Student success originates from a complex system One way of understanding this, is by thinking of student success as something originating from a complex system. The complex model of student retention presented in this post represents a first attempt to use contemporary methods from Complexity Theory to model a larger educational system, and to simulate effects in such a system.

A complex system is an interconnected system that changes and adapts as time goes on. The effect of changes in such a system are hard to predict as seemingly large changes in this system may have very small effects, while small changes in a complex system may – potentially – have cascading effects that lead to large and unanticipated effects in the system. This makes it very difficult to attain effects that can be retraced to simple, often single variable, interventions.

Drawbacks of linear modelling Understanding the parts constituting a complex system and how these parts are related is critical for understanding how and what can be done to influence and improve the educational system. More research fields begin to recognize the drawbacks of linear modelling, in which a direct link between a small number of causes and a single effect is assumed. There is a movement in educational research towards viewing education as a complex system (for example, Stephens & Richey 2013; Lemke & Sabelli, 2008).

A new and exciting field of educational research This new and exciting field of educational research offers another way of thinking of an age old problem – how to inform improvement of education. It is a way of thinking about the educational system not as a stable and predictable entity, but as a diverse and unpredictable one. It offers ways of thinking of this system that is, in a way, more close to the very nature of the educational system itself.

Example of such research. The focus of this specific study was to explore what possibilities universities have to reduce student drop out. The approach followed is to create and simulate an empirical complex model of student retention. This model was used to estimate effects of changes to this system, as a result of potential interventions to increase student retention (Forsman, et al, 2014a, 2014b; 2014c; 2014d). For this study we worked with a data set consisting of 78 variables deemed relevant for first year student retention in a university of technology. The target variable was the number of credits obtained by the students in the sample and it served as a prerequisite for student retention. The complex model of student retention is presented in Figure 1. The model consists of dots that are referred to as ‘nodes’ and links between them.

  • The red node represents the number of credits obtained.
  • The blue nodes represent variables that can be affected by institutional action and policy making, such as teachers’ behaviours, scheduling, etc.
  • The grey nodes are nodes which can be influenced indirectly, such as student study behaviour, atmosphere, etc..
  • The black nodes are those variables that are constant, such as parents’ education, previous grade achievements, etc.

All effects on students’ credits achieved caused by changes to the system are through indirect effects originating from other aspects of students’ experiences of their education The grey links indicate a positive relationship, while the red links are the negative relationships. The thickness of the links indicates the strength of the relationship.

The model in Figure 1 is based on the analysis of real data: there has been no input from any theories, other than the theories that served to select variables included in the dataset.