conversation, student S5 repeatedly said ‘Okay’ after listening to the experts’ explanations when using jargon, such as CSV files and nexus programmes. This suggests that the student was receiving information as if she were listening to terminology on programming languages without understanding the fuller meaning. However, in another conversation, students demonstrated their participation in developing follow-up questions. For instance, student S3 asked, ‘Yeah, okay, so, but have you found any indications whether your models are accurate or how accurate they are?’ Expert 2 simplified his explanation, stating a common phrase among the modelling community, ‘that every model is wrong, but that some models are useful’. Models are just imitations of phenomena that help researchers to study social dynamics by asking hypothetical questions. Expert 2 used an example of a map that can help a person to travel from Kristiansand to Oslo. The map may include the critical pathways that help the person get from one place to another without all the details, such as mountains, rivers, or trees. Similarly, computer simulations include the big things that researchers need to figure out critical characteristics for social dynamics that are like exploring the pathways to a destination in the examples mentioned above.
In this conversation, the expert used simplified language to explain models with everyday examples. As a result, the students were engaged in the interaction and asked follow-up questions. For instance, student S3 asked, ‘Yeah. What makes your model better than other models?’ Here is evidence that whenever experts use simplified language to explain M&S-based methods, the students showed a better understanding of issues by asking clarifying questions. Consequently, it becomes clearer that abstract language, codes and jargon can hinder students’ learning about M&S-based research methods.
Conclusions
This study reports the value of the meet-the-expert event, a segment of an innovative curriculum called the ‘M&S-based research methods module’, in providing opportunities for students to learn about M&S-based research. The results show the students’ growing awareness of the research methods, their background knowledge about the nature of research methods, researchers’ motivation for adopting such methods, their usefulness and the opportunities, limitations, and challenges of using the methods in their field. Furthermore, they utilised the opportunity of the meet-the-expert event by aligning it to their future professional goals and becoming more knowledgeable about M&S-based research methods.
Opportunities for studying modelling and simulation-based educational tools that are firmly rooted in the curriculum are not common in social science study programmes (Demerath et al. 2020; Holter and Schwesinger 2020; Szczepanska et al. 2020). I found that the majority of the students appreciated these opportunities. During the meeting with the experts, the students showed interest in M&S-based research methods in several ways, including active participation by asking questions about the use of M&S-based research methods practice, asking hypothetical questions about religious and social conflicts, the trustworthiness of the expert-designed models and the usefulness of M&S-based tools. The students developed a sense of the social science researchers’ motivation for using M&S-based research methods in their field through meaningful, rational and systematic enquiry. Overall, the students in the religious studies programme utilised the opportunities well.
Modules such as this one can encourage greater creativity by learners, both within the module and by taking advantage of other opportunities to engage in emerging, cutting-edge and relevant research approaches. The benefits to students in religious studies utilising M&S-based tools to experiment with religious and social behaviour are exciting! The evidence gathered in this study can be used to influence how social science research methodology courses are taught and encourage collaborative learning opportunities between students and experts. In addition, the findings of this study highlighted that future M&S-based curriculum modules can be enriched by incorporating instructional strategies to support students in developing background knowledge about computer programming and in understanding the specialist language used by the experts.
Acknowledgements
I am very grateful to the reviewers and editors of this journal for their thoughtful comments and reviews that assisted in developing and strengthening this article. I acknowledge the support from the Social Simulation Research Group, Faculty of Social Sciences, University of Agder. I want to thank Prof. Dr Pauline Vos and Prof. Dr F. LeRon Shults for their comments and feedback on the earlier version of this article. I would also like to thank Dr Ivan Puga Gonzalez and the participants for cooperating in this research.
Notes
In the European Credit Transfer System, 60 ECTS points are equivalent to one year of full-time study.
