Education and Skills Committee

Meeting date: Wednesday, November 28, 2018

Official Report 761KB pdf

---

School Support Staff (Data)

Item 2 is an evidence session on the availability of information on school support staff data collected as part of the school staff census. The session will focus on the changes to the presentation of statistics, including the merging of categories relating to additional support needs support staff and the changing of the publication status of a number of categories so that staff figures are available on request as opposed to being published as standard.

I welcome to the meeting from the Scottish Government Roger Halliday, who is the chief statistician and data officer; Alasdair Anthony, who is a statistician and head of the school, staff and pupil census statistics team; Laura Meikle, who is the learning directorate’s head of the support and wellbeing unit; and Mick Wilson, who is acting deputy director, education analysis.

I invite Mick Wilson to make opening comments.

I thought that it would be useful to put the data that we are here to discuss and the points that you have listed into a bit of context, in terms of both the statistics and the processes that are used to collect, assess and publish those data.

A range of data collection exercises are undertaken on education throughout the year. A number of those are census-type collections, which are conducted in parallel over the autumn. We are at the end of the 2018 collections across a range of things. Although all those collections are related and happen in parallel, they are distinct exercises and have important differences between them. No doubt we will come to some of those throughout our conversation.

Our primary sources of data are the pupil census and the teacher census, which are probably the most well-recognised collections across the system. We also conduct the non-teaching school staff census from which the support staff data come, the primary school class data collection and the school establishment collection. Every two years, we collect data on attendance, absence and exclusions in schools.

A range of other collections are carried out throughout the year, including collections on school meal provision and physical education provision, and a range of data exercises relating to pupil performance and outcomes.

The autumn administrative collections are supplied to the Scottish Government directly by local authorities and, in some cases, by schools, particularly grant-aided schools. The quality assurance process starts with their initial collection of the data, which they use for the day-to-day running of the education system. They hold those data on their management information systems for regular use.

Quality assurance is built into each stage of the process, so schools, local authorities and the Scottish Government all have a role to play in that process. The precise arrangements vary a little bit between the collections, depending on the nature of the data and the amount of detail that we collect. Ultimately, the data are signed off by directors in local authorities prior to being published by the Scottish Government.

During those administrative data collections, we collect and process what amounts to tens of millions of pieces of data relating to 2,500 schools, around 700,000 pupils, 70,000 teachers and more than 20,000 support staff across the system. That enables us to publish more than 100,000 statistics. In addition, we release bespoke data sets and analyses throughout the year for a wide range of users, including researchers and academics, the media, the general public, politicians and parliamentary committees.

The code of practice for official statistics sets a framework for our approach to handling data and producing statistics. It is there to ensure the quality, value and trustworthiness of the statistics that are produced by organisations such as the Scottish Government, and it provides us with guidance on specific aspects of producing statistics, such as ensuring that the burden on data providers is proportionate and that appropriately qualified professionals are used throughout the process. The ultimate responsible for ensuring that the code of practice is adhered to in the Scottish Government rests with the chief statistician.

The purpose of the administrative data collections in education, as with all other sectors, is to describe—it is to paint a picture of the subject to which the data relate. They can only ever play back practice in the system or paint a picture of the system. They do not and should not define or constrain practice in a system. However, they do and should facilitate debate and discussion about practice in the system. Any statistical collection or publication does not exist in isolation or for its own sake; it needs to remain relevant, reflecting current or likely future situations, while ensuring that as robust and accurate a picture as possible can be presented.

What we publish and how we publish it needs to be a balance between a number of factors. Those include known limits or restrictions to the data, previous practice on publication, public interest, current issues and the context in which we are presenting the data, and the availability of other sources of information on the same or similar topics. Our approach is generally to publish as much information as possible in as accessible a format as possible, without releasing data that we know to be misleading, incomplete or erroneous.

Education data are probably more in the spotlight now than at any other time in recent memory. That brings a requirement to continually improve their fitness for purpose; it also more often brings us closer to the practical limits of the power of the data that we collect. Therefore, simply replicating what has gone before will not always provide the most useful or accurate data, and it is right that we take action to investigate and address any anomalies in the data or issues raised by data providers throughout the process.

No decision to make changes to the presentation of official statistics is taken lightly. In this case, the requirement to reflect on-going changes in the relative importance of support staff data, as against other elements of education data—teachers and class sizes, for example—meant that a deeper examination was warranted when a potential issue was highlighted to us. The resulting changes provide a more reliable data set than would have been the case if we had left those unaddressed.

On the process for the timeline of publications, typically, the data is collected for a date in September, the initial publication is in December and supplementary data is published in March. For the most recent set of data, the supplementary set came in July. Was that a one-off, because this was the year in which you began making the changes that we are here to discuss, or will that be the new timeline for the publication of data?

Yes, that is right—the data was released slightly later than it has been in the past because of the additional quality assurance checks. We have announced that the supplementary data from the staff census, pupil census and whatnot will next come out in March 2019.

Brilliant—thank you.

When the supplementary data was published earlier this year, it came with a notice explaining that additional data would be provided on request. When I requested that additional data, my request was treated under freedom of information rules. That seems like an odd and onerous process to go through when soliciting requests for additional data. Will that be the process going forward?

The data that we will publish in March 2019 will be available on the website as normal as part of the published stats tables. We have not yet decided what we will do about the information that we released as management information—the additional categories of staff that were not proactively published originally. We have yet to take a decision on how we will make that information available.

It is relatively normal practice for us to treat a range of data requests as freedom of information requests. Strictly speaking, any request to the Scottish Government is a freedom of information request, under the legislation—

That is not true—

I am sorry, Mr Wilson, but that is not the case. Members of the Scottish Parliament routinely make requests of the Government that are not freedom of information requests. MSPs do that every day. The process that you described seems very odd and onerous. To treat everything as an FOI request must significantly increase the workload of your team.

No. I do not recognise that using the FOI process adds to the workload. A number of non-routine requests for data will be treated as freedom of information requests. This was the first time that we had released such a level of detail for that set of data, so it was right that we used the formal process. Doing it that way also meant that we could release that level of detail to all users at the same time, because we publish our FOI responses, rather than just releasing the information to the individual who made the request, as is the case with most of the ad hoc data requests that we get more routinely throughout the year.

We will need to come back to that.

