Evidence for school absences
Dear Department for Education,
In a BBC article of 9 July 2016 a spokesperson for the department of education was quoted as saying
"The evidence shows that every extra day of school missed can affect a pupil's chances of achieving good GCSEs...."
I'd like to know what evidence this spokesperson is referring to, and which controlled study allows them to state that a single day's absence makes a difference.
Yours faithfully,
Professor Alan Barr,
University of Oxford
Dear Mr Barr, [FOI #344752 email]<https://crm/IRIS/main.aspx?etc=2&ext...>
RE: Case reference 2016-0033454
Thank you for your request for information, which was received on 11 July 2016. You requested:
“In a BBC article of 9 July 2016 a spokesperson for the department of education was quoted as saying
"The evidence shows that every extra day of school missed can affect a pupil's chances of achieving good GCSEs...."
I'd like to know what evidence this spokesperson is referring to, and which controlled study allows them to state that a single day's absence makes a difference.”
I have dealt with your request under the Freedom of Information Act 2000.
Under section 21 of the Act, the Department is not required to provide information in response to a request if it is already reasonably accessible to you.
Information on the link between absence and attainment at key stages 2 and 4 was published in the “Absence and attainment at key stages 2 and 4: 2013 to 2014” research report, published March 2016 and available at https://www.gov.uk/government/publicatio....
The analysis of the link between overall absence and attainment when taking prior attainment and pupil characteristics into account showed that, for each KS2 and KS4 measure, overall absence had a statistically significant negative link to attainment – i.e. every extra day missed was associated with a lower attainment outcome.
The information supplied to you continues to be protected by copyright. You are free to use it for your own purposes, including for private study and non-commercial research, and for any other purpose authorised by an exception in current copyright law. Documents (except photographs) can be also used in the UK without requiring permission for the purposes of news reporting. Any other re-use, for example commercial publication, would require the permission of the copyright holder.
Most documents produced by a government department or agency will be protected by Crown Copyright. Most Crown copyright information can be re-used under the Open Government Licence (http://www.nationalarchives.gov.uk/doc/o...). For information about the OGL and about re-using Crown Copyright information please see The National Archives website -http://www.nationalarchives.gov.uk/infor... .
If you have any queries about this letter, please contact me. Please remember to quote the reference number above in any future communications.
If you are unhappy with the way your request has been handled, you should make a complaint to the Department by writing to me within two calendar months of the date of this letter. Your complaint will be considered by an independent review panel, who were not involved in the original consideration of your request.
If you are not content with the outcome of your complaint to the Department, you may then contact the Information Commissioner’s Office.
Yours sincerely,
School absence and exclusions team
Education Standards Evidence and Dissemination Division
Dear Department for Education,
I wish to know not the evidence for correlation but the evidence for causality. That information is not available in the linked report, which shows only a correlation. The statement from the spokesperson makes clear that the department believes there to be evidence that absence from school *causes* lower grades. On what study do they base that statement?
Yours faithfully,
Professor Alan Barr
Dear Mr Barr,
Thank you for your request for information, which was received on 13 September 2016. You requested:
“I wish to know not the evidence for correlation but the evidence for causality. That information is not available in the linked report, which shows only a correlation. The statement from the spokesperson makes clear that the department believes there to be evidence that absence from school *causes* lower grades. On what study do they base that statement?”
I have dealt with your request under the Freedom of Information Act 2000.
The quote queried states "The evidence shows that every extra day of school missed can affect a pupil's chances of achieving good GCSEs...." and is based upon published evidence, showing that there is a link between absence and attainment. The full report is available in ‘The link between absence and attainment at KS2 and KS4’ at https://www.gov.uk/government/uploads/sy....
The quote states clearly that missing school can affect a pupil’s chances of achieving good GCSEs and does not imply causation.
If you have any queries about this letter, please contact me. Please remember to quote the reference number above in any future communications.
If you are unhappy with the way your request has been handled, you should make a complaint to the Department by writing to me within two calendar months of the date of this letter. Your complaint will be considered by an independent review panel, who were not involved in the original consideration of your request.
If you are not content with the outcome of your complaint to the Department, you may then contact the Information Commissioner’s Office.
Yours sincerely,
School absence and exclusions team
Education Standards Evidence And Dissemination Division
Dear Department for Education,
Again, the study referred to in this reply shows only evidence of correlation, not of causality. It seems likely that some factor (such as socio-economic conditions, parental support, or geographic area) is responsible both for lower attainment and lower school attendance. The study referred to does not discuss methods for dealing with such conflating factors. The study therefore is insufficient to support the statement made by the spokesperson that every extra day's absence from school reduces attainment.
This central claim that lower attendance actually *causes* lower attainment has not been supported by any evidence presented in the responses. The continued referral to a study showing mere correlation suggests that the difference between correlation and causality has not been misunderstood.
Since the responses fail to support the claim of causation made by the official I hereby request a review.
Yours faithfully,
Alan Barr
Thank you for your message. Please note that from 1st October 2016 this email address is changing to [DfE request email]
Beccy Smith left an annotation ()
Professor Barr,
This so called causation is even more flawed than you have drawn attention to. I analysed the raw data from the data set referred to. When looking at the all types of absences (including unauthorised, sickness, exclusion as well as holiday) we do indeed see that 2014 KS2 level 4 attainment drops from 94.7% at no absence steadily down to 87.0% at 20 days absence. However, when looking at only authorised holiday absence, we see that KS2 attainment jumps up from 83.8% at no authorised absence to a reasonably level 87% attainment for 1 to 20 days absence. (Attainment ranges from 85.5% to 88.3%, but has a weighted average of 87.1%.)