The Social Simulation Research Group at the University of Agder collaborates with a network of international scholars who utilise computer modelling and social simulation techniques in researching cultural conflicts, migration, and climate change (https://www.uia.no/english/research/research-groups/social-sciences/social-simulation.html).
http://nifty.stanford.edu/2014/mccown-schelling-model-segregation/. The webpage introduces links to the assignment materials ready for adoption or review by other educators.
Naturalism is a belief that avoids incorporating supernatural elements into its interpretations of the world's causal framework in scholarly settings (Shults et al. 2018a: 221).
References
Alvarez, R. M. (2016), Computational Social Science: Discovery and Prediction (New York: Cambridge University Press).
The AnyLogic Company (2015), Anylogic professional 7.2 (St Petersburg, Russia: The AnyLogic Company).
Bakker, A. (2018), Design Research in Education: A Practical Guide for Early Career Researchers (New York: Routledge).
Boshuijzen-van Burken C., R. J. Gore, F. Dignum, L. Royakkers, P. Wozny and L. Shults (2020), ‘Agent-Based Modelling of Values: The Case of Value Sensitive Design for Refugee Logistics’, Journal of Artificial Societies and Social Simulation 23, no. 4: 1–20. https://doi.org/10.18564/jasss.4411.
Braun, V., V. Clarke, N. Hayfield and G. Terry (2019), ‘Thematic analysis`, in P. Liamputtong (ed.), Handbook of Research Methods in Health Social Sciences (Singapore: Springer), 843–860.
Demerath, L., J. Reid and E. D. Suarez (2020), ‘Teaching about the social construction of reality using a model of information processing’, in V. V. Krzhizhanovskaya, G. Závodszky, M. H. Lees, J. J. Dongarra, P. M. A. Sloot, S. Brissos and J. Teixeira (eds), Computational Science – ICCS 2020 (Cham, Switzerland: Springer Nature), 648–660.
Engeström, Y. (2005), ‘Knotworking to create collaborative intentionality capital in fluid organizational fields’, in M. M. Beyerlein, S. T. Beyerlein and F. A. Kennedy (eds), Collaborative Capital: Creating Intangible Value (Advances in interdisciplinary studies of work teams) (Amsterdam: Elsevier), Vol. 11, 307–336.
Gibson, D. C. and D. Ifenthaler (2017), ‘Preparing the next generation of education researchers for big data in higher education’, in B. Kei Daniel (ed.), Big Data and Learning Analytics in Higher Education (Cham, Switzerland: Springer), 29–42.
Gilbert, N. (1999), ‘Simulation: A new way of doing social science’, American Behavioral Scientist 42, no. 10: 1485–1487. https://doi.org/10.1177/0002764299042010.
Gilbert, N. and R. Conte (eds.) (1995), Artificial Societies: The Computer Simulation of Social Life (London: UCL Press).
Gilbert, N. and K. G. Troitzsch (2005), Simulation for the Social Scientist (London: Open University Press).
Gore, R., C. Lemos, F. L. Shults and W. J. Wildman (2018), ‘Forecasting changes in religiosity and existential security with an agent-based model’, Journal of Artificial Societies and Social Simulation 21, no. 1: 1–31. https://doi.org/10.18564/jasss.3596.
Griffiths, R. (2004), ‘Knowledge production and the research-teaching nexus: the case of the built environment disciplines’, Studies in Higher Education 29, no. 6: 709–726. https://doi.org/10.1080/0307507042000287212.
Hassan, S., L. Antunes and J. Pavon (2010), ‘Mentat: A data-driven agent-based simulation of social values’ evolution’, in G. Di Tosto and H. Van Dyke Parunak (eds), Multi-Agent-Based Simulation X, vol. 5683 of Lecture Notes in Artificial Intelligence, 135–146.
Healey, M. (2005), ‘Linking research and teaching: Exploring disciplinary spaces and the role of inquiry-based learning’, in R. Barnett (ed.), Reshaping the University: New Relationships between Research, Scholarship and Teaching (Maidenhead: McGraw-Hill/Open University Press), 67–78.