Let me move on to the change whereby information on classroom assistants and additional support needs assistants is published under a new category, “pupil support assistants”. You used to treat classroom assistants and ASN assistants as separate categories, and when I requested information in that regard you were able to provide it. Now, I understand that some local authorities list 700 staff under one category and none under the other. We can come back to that, too.

This year, you used both categories. In future years, will you request information from local authorities only under the pupil support assistants category? In other words, you will not request information under two separate categories and publish it under one category; you will request information under just one category in the first place.

The information that we collected for the 2018 census has not changed from the information that we collected in previous years. We still collect information on classroom assistants and ASN auxiliaries, as we did in the past.

If you are collecting information on both categories, why are you publishing everything under one category? The information for both categories is available on request—I requested it, and it is now in the public domain. Why publish it under the pupil support assistants category, when you have data for both groups?

We based that decision on the feedback that we had from local authorities about how they assign staff to categories and about the terminology on staff roles that they use in schools in practice. As statisticians, we took the decision that the most appropriate way to present the data that we had collected was under the category of pupil support assistants, because that is the term that local authorities tell us that they use in schools.

What alternatives to merging the data before publication did you explore? Did you explore working with local authorities on clearer guidance on the definitions and which staff fit into which category?

We thought about a range of options when the issue was first raised. It is fair to say that we had a pretty open mind about the solutions that we might end up with.

What we do not want to do is produce data on a basis that we know to be questionable. We know that local authorities struggle to assign staff to the categories for which we currently collect data. Rather than make a fundamental change to the underlying collection process—and in a hurry—we decided to take the approach of combining the categories.

Had it been clear that pupil support assistants were a new type of staff that was additional to the categories of staff on which we were collecting data, another option would have been to add a further category to the collection, and we could have reported on an extra category, on top of the two that were already there. However, it became clear very quickly from local authorities that that was not the case. The term “pupil support assistant” covers a wide range of tasks, which incorporate some of those that are undertaken by classroom assistants and ASN auxiliaries. It was therefore preferable to combine those two categories and present what we think is a more accurate overall picture of the resources in that area, rather than falsely delineating between the two categories that were initially in the collection.

10:15  

Do you understand the concern that exists when one in four young people in Scottish schools has an identified additional support need but the published data has now been narrowed to the extent that, until the further data requested is published, we would be unable to tell how many support staff are working with children with additional support needs?

I can understand the question. However, I do not think that what we have learned from local authorities tells us that the categories that we had before provided an accurate picture of the number of staff working with children with additional support needs.

How does the currently provided information provide an accurate picture of the number of staff working with additional support needs? It just provides a generalised category of pupil support assistants.

That is precisely the issue. There is a range of staff across the system now who are working in a number of ways with a number of pupils in a number of different scenarios across schools. Some of their time is spent working with children with additional support needs and some of their time is not. To delineate them on the basis of what now appear to be outdated job titles would provide a false picture. Using the category of pupil support assistant provides a more robust picture, enabling comparison between local authorities of the overall resource that is going into supporting pupils in Scottish schools.

Fundamentally, it does not tell us what resource is going into supporting pupils with additional support needs—one in four pupils. It is an acute problem that has been identified, including by this committee. The published data no longer provides information on the number of staff who are supporting pupils with additional support needs.

The question of the resources going to support pupils with additional support needs is a slightly different one. It would require assessing what resources, provided by a range of staff in education systems, go to supporting kids with ASN, whether that be classroom teachers, professionals from outside the education department, or support staff of one form or another. Trying to split out the proportion of time that individuals spend supporting pupils with additional support needs would be a difficult and onerous task. Looking at the categories of support staff or specific teaching staff—either before or now—cannot tell us the overall picture of the resource going to support pupils with ASN.

I am happy to hand over to other members at this point, convener, but I would be keen to come back in later.

Would Mr Wilson write to the committee with the procedures for handling requests? That would be helpful for our deliberations down the line.

Sure.

Mr Wilson said that concerns were raised about the statistics. Who raised those concerns? Was it the Scottish Government or the statisticians who raised concerns about the quality of the statistics?

The concerns were picked up as part of the initial quality assurance procedures that we carried out on the data. When we looked at the data, we discovered some anomalies. That was when we decided to undertake further quality assurance processes. On the library staff in particular, we responded to a parliamentary question and it was then raised with us that some of the information provided might not have been accurate. We returned to the data on that question and subsequently issued a correction to the response to the parliamentary question.

I cannot say the word “statistician”, so you will know that I am not one. Is it normal practice, when people give you poor information, to generalise the information that you are seeking? That is what it feels like has happened here. Local authorities are not giving you detailed and accurate enough information, which resulted in you saying things that you later felt were not justifiable professionally, and therefore you have generalised it. Can you think of any other example where you have gathered data where you have done that?

It is for the professional judgment of the statisticians whether time spent trying to rectify or improve the situation would be value for money. I can think of situations to do with the economy, or with our survey data. We could survey more companies or more people to get a more accurate picture of what is going on, but we have to make a judgment about what is good enough in the circumstances. We report on relatively high levels of industrial classification, for example, because our surveys of business are only so big and therefore are limited in their ability to drill down to specific industry classifications. Also, although we do quite a big population survey for our labour market statistics, we report at relatively high levels of aggregation of job titles, because we can only go to so many people, as the cost of going to more people would be prohibitively high. I guess that this is similar.

The other side of the argument is that you produce statistics that do not tell you anything. If I ask, “How many women are working in this field?” and you say, “We’re not getting a very good response to that, but I can tell you how many people are there”, that misses the point.

I want to ask Laura Meikle about the Scottish Government’s view. Statisticians do a professional job; it has already been said—and quite rightly so—that they do not define the situation, they describe it, and that is what provides the opportunity for debate.

That is right.

John Swinney has said that he is pausing some of his response to us until he has looked at discussions on “Not included, not engaged, not involved: A report on the experiences of autistic children missing school”, which covers some very profound issues including the support for a young person with autism, how much of the school day they are in school, whether they are excluded inappropriately and whether they are taught with their peers. If we cannot even say how many young people with autism have the support of an additional support needs professional, how can we possibly begin to enter that debate?

I think—

What I want you to tell me is: what did Scottish Government officials say when they were told that you were going to produce those new generalised categories? John Swinney was very clear that it was nothing to do with him.

