Large classes in STEM education—potential and challenges

Although a universal phenomenon in higher education, large classes have typically been seen as detrimental to student learning (Ehrenberg et al 2001, Cuseo 2007). The reasons for this include a belief that they are associated with weak student engagement and fewer opportunities for interactions with the teacher and with each other. Certainly large classes that are not carefully designed may create a 'transactional distance' (see chapter 2), that is, a psychological and communications space between student and teacher which hinders learning, leading them to be experienced as impersonal, with a sense of isolation and anonymity. However, others believe that with the use of active learning pedagogies, large classes can provide stimulating learning environments which lead to critical thinking and deep learning (Hornsby and Osman 2014). Despite the worry about the potential downsides of large classes, undergraduate science classes in the UK commonly have anywhere between 50 and 300 (or more) students, and in early and foundational years, numbers are usually at the upper end of this range. This situation is not likely to change as the move towards the massification of higher education has strong financial and political (Hornsby and Osman 2014) drivers.

In this chapter, as in the rest of this volume, we purposely use the term 'large class' rather than 'lecture' to discuss the learning environment encountered by STEM students. This is because a 'lecture' conjures up a particular type of pedagogy—specifically one in which a person (the lecturer) talks for the majority of the time, and students listen and take notes. In contrast the term 'large class' is free from any pedagogical preconceptions and may, therefore, take any number of forms, many of which are discussed in later chapters. Similarly the term 'teacher' is used in place of 'lecturer'.

Although there are huge challenges associated with providing high quality learning in large classes, it is our belief, as evidenced by the contributions in this volume, that it is possible, through the introduction of research informed active learning approaches, to create a good quality, stimulating, learning environment which leads to deep learning, even in STEM classes with diverse cohorts and large numbers of students. This chapter will explore how to define large classes, and discuss their challenges and potential benefits.

There is no agreed definition of a large class although various different sizes have been proposed; Winke and Rawal (2018) suggest a large class is 35 or more students and Cuseo (2007) reports it is common for classes of over 50 to be seen as large. In contrast Jungic et al (2006) define a large class as having more than 350 students and Cash et al (2017) found from a survey of students and teachers that a class was felt to be large at around 240 students. Rather than defining a large class by a numerical threshold, a more useful measure would be to consider how the number of students impacts on their experience of the class (Hornsby et al 2013). Cash et al report that students described a large class as one that is impersonal and anonymous and which has an atmosphere 'where neither their peers nor their instructor noticed whether they were absent or attentive' (Cash et al 2017). In contrast the students described a small class as a more personal learning experience which created a sense of community and where the pace of the learning could be adjusted in response to students' needs.

The factors which influence how large or small a class feels include the teaching strategies employed, how much interaction takes place, and the extent to which students feel actively engaged with their learning. Another important aspect which affects the feel of the class is the teacher to student ratio. In the UK it is common for a large class to be taught by a single teacher alone whereas in other countries such as the US the teacher is supported by teaching assistants, giving a higher teacher to student ratio. This latter scenario is more likely to make students feel visible and involved, indeed Lenton (2015) found that the staff student–ratio is a predictor of student satisfaction.

The teaching strategies employed also make a significant difference to how large or small the class feels, however it is important to note that smaller classes often use similar pedagogical approaches to larger classes. For example there is evidence from a survey of 2000 courses in North America that smaller class sizes do not necessarily result in an increase in instructional innovation and student-centred approaches (Stains et al 2018). This may be because teachers do not change their pedagogical approach when moving from large to small classes (Wright et al 2019). This can happen for a variety of reasons such as the demands of the physical space, highly structured curricula, and lack of pedagogical skills to facilitate discussions.

However, the size of the class does influence how much effort is required to create good quality interactions. For example, with a class of 20 students it is much more straightforward to generate whole class discussions where real dialogue takes place between the teacher and the students, and where most students have the opportunity to take an active part. In larger classes students can be put off from speaking in front of their peers, whereas in smaller classes students are likely to be familiar with each other and therefore more comfortable adding their voice to the discussion.

In contrast, large classes are often taught through direct instruction, where the teacher uses a continuous monologue to introduce new ideas while the students listen and make notes, with very little interaction. It is much harder to introduce whole class discussions in this context, and when that is attempted, most of the students are not directly involved in the dialogue. This style of teaching is likely to be experienced as a large class regardless of the number of students in the class. Indeed Cash et al(2017) found that 42% of student respondents felt that a numerically large class can feel small if the instructor was engaging, if they made them feel like individuals and if they incorporated small-group activities which enabled them to connect with their peers.