Held, F. and I. Wilkinson (2018), ‘Computer simulation and agent-based models as a research method’, in P. V. Freytag and L. Young (eds), Collaborative Research Design: Working with Business for Meaningful Findings (Gateway East, Singapore: Springer), 377–398.
Hmelo, C. E., D. L. Holton and J. L. Kolodner (2000), ‘Designing to learn about complex systems’, Journal of the Learning Sciences 9, no. 3: 247–298. https://doi.org/10.1207/S15327809JLS0903_2.
Holter, E. and S. Schwesinger (2020), ‘Modelling and simulation to teach (classical) archaeology: Integrating new media into the curriculum’, in S. Hageneuer (ed.), Communicating the Past in the Digital Age: Proceedings of the International Conference on Digital Methods in Teaching and Learning in Archaeology (12–13 October 2018) (London: Ubiquity Press Ltd.), 167–177.
Hsu, Y.-S., L.-F. Lin, H.-K. Wu, D.-Y. Lee, and F.-K. Hwang (2012), ‘A novice-expert study of modeling skills and knowledge structures about air quality’, Journal of Science Education and Technology 21, no. 5: 588–606. https://doi.org/10.1007/s10956-011-9349-5.
Kilburn, D., M. Nind and R. Wiles (2014), ‘Learning as researchers and teachers: The development of a pedagogical culture for social science research methods?’, British Journal of Educational Studies 62, no. 2: 191–207. https://doi:10.1080/00071005.2014.918576.
Lane, J. E. (2013), ‘Method, theory, and multi-agent artificial intelligence: Creating computer models of complex social interaction’, Journal for the Cognitive Science of Religion 1, no. 2: 161–180. https://doi:10.1558/jcsr.v1i2.161.
Lane, J. E. and F. L. Shults (2020), ‘The computational science of religion’, Journal for the Cognitive Science of Religion 6, nos. 1–2. https://doi.org/10.1558/jcsr.38669.
Lave, C. A. and J. G. March (1975), An Introduction to Models in the Social Sciences (New York: Harper & Row).
Lave, J. and E. Wenger (1991), Situated Learning: Legitimate Peripheral Participation (Cambridge, UK: Cambridge University Press).
Lehrer, R. and L. Schauble (2012), ‘Seeding evolutionary thinking by engaging children in modeling its foundations’, Science Education 96, no. 4: 701–724. https://doi.org/10.1002/sce.20475.
Lovell, D. (2019), ‘Promoting original research: Designing a course to produce quality undergraduate political science research in the community college setting’, Journal of Political Science Education 15, no. 3: 392–398. https://doi.org/10.1080/15512169.2018.1487304.
McKenney, S. and T. C. Reeves (2012), Conducting Educational Design Research (New York, NY: Routledge).
Meld. St. 16 (2016–2017), Quality Culture in Higher Education. White Paper from the Norwegian Ministry of Education and Research, https://www.regjeringen.no/en/dokumenter/meld.-st.-16-20162017/id2536007/.
O'Neill, D. K. (2016), ‘Understanding design research–practice partnerships in context and time: Why learning sciences scholars should learn from cultural-historical activity theory approaches to design-based research’, Journal of the Learning Sciences 25, no. 4: 497–502. https://doi.org/10.1080/10508406.2016.1226835.
Plomp, T. (2013), ‘Educational design research: An introduction’, in T. Plomp and N. Nieveen (eds), Educational Design Research, (Enschede, the Netherlands: Netherland Institute for Curriculum Development [SLO]), 10–51.
Poudel, A. B. (2021), ‘An exploration of how social science students utilise an opportunity to learn about simulation-based research methods: A design-based study’, PhD thesis, University of Agder, Kristiansand, Norway.
Poudel, A. B., P. Vos and F. L. Shults (2020), ‘Students of religion studying social conflict through simulation and modelling: An exploration’, in H. Verhagen, M. Borit, G. Bravo and N. Wijermans (eds), Advances in Social Simulation: Looking in the Mirror (Cham, Switzerland: Springer), 379–383.