A very important thing to reflect on is that although statistics are an important part of the evidence base that we use to implement policy, they are not the only part.

If they are important, if they describe rather than define and if they offer the opportunity for debate, which is seen to be their purpose, how can generalising the categories in that way help define the debate?

On the specific question about the decision, we have an advisory group for additional support for learning, which involves a wide range of stakeholders such as children and young people, parents and service delivery people—it is a vast array. We capture the perspective of a range of people when we think about implementation. In that arena, we have had discussions around data in a slightly different way than we are describing here with regard to this specific change, and the information that came through was that the term “pupil support assistant” was more appropriate. The question of the terminology of “additional support needs auxiliary” or “classroom assistant” has been raised in those arenas and in my team’s discussion with a wide range of stakeholders. The concern is with ensuring a proper reflection, so when the suggestion was made about joining the two categories, I was comfortable with the decision, because it linked back to what our stakeholders had told us.

Are you seriously saying that about stakeholders who have already said that their children, in describing their school experience, talk about not having a full school day, not being properly supported and having the additional support that they are entitled to pooled with other young people?

No.

Are you seriously saying that they told you, “It’s okay to generalise those categories. We’ll leave the policy to be based on anecdote by this group rather than on the evidence that should underpin any policy change”?

No. I am saying that my discussions with the education authorities and the Convention of Scottish Local Authorities, as part of the wider discussions with all the stakeholders, led to our being able to agree that the two categories could be joined together, given that the terminology that is predominantly used in the system is “pupil support assistant”.

However, do you accept what families campaign groups have said, which is that the teaching unions and people who work in staff support are saying that there is a lack of support in schools and that that is a burden on teachers and a general pressure on the workforce? They say that although that is happening, the Government and local authorities are saying, “It is not really as bad as that.” What we do not have is what you would then go to—the evidence base.

Absolutely.

You have generalised everything now.

No—

But you cannot answer the question. How many young people with additional support needs have somebody professionally trained to support them instead of there being generalised classroom support from which they are just given some of a person’s time? You cannot answer that question any more.

I am sorry—what I started to say was that the information from the statistics is just one part of a wide range of evidence.

But that one part no longer answers the question.

Please let the witness answer the question.

We use a wide range of information to consider the implementation of additional support for learning, including information from Education Scotland inspections and views from parents and young people on the support that they receive in school. We engage with specific stakeholders. For example, in relation to the evidence that you have referred to, officials engaged directly with the National Autistic Society Scotland, Scottish Autism and Children in Scotland to have a discussion about which actions needed to take place. We do not have to rely entirely on the statistical evidence in order to take action to improve implementation. Indeed, that is what happens routinely.

If you do not know what the picture is, and if you believe in—

We ask stakeholders to give us the information, and it is a much broader range of information and evidence than that from statistics.

The logic of that position is that you would not employ statisticians at all, but simply ask people how they felt things were going.

I understand that there is more to your job than looking at the stats—I respect that. However, if people say that the system is not working, one way in which you can establish whether that is the case is by looking at the evidence and saying to families, “Actually, there is something different here.”

Absolutely.

The challenge—or the problem—is that we are told that, in our schools, young people who need to have personal additional support needs professionals working with them are now part of a wider group of young people being given broader classroom support. The two things are not the same. People are fearful of the consequence—as was set out in the report by the National Autistic Society Scotland, Scottish Autism and others—which is that young people are put on part-time timetables or excluded within the school estate, or do not have access to the same level of support that they might otherwise have expected. You cannot even rebut that, because you do not have the evidence to do so.

I am saying to you that we do not use the statistical evidence that we are talking about here today to try to address such issues. If there is a problem over part-time timetabling or exclusion from school, we go and talk directly to people about the underlying issues that caused that problem, in order to establish the actions that we need to take to resolve it.

The “Not included, not engaged, not involved” report is one of many reports; Enable Scotland did one about the experience in schools of young people with additional learning needs, and the National Association of Schoolmasters Union of Women Teachers produced another that said the same thing. The Scottish Government would respond to that by saying, “Well, actually, our figures do not suggest that that is the scale of the problem.” However, it did not respond to the reports; it cannot rely on them now and say that the approach is simply about that dialogue and engagement. Do you understand that frustration?

I do, but I think that you are reflecting a position in which we would respond to each individual report and not think about the collective evidence that had emerged from all of them. We would consider all that information as part of our evidence about importance.

And would you test it against the statistical evidence, which has been weakened?

We would test it against the views of our stakeholders—the large group that I talked about—by asking them whether it feels real to them. The ultimate question that we ask is about what actions we should take; we do not sit and challenge everybody’s evidence, and we do not rip those reports apart. We ask that question and then we start to look at our implementation requirements with our stakeholders.

I was not suggesting that. You cannot argue on the one hand that those reports provide you with evidence to respond to the problem and then, on the other, say that you do not have to worry about stats because you have dealt with that. The reports are pushing back, because the Government has said, “No—actually, the scale of the problem is not something that we recognise.”

We use all the information that is available to us as an evidence base for informing our actions. That includes information from statistics, stakeholders—including all the reports to which you have referred—and Education Scotland, and we also engage regularly with the Association of Directors of Education in Scotland and COSLA. Specific groups look at children and young people with additional support needs and engage with additional support for learning officers. We do so to ensure that we have a range of information at a number of levels in the system, which informs our policy decisions. We do not rely on any single piece of evidence at any single point in time. It is about having a holistic approach to information.

10:30  

But you were content to have less statistical information. You were content for the information to be more generalised than it used to be.

I do not have less information.

On the specific point about the data, I do not accept that we have less information.

The premise behind Ms Lamont’s question is that the categories that we had before were an adequate description of what staff did. However, what the collection does is count members of staff who have various job titles or roles; what it does not do, as I have said, is account for their time and what they do to support individual pupils or provide broader support to a teacher in a classroom or school setting, nor does it account for the time that individual classroom teachers spend directly supporting pupils with additional support needs, which is a very different exercise. As I have said, considering the job titles or roles that we collect as part of the administrative data uplift is not the right way to answer your specific question about the amount of support that pupils with additional support needs receive in schools.

My point is that we need to know that, to ensure that pupils’ needs are being met.