One of the key drivers of large classes in post-secondary STEM education is the 'massification' of higher education (Hornsby and Osman 2014), a term used to describe the rapid increase in student numbers that took place at the end of the 20th century, and which has continued into the 21st century (Scott 1995). Massification is a global phenomenon, purposely pursued with the aim of improving health, empowerment and economic development (Bloom et al 2006, Santiago 2008). It has been argued that increasing enrolment in higher education is a consequence of an increased sense of social justice and fits with the democratization of education (Altbach 1992). Massification has led to increased class sizes, but it is worth pointing out that this is not a necessary outcome, rather, a consequence of the lack of a corresponding increase in financial (and therefore human) support from public sources in the higher education sector (Mohamedbhai 2008). The increase in class sizes has resulted in concerns that academic standards will slip, as, it is argued, while large classes are suited to information transfer, they are not suited to the acquisition of disciplinary knowledge, because this requires contact between lecturers and students, which is limited in large classes (Allais 2014). Yet some researchers, such as Arvanitakis (2014) discuss the benefits of massification, arguing that it offers the potential to 'promote the emergence of citizen scholars in sections of the population that have for far too long been ignored', and that the challenges of large classes can be overcome by pedagogical approaches.

While large classes are likely to be a feature for at least the foreseeable future, it is worth considering the arguments both for and against their use. One of the most influential works concerning the utility of lectures is 'What's the Use of Lectures' by Bligh (1998). One quote in particular notes that, while the lecture 'may be used appropriately to convey information ... it cannot be used effectively on its own to promote thought or to change and develop attitudes without variations in the usual techniques' (p 13). This quote is often used as an argument to show that there is no value in the lecture as it does not encourage thought or inspire students. However, in reality, Bligh's discussion is more nuanced than this: rather than arguing against all lectures, he is instead saying that the lecture should be given less prominence in the curriculum, and that other learning experiences should be seen as important. Indeed as we will see in this volume, modern STEM courses will often combine lectures with workshops, tutorials and laboratory experiments to provide a holistic and varied learning experience. Furthermore, the final part of the quote 'variations in the usual techniques' is particularly relevant, implying that techniques can be employed which will promote thought, change attitudes and support the development of critical thinking. These techniques are a key feature of this volume.

However, discussions in the popular press have recently been particularly polarized with a strong movement both for and against the use of lectures, for example articles discussing how the 'death of the lecture' is imminent (e.g., Campus Review 2019), are counterbalanced by articles singing the praise of lectures (e.g., The Guardian 2013).

The arguments against the lecture are mainly that they do not engage students, that by a few weeks into the semester, students no longer attend, that students who do come have a short attention span, and rather than listening to the lecturer, are likely to be using social media, and that the teaching method of 'stand and deliver' is not effective. In contrast those in favour of lectures cite the way that they provide meaning about the subject, synthesize important ideas, communicate passion and enthusiasm, create a community of learners and a common intellectual experience, demonstrate expert-like thinking and problem-solving, and provide academics with a new way of thinking about the subject. What pieces such as these tend to do however, is to conflate two things: class size and pedagogical approach. As we have discussed, large classes are likely to be here to stay, therefore if we can't change the number of students, we can at least change the pedagogical approach. Good pedagogical approaches can overcome the challenges cited, while also providing the advantages of large classes, such as creating a community and common experience for students (Lawrence 2022).

Although we might assume that large classes are always highly detrimental to students' learning, the literature on the subject is mixed, with some research showing a large negative impact on grades and high levels of attrition (Cuseo 2007, Kara et al 2021), others finding more minimal effects (Ake-Little et al 2020) and some demonstrating no negative effect at all on student learning and achievement (Gleason 2012). For example in a large study comparing STEM and non-STEM classes, STEM classes were found to have the biggest negative effect on grades with respect to class size (Kara et al 2021) and Cuseo (2007) reports that the drop-out rate for first year students was more than 25% at institutions with four year courses. In contrast, Gleason (2012) compared mathematics classes with around 30 students to those with over 100 students, with a similar curriculum, and found very little difference in outcomes, and a small positive effect for student satisfaction in the large classes. Gleason hypothesized that these findings were due to the extensive use of technology both in and out of class. Care needs to be taken when using student perceptions, as students tend to assign higher ratings to instructors and courses when the class size is smaller, even when accounting for the pedagogical approach (Wright et al 2019).

The conclusion from the literature is that large classes can be detrimental to student learning, achievement and completion rates, but that this doesn't need to be the case. With carefully thought out student-centred pedagogical approaches large classes can be just as successful as smaller classes and should not be feared as being second best.

1.5.1. Opportunities for interactions

1.5.2. Anonymity

1.5.3. Heterogeneous background of students

1.5.4. Assessment and feedback

1.6.1. Social aspects

1.6.2. Diversity

A systematic review of research into what influences the effectiveness of large classes by Jerez et al found five conditions which make large classes successful: (1) student–teacher and student–student interaction, (2) implementation of active learning strategies, (3) classroom management, (4) students' motivation and commitment, and, (5) the use of online teaching resources (Jerez et al 2021).