Schelling, T. C. (1971), ‘Dynamic models of segregation’, Journal of Mathematical Sociology 1, no. 2: 143–186. https://doi.org/10.1080/0022250X.1971.9989794.
Sengupta, P., A. Dickes, A. V. Farris, A. Karan, D. Martin and M. Wright (2015), ‘Education programming in K–12 science classrooms’, Communications of the ACM 58, no. 11: 33–35.
Shults, F. L. (2018), ‘Can we predict and prevent radicalization?’ in G. Øverland, A. J. Andersen, K. E. Førde, K. Grødum and J. Salomonsen (eds), Violent Extremism in the 21st Century: International Perspectives (Cambridge: Cambridge Scholars Press), 45–71.
Shults, F. L. (2019), ‘Modeling metaphysics: The rise of simulation and the reversal of Platonism’, paper presented at the Simulation Conference, Society for Modeling & Simulation International, 29 April–2 May 2019, Tucson, AZ, USA.
Shults, F. L., R. Gore, C. Lemos and W. J. Wildman (2018a), ‘Why do the godless prosper? Modeling the cognitive and coalitional mechanisms that promote atheism’, Psychology of Religion and Spirituality 10, no. 3: 218–228. https://doi.org/10.1037/rel0000198.
Shults, F. L., J. E. Lane, W. J. Wildman, S. Diallo, C. J. Lynch and R. Gore (2018b), ‘Modelling terror management theory: Computer simulations of the impact of mortality salience on religiosity’, Religion, Brain & Behavior 8, no. 1: 77–100. https://doi.org/10.1080/2153599X.2016.1238846.
Shults, F. L. and W. J. Wildman (2020), ‘Artificial social ethics: simulating culture, conflict, and cooperation’, in H. Verhagen, G. Bravo, M. Borit and N. Wijermans (eds), Proceedings of the SpringSim 2020 conference 18–21 May 2020, Fairfax, VA, USA, 1–10.
Szczepanska, T., M. Priebe and T. Schröder (2020), ‘Teaching the complexity of urban systems with participatory social simulation’, in H. Verhagen, M. Borit, G. Bravo and N. Wijermans (eds), Advances in Social Simulation: Looking in the Mirror (Cham, Switzerland: Springer Nature), 427–439.
Taillandier, P., D.-A. Vo, E. Amouroux and A. Drogoul (2012), ‘GAMA: A simulation platform that integrates geographical information data, agent-based modeling and multi-scale control’, in N. Desai, A. Liu and Michael Winikoff (eds), Principles and Practice of Multi-Agent Systems. PRIMA 2010 Lecture Notes in Computer Science, vol. 7057 (Berlin, Heidelberg: Springer), 242–258.
Van den Akker, J., B. Bannan, A. E. Kelly, N. Nieveen and T. Plomp (2013), Educational Design Research, Part A: An Introduction (Enschede, Netherlands: Netherland Institute for Curriculum Development [SLO]).
Van den Akker, J., K. Gravemeijer, S. McKenney and N. Nieveen (eds) (2006), Educational Design Research (London: Routledge).
Wildman, W. J., P. A. Fishwick and F. L. Shults (2017), ‘Teaching at the intersection of simulation and the humanities’, in W. K. V. Chan, A. D'Ambrogio, G. Zacharewicz, N. Mustafee, G. Wainer and E. Page (eds), Proceedings of the 2017 Winter Simulation Conference, Las Vegas, NV, USA, 3–6 December (Las Vegas: Society for Modeling & Simulation International), 1–13.
Wilensky, U. and W. Rand (2015), An Introduction to Agent-based Modeling: Modeling Natural, Social, and Engineered Complex Systems with NetLogo (London: MIT Press).
Wu, H.-K. (2010), ‘Modelling a complex system: Using novice-expert analysis for developing an effective technology-enhanced learning environment’, International Journal of Science Education 32, no. 2: 195–219. https://doi.org/10.1080/09500690802478077.