I do not blame you if you find it difficult to collect the data and have to make a judgment call. However, you would never go into a secondary school and just say, “Gonna tell us how many teachers are here.” If you want to know about a secondary school’s capacity to teach science, for example, you need to know how many science teachers there are. Why would it be any different for ASN?

I accept that you are operating in a policy framework that has not been decided by you. The Scottish Government has said that it does not want to disaggregate the information, because it does not think that that is useful. The information can be disaggregated, but, in policy terms, there is no necessity for or obligation on you to do that now.

I think that we need to separate the statistical nomenclature that we use from employment practice in schools. Through the statistical collection, we can try to reflect, as accurately as we can, practice in local authorities and in schools, by reflecting the specific roles, responsibilities and job titles that people in the system have. We have learned, through this exercise, that the categories that we had before—in particular, the two that we are talking about—no longer reflect accurately how support staff function in schools, given their roles.

You mentioned science teachers, and you are quite right to say that we do not simply ask schools, “How many teachers have you got?” We have a detailed and long-established process for collecting the number of teachers in the system on a given day in September, and—yes—we collect information about the primary subject that they teach. If someone’s main subject is physics or maths, we collect that information. However, we do not collect information about everything that that teacher does and how they spend all their time in school—they will no doubt undertake other duties that are not directly related to teaching their primary subject. That would be a very different statistical collection exercise.

With respect, you are making my point for me. There is a difference between a classroom assistant and an additional support needs professional, and if we are capable of distinguishing between a physics teacher and a history teacher, we should be able to distinguish between those two categories, too. We need to do that, frankly, if we are to ensure that all needs in the classroom are met.

When we discussed the issue with education authorities, they reflected that those two things are not as distinct as they have been previously and that there has been a move to use a role that authorities call “pupil support assistant”, which merges two functions. That is happening.

And we know why it is happening. It is being driven not by the needs of children with additional support needs but by the pressures in schools and on budgets. That is why schools are pooling and sharing classroom assistant resources, as opposed to directing to individual young people the ASN support that relates to their support plans. That is what parents are telling us. Someone who is supposed to be supporting one young person in a class will now very often be pulled across the classroom.

We can understand why schools are doing that, given the pressures on their budgets, but that does not make it right. For the stats to follow a decision that has been driven by budgets creates a problem in terms of our ability to understand what is going on in schools to support young people who have additional support needs.

My position is different from yours. In my discussions about the role with education authorities, discomfort was expressed about the two roles being described in that way. In practice, different terminology is used in authorities, and we should reflect that. That was the basis on which the discussion was held and the decision made. It went no further than that.

Have you come across families with a young person identified as needing additional support who have said that it is entirely acceptable for the pupil support assistant, who comes into the classroom to specifically support that young person, to be pulled into assisting the whole class? Are you comfortable with that as a policy?

I cannot comment on the way in which education authorities deploy their resource. That is entirely a matter for them.

Does the Scottish Government not have a view?

In terms of policy, the requirement on an education authority is to identify, provide for and review the additional support needs of the pupils in the authority. The provision and support should be tailored to the needs of the individual children. That is our position, which is enshrined in law. It is for education authorities to decide how they resource that requirement.

It feels that we are now in a position in which if we do not ask, we do not know.

We will move on. A number of members have indicated that they want to ask supplementary questions, but I must ask them to keep them quick.

I have a couple of quick questions about data quality. To ensure that we have the correct level of support in place, we need accurate data for the number of young people with additional support needs. If we compare the first bulletin that came out with the latest one to be published, we see that the school roll has increased by 2.3 per cent in eight years, while the percentage of pupils requiring additional support has increased from just over 10 per cent to 27 per cent. Is there a reason for that substantial increase? Has the quality of the data or the methodology changed?

Prior to 2010, the data collection on additional support for learning focused on the children and young people who were learning in special schools as well as those who had a formal plan such as a co-ordinated support plan or an individualised educational programme. That is a very narrow group of children and young people.

In 2010, we changed the statistical collection to include anyone who was receiving any type of support, whether in or outwith a formal plan and regardless of the education setting that the child or young person was learning in. As a result, there was a sharp increase in the number of children and young people who were recorded as having additional support needs in 2011 and 2012.

The data stabilised during the 2012-13 timeframe, which means that, in effect, we have a five-year period with a run of broadly consistent information. The sharp increase was caused by the expansion in the number of children and young people collected in the data.

In 2013, 131,000 pupils had additional support needs. That number is now 183,000, so it is still increasing.

Yes, there has been a continued increase. Children are continuing to be identified as having additional support needs and we continue to record that information, so the picture is one of rising numbers.

Looking at the different categories, I note that from 2010 to 2017 there was a 37 per cent reduction in the number of co-ordinated support plans, as you have mentioned, and a 19 per cent reduction in the number of individualised educational programmes, but the number in the “Other” category increased fivefold.

There has also been a significant increase in the number of children and young people with a child plan, which is another planning mechanism. It is a balancing out of rather than a reduction in the numbers, and a spread of different planning mechanisms is used.

Given that the “Other” category makes up 78 per cent of the pupils with additional support needs, is there a need for more categories to ensure that support is targeted properly?

In practice, the support is targeted properly. The “Other” category contains all the plan types that are not named. For example, educational authorities might use what is called an additional support plan. To me, that would be an individualised educational programme, but the authorities might not use that category when they respond to the collection, because it does not exactly match, and would therefore use the “Other” category. The personalisation and planning of support are there; it is just that the actual title of the plan is not there for the purposes of data collection. However, that allows us to collect information about a broad range of different planning approaches within our collection instead of not having the information available at all.

Will that level of detail be in future stats bulletins for education, because they will carry more categories?

No. That is not the intention at the moment.

It is worth adding that, with regard to those in receipt of a specific plan, we also collect data on the reason for an individual’s additional support needs—for example, mobility issues or autism. Most pupils are in the group that is outside those formalised plans. We have that level of detail, and we do not plan to change the method of collecting that information.

Just for absolute clarity, Mr Wilson, were the two previous categories brought into one because the individual categories were not consistent? With the additional breakdown, could you still get two councils reporting ASN as a category but having completely different support levels for that job?

I am sorry, but is that a question about the pupil data or the staff data?

It is a question about the staff data and ASN.

