Does a longer school day raise achievement?

15th January 2012

 Since January both sides of Parliament have called for an extension to the school day.  Stephen Twigg argues that longer hours will give poorer students a haven from their chaotic homes and provides a model of hours endured in an adult’s working life. Michael Gove is less clear on why he wants to lengthen school days but most likely because it’s one way US Charter Schools argue they are ‘closing the gap’ between rich and poorer students.
But do longer hours actually improve outcomes for students?
In the mid-1990s the Chilean government extended the school day by 30% when it moved from a 2-shift system – where some students were taught in the morning and others in the afternoon – to everyone being taught for the complete day.  The change added an extra 25% to the education bill and parents were pleased that the majority of the extra time went on academic study and homework support. The change therefore provided a natural experiment where results before and after the move could be examined.  The longitudinal data on those cohorts is now becoming available and makes for interesting reading.
Cristian Bellei’s paper compared the 10th grade Maths & Language scores of 180,000 students schooled just before and just after the change.  The headline is good news for longer-day advocates: both Maths and Language scores significantly increased after the change.  As per Twigg’s ideas, the number of students hanging around on the streets or getting into trouble also dropped. The bad news for Gove, however, is that the achievement gap increased.  Well-motivated learners seemingly took greater advantage of the increased hours whereas disaffected students got further and further behind. 
Beyond students, the longer day led to a teacher shortage and increased reported discipline issues in schools.  Though it is convenient to believe that the reason for poor performance at GCSE is the cruelty of teachers locking knowledge-seeking children out at 3pm, the reality is that some students can barely self-regulate for the 6 hours we currently ask of them. If increases to the school day go ahead these will arise and solutions must be sought ahead of time.
A more recent longitudinal study of the Chilean change supports Bellei’s positive findings and shows that the longer-school day: improves exam scores, reduces anti-social behaviour and lowered the number of adolescent mothers.  But the authors note these improvements did not translate to greater employability or increased wages, putting to bed Twigg’s ideas that a longer day improves one’s job-getting abilities.  However, the research unequivocally states that the longer days best helps students with uneducated parents, as the longer day provides more time with educated role models and greater access to educational materials than is available at home.  On the down side where students have highly educated parents the longer school day can be more negative for their cognitive scores than being at home.
Interestingly, in both papers the differences shown in exam outcomes are small. Though the results are ‘significant’ in the scientific sense of showing the longer day is the main cause of the change, the size of the difference in scores are not massively significant. It looks like the difference of only a few marks per student.
Giving students more access to schooling is surely a good thing – this is why I was a fan of the ‘Extended Schools’ agenda. But it does cost a lot and it is not solving several of the problems Gove and Twigg purports it does.  If the politicians keep onwards (as I hope they do) then they must consider what it will take to make these reforms work well and have the desired outcomes, otherwise it will be a very expensive way to not get very far.