Student approach to learning

Student approach to learning: An empirical investigation of factors associated with student approach to learning

Purpose/research question

To identify the key predictors for the students' choice of learning approach, including: background characteristics, institutional and contextual background variables, motivational factors, including the students' experience of the teaching environment, academic self-efficacy and exam angst. 


The universities are currently facing a pedagogical challenge. The students represent an increasingly heterogeneous group, on the one hand because of differing social backgrounds, and on the other because the wide range of entrance requirements implies greater diversity in the students' educational backgrounds. This development brings new challenges for the institutions, teachers and the universities' educational development units due to a need for studies that help to reduce drop-out rates, and conditions that stimulate and motivate the students to adopt appropriate learning practices.

Research shows that the students' learning outcome is linked to how they learn, and the way in which they learn is determined by their response to the learning environment in which they find themselves. Even though all teaching involves acquiring a body of factual information and new concepts, it should be remembered that rote learning is generally regarded as an unsatisfactory outcome of research-driven education at university level. In fact, studies show that this form of learning is generally associated with poor academic performance.

To support teachers in developing and implementing teaching methods that encourage and stimulate the desired learning efforts, it is necessary to investigate the factors that affect the students' approaches to learning.

Data and analyses

In total, 1,350 questionnaires were distributed at lectures and seminars to undergraduate students (third semester) and Master's degree students (seventh and ninth semesters) recruited from four faculties at Aarhus University (response rate: 88.3%).

The provisional reliability and validity of the Danish translation of R-SPQ-2F was examined using descriptive statistics, reliability analyses (internal reliability: Cronbach's Alpha) and confirmatory factor analysis. In the following studies, predictors for deep and superficial learning were analysed using non-corrected bivariate regression analyses and multiple hierarchical regression analyses with deep and superficial learning as dependent variables, adjusted for the other independent variables which were examined.


The subsequent analyses showed that the most important independent predictors for a higher degree of deep learning were the following: that the students were older, were female, had higher upper secondary school exam results, were self-motivated in their choice of study programme, studied at the Faculty of Arts or the School of Business and Social Sciences faculty and that the teaching took place in smaller groups.

The independent predictors for a higher degree of superficial learning included: younger age, lower upper secondary school exam results , studying at the Faculty of Health or Faculty of Science and Technology, lower average marks on the study programme and lectures as a teaching method.

The findings also showed that motivational factors (e.g. self-efficacy, test anxiety  and the experience and importance of the teaching environment) were strong independent predictors of the students' choice of learning approach.

Several of the findings proved to be in line with a number of previous results reported in the international research literature. For example, that deep learning was associated with higher age, higher qualifying upper secondary school exam results, current grade average on the study programme, being intrinsic-motivated in the choice of study programme, and  studying subjects associated with “soft” sciences such as the Arts

Moreover, the findings confirmed that a number of contextual factors, including teaching in small groups such as seminars or lectures combined with seminars, were associated with an increased tendency to use deep learning.

Even if the learning outcome depends largely on acquired knowledge and the ability to apply this knowledge, the study confirmed that motivational factors such as academic self-efficacy, test anxiety were important predictors of approach to learn, adding considerably to the explanatory power of the models . The findings also indicated that experiencing a learning environment which encourages problem-solving, scientific thinking and exam preparation in accordance with a deep learning-based approach to learning, was an independent predictor on an equal footing with self-efficacy.

Discussion og perspectives

Teaching and learning environments are not perceived in the same way by students, and both teachers and administrators on higher education degree programmes should be aware of the factors that either encourage or prevent students from becoming actively involved and developing their own individual understanding of the subject matter as well as the student's perceived ability to successfully handle a particular learning task. It is therefore an important task to develop and promote problem-solving learning activities, where the students have the opportunity to regulate their own learning efforts.

Status: finished

PhD defence june 2012

Berit Lassesen

Associate professor