We think that, at the level of detail at which we collect the data, the descriptions of “ASN auxiliary” and “care assistant” that we had in the past do not match with the staff that authorities have in place now. Because there was no “pupil support assistant” option on the collection, some authorities were randomly allocating their pupil support assistants to one of those categories.

At that level of detail, there can be a difference between authorities that might have individual staff members doing precisely the same job in schools. One might record their staff member as undertaking the “ASN auxiliary” role, because it thinks that that is the closest fit or because that is what it has done in the past, while another might record all those staff against another category that it thinks might be a closer fit or because it is unable to differentiate between the two. That has led to the decision to amalgamate the two roles for publication purposes but not for collection purposes.

I return to the answers that were given in response to Johann Lamont’s questions. I wonder whether I am understanding correctly. In effect, from a Scottish Government policy point of view, do you no longer see a distinction between the two roles or do you think that the difference is not significant?

The roles are determined by the education authority; they are not determined by us—

You must have a view, from a policy perspective, as to whether the distinction is significant. I imagine that the Government is looking—as parliamentarians are—at the performance of local authorities and wants to compare practice and to find best practice to support the work of Education Scotland, and so on.

We recognise that there are a number of different roles—

But do you think that the distinction is important?

I will answer the question.

There are a number of different roles. There are pupil support staff, behaviour support staff, home-school link workers, school nurses, medical professionals, educational psychologists, classroom assistants, and ASN auxiliaries. Some authorities use those two specific categories, but the majority of them use the term “pupil support assistant” to describe the functions of a role that might formerly have been called a “classroom assistant” or an “additional support needs assistant”.

The way the authority uses the resource is determined by the support that it needs to give the individual children and young people in its classrooms. That is not determined by me and it is not determined by our policy, specifically. The Education (Additional Support for Learning) (Scotland) Act 2004 requires that provision be made for the individual child—

That is still not really an answer to the question whether the Government sees a distinction. Does the Government see the two roles as distinct and different in the classroom?

I recognise that there are a number of roles in a classroom—

No. I am talking about the ASN auxiliary role and the classroom support role. Do you think that those are two different and distinct roles or do you think that the difference between them does not matter any more?

In the past, they have been very distinct roles. Current practice means that they are not as distinct as they were previously.

10:45  

I have never been good at statistics and am now completely bamboozled by all the different categories that the panel has been talking about. Is there a need for clearer instructions to go out to authorities about how to make those important distinctions?

That goes back to the point that we made earlier, that we have to separate this into two parts. One part is to provide sufficient guidance to local authorities and grant-aided schools directly on how to complete the statistical return. We discovered some errors within the returns, as part of the process: authorities had not followed the guidance correctly and had recorded the wrong information.

Does the guidance that goes out make the clear distinction between support staff in general and staff who support additional support needs pupils?

No, and—as I said in my opening remarks—that is not an issue that should be determined by a statistical collection. The statistical collection should reflect the practice that is implemented in the system. It is for the system to tell us what the distinction is and what rules are in place, and it is for us to try to accommodate that as best we can in the statistical collection, rather than to say that we have in mind specific job titles or categories and what they mean for the collection, so authorities must sort out their staff return to match that.

Who gives that guidance and tries to make that distinction?

Do you mean in terms of the rules in schools?

Yes. Is it Government policy?

That is a matter for local authorities.

The local authority, as the employer, determines the roles that are carried out by their staff under the titles.

But who is giving guidance to local authorities? Should that come from COSLA or from the Government?

It is not appropriate for the Scottish Government to determine those roles.

I understand that. Who would give the guidance?

It is for the education authority itself to decide what the roles of its employees are. The authority guides itself as to what the roles are, in response to children and young people’s needs.

When roles are advertised, they are advertised for specific schools and establishments, and have attached to them a series of functions that the person will carry out under whatever title. A number of titles are used, even beyond the ones that have been discussed.

At the point that the survey is sent out, is any guidance issued with it?

Yes, there is guidance. We have a specification that is publicly available on the Scottish Government website that all local authorities will use when they decide how to assign their staff to the categories on which we collect information.

I am not any clearer.

Mr Halliday, if I picked you up correctly, you said that to extrapolate the information would be too much work and not value for money. Would you expand on that?

The point that I was making was that whatever statistics we produce, we must do so in a way that gives value for money. We must make a judgement about how much effort to spend in making sure that everything about a particular data set is absolutely right versus making sure that the vast majority of it is right.

As Mick Wilson said, the return forms part of a wider set of data collections about schools and pupils. Alasdair Anthony and his team are making judgements about whether to go back to local authorities, on the basis that practice within local authorities is clearly different and mixed.

I am struggling to understand how producing general data that does not give the Government the information that it needs represents value for money. I do not see the point of that.

Again, that goes back to a couple of the answers that we have given already about what the data are for. There are two questions. The first is whether we can, now that we have alighted on the pupil support assistant category, retrospectively create a pupil support assistant time series, if you like. We have learned from discussions with local authorities that it would be extremely difficult for them to go back and pinpoint when people moved from particular job titles or when the roles changed sufficiently, as Laura Meikle said.

The other question would be about how staff spend their time, rather than what staff the authorities have. That would be an incredibly difficult exercise for any organisation to undertake, whether it is a school, a local authority, the Government or anything else. To assign to specific tasks or actions the time of individuals who necessarily have broad remits would be incredibly difficult, and I question whether it would be worth putting resources into that.

You do not consider that to be a priority.

We have in place processes to identify and listen to the needs of users of the statistics, and to make judgments about how we can best marshall our resources to meet their needs. That might be through adjusting the survey or through collecting information in a different way.

I am picking up from this conversation a different set of needs to what we have been hearing from local authorities and others. I guess that the question for us is how we can factor the things that members are telling us into planning for future collections of data, and whether there would need to be some other way of getting at the information that you are describing. It would not necessarily be done by adjusting the current collection. We could spend ages trying to think about how to do that, but practices would make that difficult, given how the data collection is set up.

The consequent demand on local authorities’ time of trying to undertake such an exercise would be a very serious ask, and we would need to balance that against the quality of the information that we would likely get from the process. That is where Laura Meikle’s comments on the wider range of information and evidence that we can bring to the discussions become really important: there might well be other more appropriate ways of collecting the type of information that I think members are getting at, rather than through adjusting a formalised administrative data collection exercise.