As discussed above introducing high quality and diverse types of interactivity is a challenge in large classes. One approach, which will be explored in detail in later chapters is to introduce small-group activities, through pedagogies such as Think-Pair-Share (Kothiyal et al 2013) and Peer Instruction (Mazur 1999). This gives students the opportunity to interact with their peers, to learn from each other, to discuss the material in depth and to practice making their thinking visible (Smith et al 2009, Wood et al 2014). Beyond the clear learning benefit, this type of approach also increases the opportunities for social interactions which reduce the feeling of isolation.

Introducing teacher–student interactions is harder, both because of the natural reticence of students to speak out in front of hundreds of their peers and because only a small number of students will ever get the chance to take an active role in the discussion—most will be passive observers of the interaction. Nevertheless, this approach has an important role to play in connecting the teacher with the student group and in creating a community of enquiry (Garrison 2009). It is also important for 'sense-making' and, as pointed out by Turpen and Finkelstein (2010) it can be particularly valuable if students are encouraged to explain their thinking for incorrect answers as well as for their correct answers.

The strategies discussed above can all be described as active learning approaches. Not only do they increase the opportunities for student–student and student–teacher interactions, they have been shown to have a range of benefits for students. These include improved performance on exams and concept inventories (Freeman et al 2014), reduced attrition rates (Lasry et al 2008) and helping to close the gap between under-represented minority students (URM) and well represented student groups (Ballen et al 2017).

Classroom management also plays an important role in large classes, and ultimately the success of active learning approaches depends on what is termed the 'idioculture' which can be described as 'a system of knowledge, beliefs, behaviours, and customs' (Finkelstein 2005) shared by all members of the group. When applied to a learning environment the idioculture affects the class climate, attitudes and approaches to and within a class, influencing the nature of the interactions between students and teacher. These class 'norms' can be set at the start of the course by making it clear to students what to expect in the classes (i.e., that questions are welcomed, that mistakes are learning opportunities), by modelling the behaviour that teachers want to see, and explaining to students the reasons behind a particular pedagogical approach.

Another way to support large classes is the provision of online teaching resources. These include in-class voting systems which encourage interaction such as Kahoot, Mentimeter and Socrative, out of class learning systems such as PeerWise (Bates et al 2011) which encourage students to create their own questions, STACK (see, e.g., Sangwin 2007) for solving mathematical problems with immediate feedback and Padlet (see, e.g., Ellis 2015) for opportunities to discuss and connect with other students. Online resources also include digital notes and lecture captures which, when used appropriately, can support students to take control of their learning (Wood et al 2021).

As discussed above, designing appropriate assessment and feedback opportunities is another challenge in large class teaching. The approaches already mentioned such as Think-Pair-Share and Peer Instruction provide instant feedback, both to the students about their level of (collective) understanding and to the teacher who can use that information to make decisions about class activities. In addition, large classes can benefit from the use of computer marked assessments, which can provide both formative and summative assessments in a less resource intense way, while also giving timely feedback to students (see chapter 6 for a detailed discussion of computer marked assessments).

The drive to increase the number of students attending higher education in recent decades, coupled with the proportional lack of investment in the number of teachers, means that large classes are likely to be a part of the higher education STEM landscape for the foreseeable future.

For large classes to be effective teachers need to be aware of both the challenges that exist and the way in which pedagogies such as active learning can offset those challenges. In this chapter we have discussed the difficulties of creating high quality student–student and student–teacher interactions in large classes, the potential for students to feel isolated and anonymous, the challenge of teaching classes with students from a mix of backgrounds, educational achievements and motivations, the difficulties of providing feedback to large numbers of students and of designing suitable assessments. We have also pointed to approaches, many of which will be fully explored in the following chapters, which can not only ameliorate these challenges but provide a high quality learning experience which encourages deep learning and critical thinking. These include pedagogies such as Peer Instruction which encourage student–student interactions and provides feedback to both the students and the teacher, and technology based strategies to provide online resources to students, to enable them to connect outside of class, to give instant feedback on their level of understanding and to provide effective assessments.

In conclusion, while this chapter has discussed many of the challenges of teaching, assessing and learning in large cohorts, we have also provided examples of ways in which large classes can not only overcome these challenges but can become high quality learning environments in their own right. Indeed as Hornsby and Osman (2014) observe, 'Large classes in and of themselves are no longer insurmountable obstacles in efforts to foster higher order cognitive skills and in the process to achieve innovation and knowledge based economies. Indeed they can offer as many opportunities as they do challenges'.