In order to make deliberate choices about their education, students need to have information about the options that are available to them, about which of these might be appropriate for them given the previous choices they have made, and about potential consequences of making certain choices. It is only by having such information can one make an informed choice. On the one hand, these decisions might be related to which courses they should take, which graduate programs they could apply for, or which careers might be suitable for them. On the other hand, they might also relate to a more general issue, such as about where to focus their attention in their future development. In considering such matters, one might ask what others in a relatively similar situation have done, and what the consequences of those choices have been. This type of information is relevant, as it gives some insight into what might be appropriate options and what their possible consequences are.
Guided pathways are tools that provide such insights by producing personalized recommendations and tailored information relevant to a student’s situation which are based on an analysis of the choices others have made and their consequences. By using methods from data science and learning analytics, it is possible to create an online system that takes input provided by a student, such as the courses they have taken or their interests, and compares that input to a database which contains information about a large number of students who have faced a similar choice. The system looks at the extent to which the input provided matches students in the database, and presents information about the choices these other students have made, thereby generating a personalized recommendation. Effectively, such a system can tell a student what relevantly similar students have done in the past. It is comparable to systems that are used in many online shops where recommendations for future purchases are based on what one has previously bought and compares it to other people’s purchasing history.
These systems can help students gain an understanding of what might be interesting options for them as well as their outcomes during and after their study program. For example, it generates information about what courses students with similar curricula or interests have chosen which is helpful in order to arrive at a decision for future courses. Additionally, it provides information about what other students with similar profiles have chosen regarding their graduate programs and future careers. As these systems can be based on significant amounts of data, they can provide insight and options that students and their advisors may not have thought of on their own. Moreover, students can use these systems to get a sense of the possible consequences of certain choices, by varying their input into the system based on options they are considering, and seeing how this affects the recommendations and information the system gives them.
Needless to say, such systems come with a number of limitations, which are important to discuss. Recommendation systems are only as good as the data they are based on. Small sets of data, inaccurate data and the like can all yield skewed results. Moreover, students who might be similar in the data might actually be very dissimilar in reality. Two students might have made the same choice but done so for different reasons. Some students might regret choices they have made or have changed their mind since making that choice, which would not be represented in this system. These are some of the aspects that might make the recommendations and information only partially accurate or inaccurate. However, the system will present them anyway, and the students using them will have no way of knowing whether they can have confidence in the results. Hence it is important to place these systems in context, acknowledge their limitations, as well as clearly explain to both students and faculty members how they work. Despite the potential drawbacks, such systems can helpfully supplement other forms of advising, and provide input and suggestions for those conversations.
Two examples of self-advising tools that provide guided pathways for students, both developed at UCM, are discussed below. They are recommender systems that help students choose courses and UCM Compass, which suggests graduate programs that are appropriate for their interests and curricula.
For more information about guided pathways and their use, please see the bibliography.