Okay. Thank you.

We are still on supplementary questions. Mr Gray is next.

I want to go back to Laura Meikle’s point about the roles being a matter for the education authorities—the local authorities—as the employers. If you went to a local authority and asked to count teachers and it said, “We don’t call them teachers any more, so we don’t have any”, would that be okay?

No. That is not what I was saying. In taking the decision, we were trying, in effect, to align the data that we have more closely with practice out there in the education system.

That is not really the point, though, is it? The Scottish Government says to local authorities, “You must employ a certain number of teachers. If you fail to do that, we will claw back money from you.” Why is it acceptable, when it comes to pupils with additional support needs, to say that you are not going to count the support that they are given?

It is unfair to say that we do not count the support that they are given. We have a statistical collection—

Yes, but it is worthless.

—and a range of other evidence. It is not worthless—

Hang on. We have been told that the statistics that have been collected do not tell us how much support is being provided because pupil support assistants do other things as well. The support is not being counted.

Every teacher in Scotland provides support to children and young people. If we were to collect the range and level of detail that you are seeking, we would have to apportion part of a teacher’s time to one of the collections. We would not manage to count that—

But that is nonsense. What is asked for is data on additional support, which is support above and beyond what is provided by the classroom practitioner. That is what people assume you are counting when you count additional support in a classroom. However, we are told that authorities are saying, “Well, we don’t really deliver it that way”, and the Government has responded, “That’s fine.”

On your point about additionality, support that is provided to an individual pupil and not to the rest of the class would be considered to be additional, because it is additional to the support that is provided elsewhere. We would require to count a proportion of the person’s time, and we cannot collect that level of detail—it is not possible.

What we can do, and have done, is collect a range of information that gives us a baseline of information from which to work. We also rely on a far wider range of information to inform our policy decisions, as I said. We work with a range of people to inform our positions on policy. It is not just about the statistics. The statistics are important, but they are not the only—

The member who is sitting next to me has just shown me a live advert for an ASN auxiliary job at Knox academy, in Haddington, in my constituency. That is manifestly a member of staff being recruited to provide additional support. All the committee is doing is asking why the Scottish Government will not count those employees. You said that it is because the education authorities say, “We’re not going to tell you”—

No, that is not what we said—

No. I am sorry, but that is not true—

You do not accept that for teachers—

May I just cut in here? Teachers are a specific category—teaching is a profession and a recognised role. Are ASN auxiliaries doing different roles in different authorities?

Yes.

Over the past two years, the committee has taken a lot of evidence on a range of educational issues, and we have heard from organisations such as the Organisation for Economic Co-operation and Development and the Royal Society of Edinburgh, and from people such as Professor Lindsay Paterson and Keir Bloomer, all of whom have asked questions about the effectiveness of the data that is being collected about Scottish schools. This morning’s meeting seems to be flagging up considerable concerns about how effective the data is in relation to a specific issue.

How effective do you think that data collection is on all education issues? Concerns have been raised about the data in the context of curriculum for excellence, additional support for learning and a range of other areas. Data is crucial to informing policy. Are you content that we are collecting data in as accurate and as comprehensive a way as possible?

I am content that we have taken the correct steps to ensure that the data that we have is as accurate as it should be, within the confines and limits of the collections that we undertake—they all, necessarily, have limits and restrictions, and we have talked about some of those in relation to support staff.

I am familiar with some of the issues to which you refer, and they relate to a range of things—it would take us a long time to go through them all. The range of data that we have across Scottish education is very comprehensive and detailed. We have vast amounts of information about the mechanics of the system, by which I mean schools, pupils, teachers and other staff, and we have huge amounts of information on the performance of pupils and of the system itself. We take in all the data that the Scottish Qualifications Authority produces every year—all the exam results. We have data from Skills Development Scotland about outcomes for pupils at the end of the school process. We have the new collection, which looks at achievement of curriculum for excellence levels throughout the broad general education phase, and which lines us up with the national improvement framework and other things.

There are always questions about a system, whether for education of anything else. The data that we have—

11:00  

I will just interrupt, Mr Wilson, to ask whether you have an answer as to why individuals and groups that have strong pedigrees in educational research and data collection are raising serious questions. This morning’s meeting has also raised serious issues about this point, in my view and that of several colleagues. Why do they have those concerns, if you are telling us that everything is as good as it could be?

I do not think that that is what I said. I said quite clearly that there are always limits to the data that we have and there are always questions for which we might want evidence but which are not answerable through a data collection exercise. They may be answerable through other exercises, such as evaluations, research, surveys or discussions with stakeholders, in the way that Laura Meikle described earlier.

Some concerns that have been raised externally about information on Scottish education relate to how we use the information, rather than the information that we have. Some relate to the questions that we ask with regard to our data, rather than the scope of the collection. Some relate to a desire to maintain what are now historical collections on the system, whether that be the Scottish survey of literacy and numeracy, which comes up from time to time, or some of the school leaver surveys that used to be undertaken. Those views come from a range of perspectives, depending on the use to which people want to put the information.

Objectively, what improvements would you like to see in the data collection and the questions that are asked; you have just said that you do not think that they are always the ones that people might want to have answered. What improvements do you need to make to ensure that the data that we are provided with, as politicians, is better able to inform policy making?

The first thing is to make sure that we keep pace with the system, to make sure that the data collections match what happens in the system.

Do they not match now?

We have done our best to make sure that they do, but the questions in this meeting suggest that people are not necessarily content with how we are doing that.

Broader developments in data and how we use it as a Government would increase the power of the information. We now have a much clearer base to match data sets together; legislative changes have enabled us to look at the potential for matching information across systems. We will have the ability to match individual information and are looking, for example, at matching attainment records in schools to subsequent employment patterns and earnings potential—that is a particularly powerful piece of information that we cannot get directly through data collections at the moment, although we can try through other ways.

The impact of particular measures, policies, programmes and approaches to delivering education are always good questions. I am not a statistician by profession; I am an economist and have a broader interest in evaluative and performance information. Evaluation questions about the performance of interventions are always of interest and they are not generally answerable purely through data collection exercises. They require further exercises, surveys, research and statistical analysis.

Do you agree, though, that your work to interpret the information is very dependent on the data collection? The facts have to be there, both qualitative and quantitative, for you to be able to make an accurate interpretation.