This is not a case of wondering whether the negative correlation is causation; rather, why the positive correlation has been advertised as negative correlation.
Dr Beccy Smith
Professor Alan Barr left an annotation ()
I undertook a brief re-analysis of the raw KS2 data from the study. I confirm Becky's findings.
The four conclusions I draw from these data are that:
(1) Long periods of absence are dominated by pupils' illness
(2) Children who have long periods of illness unfortunately tend to perform significantly worse in the tests
(3) Authorised holiday absence has almost no effect
(4) Except that those pupils who take no authorised holiday absence at all are likely to do worse than those who do take at least one day
Dear Mr Barr,
RE: Internal review of Freedom of Information request - Evidence for school absences
I refer to your complaint/request for an internal review which was received on 26 September 2016. You requested
“Again, the study referred to in this reply shows only evidence of correlation, not of causality. It seems likely that some factor (such as socio-economic conditions, parental support, or geographic area) is responsible both for lower attainment and lower school attendance. The study referred to does not discuss methods for dealing with such conflating factors. The study therefore is insufficient to support the statement made by the spokesperson that every extra day's absence from school reduces attainment.
This central claim that lower attendance actually *causes* lower attainment has not been supported by any evidence presented in the responses. The continued referral to a study showing mere correlation suggests that the difference between correlation and causality has not been misunderstood.
Since the responses fail to support the claim of causation made by the official I hereby request a review.”
The Department has now completed its internal review process and has carried out a thorough review of the case, chaired by a senior officer who was not involved with the original request. The Department has decided to uphold the original response for the following reason:
The Department took note of your comments when considering the scope of your request. We understand that you do not consider that the evidence demonstrates a causal link. However, the Department has not claimed a causal link, and we have previously explained that in the statement the spokesperson said that absence ‘can’ (rather than ‘will’) affect GCSE performance,
As you will be aware, the Freedom of Information Act requires requesters to describe in writing the information they seek – in this case, the evidence being referenced by the spokesperson – and public authorities to consider the information that they hold in scope of that description. They must then release any information held unless there is any reason set out in the Act which would prevent its release. Accordingly, the matter at issue for the review was whether the Department handled your request properly under the FOI Act – i.e. whether we could identify information in scope of your request and if so, whether we should release it.
Your request was as follows “In a BBC article of 9 July 2016 a spokesperson for the department of education was quoted as saying "The evidence shows that every extra day of school missed can affect a pupil's chances of achieving good GCSEs...."
I'd like to know what evidence this spokesperson is referring to, and which controlled study allows them to state that a single day's absence makes a difference.”
The evidence that the spokesperson was referring to when quoted as saying that a day’s absence can affect a pupil’s chance of achieving good GCSEs was the report provided to you in response to your initial request. For that reason, the Department is clear that we have already identified and provided the information that you described in your request, and that we handled your initial request in accordance with the Act. We recognise that you will be disappointed with the response, but hope that this helps to explain our decision.
If you are unhappy with this decision, you have the right to appeal directly to the Information Commissioner. The Information Commissioner can be contacted at:
The Case Reception Unit
Customer Service Team
Information Commissioner’s Office
Wycliffe House
Water Lane
Wilmslow
Cheshire
SK9 5AF
Further information about the Information Commissioner’s complaints procedure can be found on the Information Commissioner’s Office website: https://ico.org.uk/for-organisations/gui...
Yours sincerely,
School absence and exclusions team
Education Standards Evidence and Dissemination Division
Jon Platt left an annotation ()
Thank you again Professor Barr and Dr Smith for this analysis. I see that the DfE has responded and in effect is now acknowledging that there is NO CAUSAL link and seems to accept that the divergence of opinion is between 'CAN' & 'WILL'. That seems, to me, to be clear acceptance that the statements regularly put out by the DfE (when questioned about the impact of HOLIDAY absences) were only true in respect of other types of absence. Why, I ask, do they pretend not to grasp that different TYPES of absence have a different impact on attainment and keep referring to 'absence' as if the different types of absence had a uniform impact on attainment. Clearly the annotations from Professor Barr and Dr Smith show clearly that the best that can be said of absences for family holidays in relation to attainment is that there is no evidence at all of a causal link and at best there is a correlation between improvement in attainment and family holiday absences.
Regards
JON PLATT
James Coombs left an annotation ()
Dear Professor Barr,
Firstly I'd like to thank you for pursuing this. I am very concerned about the lack of objectivity in the way education policy is formed. I haven’t looked at the data myself but might include a request for absence details in my next NPD request.
This has probably already occurred to you but I think is worth pointing out to the wider world. From a Freedom of Information perspective, this is concluded. You asked the DfE to provide the information they based this claim on. They provided it. The FOIA places no obligation on an authority to publish information which is factually correct. Maybe we need new laws in our post-reality society. (Who’d want to turn down the an additional £350m/week on the NHS?) It would have been encouraging if the DfE conceded the valid points you have raised and offered to publish a correction but you’ll probably find that your Freedom of Information request was handled by the Freedom of Information department who take a rather narrow view. Hopefully Mr Platt’s high profile case will get these important points aired but you could raise this with Mr Tim Leunig who has come across as being an objective person in my dealings with him.
Best wishes
James Coombs
We work to defend the right to FOI for everyone
Help us protect your right to hold public authorities to account. Donate and support our work.
Donate Now
Jon Platt left an annotation ()
Professor Barr,
I must say I am absolutely delighted that a man of your stature is pointing out this CRITICAL flaw in the DfE argument.
I'm the father on the Isle of Wight imminently due in the Supreme Court (IW Council v Platt). All along I have argued that the DfE have been citing statistics that show correlation NOT causation. It is WONDERFUL to read this FOI request.
Thank you
JON PLATT