Skip to content

A wide range of literature has found that homework is positive for students’ academic achievement (see for example Murillo & Martínez-Garrido, 2013; Fernández-Alonso, Álvarez-Díaz, Suárez-Álvarez, & Muñiz, 2017) – a conclusion which is rooted on the assumption that children become more responsible when they perform it at the same time as they learn. However, in recent years students have been charged with high loads of homework, a practice which has been widely criticised by both parents and teachers (Inchley et al., 2016). This situation is so alarming that in 2016 Spanish parental associations fostered a ‘homework strike’ (see Jones, 2016; see also Cooper, Robinson, & Patall, 2006). This excessive amount of homework is not an exclusively Spanish problem: it is also present in other countries (see Associated Press, 2016; see also Zuzanek, 2008).

Because of this situation, our research work intends to shed further light on the association between homework and students’ academic performance. Our work uses census and longitudinal data from the Andalusia region of Spain. This data has been gathered by the Andalusian Education Assessment Agency to assess students’ competences in the 5th and 8th grades using cognitive tests and background questionnaires similar to those of the Programme for International Student Assessment (PISA). Andalusia is a particularity in that it is the most populated Spanish region, but also one of the worst performers in international large-scale assessment tests such as PISA.

Our study makes use of a twin fixed-effects approach combined with value-added models. Our objective in using these methodologies is to control, as much as possible, for unobserved variables that may have potentially biased the analysis. To the extent that twins are similar in most of their characteristics, then we employ within-twin variation in homework time to analyse its influence on this same variation in terms of academic performance, also controlling by students’ previous academic performance (using a value-added model).

‘The results indicate that homework is not positively associated with students’ academic performance, and may not be helping them to improve their skills.’

The main results derived from the estimation of these models indicate that homework is not positively associated with students’ academic performance (as found by research works including Cooper, Robinson, & Patall, 2006; and EEF, 2018). Hence, in this context our finding highlights the need to re-evaluate the particular topic of whether and in what ways homework influences students’ academic performance within each particular country setting, expecting that this influence may vary by country.

Furthermore, the variables under analysis in our research work were students’ competences – that is, their ‘real-life skills’ – rather than the content-based knowledge that is more usually studied, as we assume that these competences may be relevant for students’ later outcomes, such as their incorporation into the labour market. Our findings imply that the homework that Andalusian students are receiving may not be helping them to improve their skills. This may seem unsurprising in the Spanish context, as students are usually assigned homework that consists of memorising content-based knowledge without any kind of practical application of that knowledge to real life.

Our study has a number of limitations: it relies on students’ self-reports of their homework time; it is difficult to exactly identify all the twin students in the census; there is a higher representation of twins among families of higher socioeconomic status; and there are unobservable factors which remain uncontrolled in our estimations. Therefore, our results should be interpreted with caution and considered correlational ones. Nevertheless, the results of our study open the gate for analysis of the influence that higher quality homework (in the sense of competence-oriented homework) may have on students’ academic performance.

This is relevant to the extent that students may have more free time to perform extracurricular activities and, some may argue, even enjoy their childhood/adolescence. This issue goes beyond homework time alone, as authors such as Zuzanek (2008) have highlighted:

‘Pressures experienced by teens do not stem from homework alone but from the way society structures its employment, educational and temporal. In other words, we face not so much a ‘homework problem’ but rather, in a broader sense, a societal time use problem.’
(Zuzanek, 2008, pp. 114–115).


Associated Press (2016, September 26). Goodbye to homework for some elementary schools and classes. Daily Mail. Retrieved from

Cooper, H., Robinson, J. C., & Patall, E. A. (2006). Does Homework Improve Academic Achievement? A Synthesis of Research, 1987–2003. Review of Educational Research, 76(1), 1–62.

Education Endowment Foundation [EEF] (2018) Homework (Primary): Low impact for vert low cost, based on limited evidence [Webpage]. Retrieved from: evidence-summaries/teaching-learning-toolkit/homework-primary/
(Spanish version:

Fernández-Alonso, R., Álvarez-Díaz, M., Suárez-Álvarez, J. & Muñiz, J., (2017). Students’ Achievement and Homework Assignment Strategies. Frontiers in Psychology, 8(286), 1–11.

Inchley, J., Currie, D., Young, T., Samdal, O., Torsheim, T., Augustson, L., Mathison, F., Aleman-Diaz, A., Molcho, M., Weber, M. & Barnekow, V. (2016). Growing up unequal: Gender and socioeconomic differences in young people’s health and well-being: Health Behaviour in School-Aged Children (HBSC) Study: International Report From the 2013/14 Survey (Health Policy For Children And Adolescents, no. 7). Geneva: World Health Organisation.

Jones, S. (2016, November 2). Spanish parents urged to put children on weekend homework strike. Guardian. Retrieved from

Murillo, F. J. & Martínez-Garrido, C. (2013). Homework Influence on academic performance: A study of Iberoamerican students of Primary Education. Revista de Psicodidáctica, 18(1), 157–178.

Zuzanek, J. (2008). Students’ Study Time and Their “Homework Problem”. Social Indicators Research, 93(1), 111–115.

More content by John Jerrim, Luis Alejandro Lopez-Agudo and Oscar David Marcenaro-Gutierrez