Yes.

My question is about the borderline between what are considered to be official statistics and their background data, and how that line is drawn. In recent weeks, for instance, the committee has taken an interest in instrumental tuition in schools. Am I right that some statistics on music tutors in schools would be considered to be background data, or would they be considered to be in the scope of official statistics?

I will say something about the general case and will leave my colleagues to talk about the specifics of music tuition.

My role as chief statistician for Scotland is to raise the quality of our statistics, so that we can produce trustworthy, high-quality numbers that support decision making. I take a decision on whether something is an official statistic or does not meet that standard. We have strong processes across our teams to make sure that our statistics are trustworthy. We have good processes to make sure that our statistics have value across the piece.

It comes down to judgments about the quality of individual data collections. That is about our understanding of the whole statistical chain, from when the data is first recorded and what happens at that point, what happens to refine the data and how we assure ourselves of the quality of the information by the time it arrives with us, to the way the data is published to allow the users of the statistics to understand, interpret and use the information properly. That is my general approach. My colleagues will talk about the particular case.

In previous years, the data on music instructors has been released as supplementary statistics to the main headline statistics released in December. For the 2017 data, based on the information that we received from our quality assurance processes, as well as the changing context and environment into which we are releasing the data, we took the decision to release only certain categories as background statistics. The rest of the information we made available as management information. That was an interim position, while we considered the most appropriate way to make that data publicly available and serve the need and demand for the information.

I have a point about quality assurance. Alasdair Anthony said earlier that the title “pupil support assistant” is what is used in schools. I used to work as a teacher, and in 2014 the City of Edinburgh Council had a policy whereby classroom assistants provided administrative support to principal teachers. Every two weeks, they would come to my department and help with data entry. That role of classroom assistant in Edinburgh was different from what it was in Fife, when I worked there. In Fife, “classroom assistant” meant somebody in the classroom supporting the class or providing pupil support. How do you quality assure so that people who are not PSAs do not end up in the category?

To an extent, we rely on what the local authorities tell us. They are the ones that are categorising staff. That information will be based on how they advertise the roles and the types of role they are using in the schools.

On quality assurance, we look at how the information has changed over the years, how it compares to information from other local authorities and how it compares between schools within a local authority. We seek to draw together a coherent picture of how the categories are being used. Through that work, we have taken the decision to present the data for ASN auxiliaries and classroom assistants under the new category of pupil support assistants.

I am saying that the same job title can mean different things in different parts of the country. Do you check that what you are gathering is the right data?

We do not undertake specific examinations of whether an individual member of staff is performing a particular role within a school. That would almost be an audit of the employment practices of the authority. The guidance that we issue describes what is in scope for the data collection. Volunteer parents supporting pupils within classrooms, for example, are excluded from scope. We do not count them. We are clear to local authorities that they should not be assigned to one of the staff categories. As Alasdair Anthony said, we are reliant on local authorities providing us with correct information about the way they use their staff.

I will ask about the background data for school technicians. The committee is about to undertake work on the Scottish Government’s science, technology, engineering and mathematics strategy. As recently as July 2018, there have been articles in Tes Scotland about a reduction in school technician numbers by nearly 500 since 2005. Is the situation for lab and school technicians the same as for pupil support, in that roles are different in different schools and they are changing?

We have not undertaken specific additional quality assurance on technicians and laboratory assistants, which is why we have made that information available separately as management information at this stage. When we look at it in more detail, we will be able to assess how those categories are being implemented by local authorities.

Some school technician roles are advertised as term time only, and some local authorities employ people full time. The role involves a significant amount of repair and preparation, so a full-time role is different from one authority to another. Does your information capture that difference in any way?

We have made information and guidance available to local authorities on how they should calculate a full-time equivalent, which takes account of roles that are term-time working and those that are throughout the year.

In both technician and pupil support roles, does the data collection categorise the professional qualifications of the person doing the role?

That is left to local authorities to determine; if a person is described as a technician by the local authority, that is how we would expect them to be recorded.

I assume that Mr Halliday is in charge of statistics across all of Government. When you make a significant change to how statistics are collected, how do you tell the world that you are doing that?

We have a ScotStat network, which has a few thousand users of our statistics. The process has a series of themed groups, including school education. First, a consultation would happen with that group, partly online and partly face to face—it would depend on the individual process—

Did all the process that you have described happen with the example on ASN that we have discussed all morning: the merger of two categories into one?

I will leave my colleague to talk about that specific example.

I would first like to clarify that we have not made any changes to the data that has been collected. The categories that are collected from local authorities are the same as they have been for a number of years. The changes that have been made are to the presentation of the data; when we presented it, we highlighted that additional quality assurance had been undertaken on the data that we were publishing and that the rest of the information was available on request.

However, you did not explain why you changed the presentation.

That is something that we want to improve the next time we publish the data, based on feedback from here and other places.

We could not find any minutes or anything else that explained why the presentation changed. Do you accept that that was not good?

We could have made it more explicit to users that we had combined the classroom assistant and ASN categories.

That is fine. Mr Halliday, in future, if you make a statistical change to the presentation of information in a policy area that a parliamentary committee has been poring over, such as ASN, can I take it that someone in some part of your organisation will say,“Wait a minute—there will be a lot of parliamentary interest in this, so we should make clear why we are changing the presentation of that information”? It could be the same for economic statistics or anything. It is a matter of judgment, of course.

11:15  

I would certainly expect that from my colleagues.

It did not happen here, so that obviously has to change.

I clarify that the change with regard to support staff was reactive to issues that were unearthed during the quality assurance process. We cannot tell people what we are going to do about that until we have gone through the process of working out what the issue is and the correct solution. That is different from a planned change that we might make to collections, such as stopping some collection or changing timing or methodology. We are required to consult more broadly on bigger changes to our collections for purposes of official statistics or broader legislation.

I understand the differences.

This is a slightly different case and we did put out information at the time of the change.

I also want to ask about your point earlier to Ross Greer. Are you seriously saying that you treat every request for information as an FOI request? Is that a matter of Government policy?

We treat non-straightforward requests for information—

What is the difference with a non-straightforward request?

They are requests that are not trivial to answer. The issue is the amount of time—or rather effort—that is taken to respond to those questions.

