Despite the fiery debates for and against homework (and the inferno at this time of year, as parents begin to form opinions of their children’s nightly workload, can be particularly scalding), few teachers are in possession of reliable evidence on what, exactly, works best.

Indeed, I have set homework for my entire teaching career, with more to come in the school year ahead, but I couldn’t confidently state how much homework I should set and how frequently I should set it.

Old habits can die hard and unthinking homework habits appear to plague our profession. But the days of parents taking a painstaking role building motte and bailey castles, or coaxing their younglings through interminable spelling tests, need to be put behind us. Homework horror stories can be avoided with a better understanding of the evidence that attends effective homework setting and applying it with wisdom.

So here’s what I found on a journey into the research.

Not all homework is good homework

Evidence synthesised by the Education Endowment Foundation, and compiled by academics at Durham University, has revealed that the benefits of homework at primary school are modest at best. And it would appear that we may be overloading young children with too much homework.

The evidence of the positive impact for older students is clearer. In their 30-year review of homework, Fan, Xu, Cai and Fan (2017) found that academic achievement in maths and science correlated with time spent on homework, homework frequency and homework effort.

Of course, not just any old homework will have a positive effect on our students’ learning, but research can guide us in the right direction. For example, Cathy Vatterott (2010) pin-pointed the five key characteristics of good homework:

• Purpose: homework is meaningful and students understand its value to their development.
• Efficiency: homework is not too time-consuming, but it does demand some hard independent thinking.
• Ownership: students who see the value of the homework task are better motivated and focused. Providing choice can offer a greater sense of ownership.
• Competence: students need to be able to complete the task independently and well.
• Inspiring: well-designed homework can better motivate students and encourage greater focus and effort.

Now, can I honestly say that all the homework I set my students meets this testing criterion? And can you?

Probably not, so the list is a great checklist to help avoid setting ineffective homework for the school year ahead.

But how often you should set it, and how much should you set?

Is bigger really better?

It wasn’t too long ago that a Labour government set school homework guidelines of one hour a week for five- to seven-year-olds, rising to over two-and-a-half hours for 14- to 16-year-olds. These guidelines, subsequently scrapped by Michael Gove, have seemingly been forgotten and each school is now left to their own devices in struggling to create its own homework policy.

In our super-sized modern world, bigger is too often mistaken for better. When it comes to homework, too, the best available evidence would indicate that a large amount of homework could bring diminishing returns. Instead, a relatively small amount of homework undertaken each night appears to have a positive effect. A mere 60 minutes may prove to be the sweet spot.

Meanwhile, a recent study, based in Spain (Fernández-Alonso, Suárez-Álvarez, and Muñiz, 2015), revealed that when students were given more than 90 to 100 minutes of homework a day, their maths and science achievement began to decline. Tales of tired students being turned off school become validated by such findings.

Judging the time a task will take, though, is difficult. When you begin to dig beneath the headlines, it’s important to note that there are many pupils in our classes who lack the prior knowledge of our most successful students. Any homework issued can bleed into endless hours of toil with significantly diminished returns.

And paradoxically, when well-meaning parents help their children it can have a negative impact over time and hamper their development (Trautwein, 2007).

The onus is on us, as teachers, to set homework that achieves the Goldilocks principle: it should not be too easy, nor too hard, but just right.

If we are to consider the best evidence on homework, we may use the analogy of homework being a beautiful diamond: small, expertly crafted and precious to its owner.

Of course, fashioning some of our dull school curriculum into a polished diamond will prove to be an enormously skilful craft.

Alex Quigley is director of Huntington Research School in York. He is the author of The Confident Teacher, published by Routledge