Are your criteria to do with how long it will take your statistical unit to respond to the question? Will that determine whether the request will be treated under FOI or whether you will just provide the information?

I guess that that is how it is done in practice.

Mr Scott, I think that the FOI legislation puts specific responsibilities on to any Government about what is expected—

I know that, but I am asking about the culture.

I understand that, but the question is more general and perhaps the committee could explore it across Government to get some of the information.

Mr Halliday is the head of statistics, convener.

He has given the answer as the head of statistics, but the issue that the committee can explore is wider.

Can I ask my final couple of questions?

Yes, absolutely.

With regard to the changes that you are contemplating, which will help us to understand the presentation of your figures in future, can you make sure that the committee understands that that will be in place from now on? Will we understand who makes the call in the process by which changes will happen across important policy areas?

I expect of our colleagues that they will, as part of our publications, describe changes that have happened to data collection or to significant changes to presentation—not necessarily all the minutiae.

We have said a couple of times that the landscape shifts regularly with regard to interest in particular pieces of information. We collect an awful lot of data. Some of it is of particular interest to a minority of people; some stays like that throughout its life and some does not. This falls into the category of data that has come up the level of interest tables. If we know in advance that we are making substantial changes to any data collection, we let the relevant people know. That is certainly what we will do when we have been through the process to look at our data on support staff categories. If we end up looking at making substantive changes, we will make sure that the broadest range of people, including you, are aware of it.

I appreciate that. Given how important and topical public policy in the education sphere is at the moment, do you pay particular attention to how education statistics are presented?

Yes; it is our job to do that—that is what we are there for. The increased interest in this information is double edged for us. It is great that our data is in the spotlight and that we can bring it to bear in the debates and evidence. However, as I said earlier, that brings a responsibility to make sure that it is fit for purpose and keeps pace with practice in the system.

Therefore, you understand why it is difficult for us to understand why Ross Greer did not have that information provided, given that it was a matter of such interest to this committee and, indeed, the wider public policy sphere.

I am not clear what information you think he was not provided with. We answered his question, given the—

We have been through it for an hour this morning. I am not going to go over it again.

I add one further point about future data collection. Last year, a regulation that was specifically about collection of data on additional support for learning was agreed to. It places a requirement on the Scottish Government to consult publicly if we seek to make substantial changes to the collection of that data. We would consult the committee as part of that process.

We have been informed that the number of categories to be published is being reduced from 21 to five, but management information will be available on request. Is it the case that we will not have to put in an FOI request for the narrowed information, because it will be management information that is available on request, so we would not have the same circumstance again?

I think that you are making a slightly false distinction between the processes. How we handle a request depends on the nature of the request. Routine requests for standard information—

With respect, this is information that used to—

I am sorry, Ms Lamont. Can you let Mr Wilson finish his point?

My apologies.

As I said earlier, in the particular case, because we were asked to release information that we had previously deemed to be of a lower quality in terms of the distinction between the two categories, we felt it was appropriate to go through the formal process for FOIs, for which there is a set of requirements and restrictions on us.

We receive a vast number of requests for information and data throughout the year and a lot of them are handled as freedom of information requests. However, some of them require protracted discussions and debates with the requester about the nature of the information and the level of detail that they require—whether they want personally identifiable information and so on. We cannot handle those requests through the FOI process, because that does not enable us to have that on-going debate.

That was not my point. Are you giving us the reassurance that you are reducing the number of categories from 21 to five but you are still gathering the management information and it will be available on request?

Yes.

It is not complex. It is what is already there. We want reassurance that, although you are reducing the number of categories that you report on, you still have the other information available.

On the matter of additional support needs—the nature of the support that a young person has and why it matters—Jenny Gilruth highlighted well how the catch-all description may not capture what is happening in our schools. When we look at teachers, we can identify all sorts of things about their age, gender, main subject, contact time, whether they are training as a chartered teacher or on headship training programmes and whether they can teach in a Catholic school or a Gaelic-medium school. We gather all that information, but are we able to gather information about the skill set of people who are working with young people who have additional support needs? That is a question for Laura Meikle.

It is not; it is a statistical question.

No—I am asking you whether it is possible. Do you want information that tells you that somebody who is a pupil support assistant has received training or qualifications in autism awareness, working with young people with learning disabilities or whatever? Given the scale of the information that we collect from teachers, my concern is the implication that there are schoolteachers and everybody else just does stuff around them, when in fact those jobs are highly skilled professional jobs in their own right. Do you try to capture the scale of professionalism among people offering additional support?

As I said earlier, I use a wide range of evidence to inform our policy decisions.

Would you ask local authorities about the qualifications and professional training of people who are supporting young people with additional support needs? This relates to the point that Gordon MacDonald made earlier about the importance of a match-up between the identified additional support needs and the support that is being offered. Do we attempt to capture the qualifications and skills of people who are working with young people with additional support needs?

The discussions that I have with a wide range of people, including education authorities and parents and families, touch on those issues. I would not use a national statistical collection to do that.

So you do not touch on those issues in discussions about teaching, but you do when it comes to the really important issue of the scale of expertise and professional qualification of the people who are offering additional support in schools.

It depends; how I use the information may be different. For example, if we are considering what additional training is required nationally for people working with children and young people around inclusion or autism, I would ask specific questions. I would not necessarily use a national statistical survey to do that, because I want to find out a whole range of other information at the same time. I would use the engagements that I have talked about to cover a whole host of issues. When we are talking about training, that could include how we can best deliver the training, what people need to know, the balance of information that people require and how we can ensure that any training that we provide will be given professional recognition through the General Teaching Council for Scotland. Those are the things that we do.

Given that we are suggesting that support for staff in classrooms with young people with additional support needs—the demand for which is increasing, as Gordon MacDonald identified—is becoming more generalised support for staff, one way in which you could establish whether it is just the job title that is different, rather than the support, is to ask people in the ASN support category whether they have a professional qualification or training. It would be for other people to decide how that question should be framed, but do you think that it is a reasonable question to ask? Do you want to know the scale of professional qualification among people who are working to deliver ASN support in our classrooms?

The requirement would be to move from an aggregated return, which gives us the number of staff in these categories, to an individual-level return for all types of support staff. We have that information for teachers—and for pupils, as it happens—because we have an individual-level return for each teacher working in the system. That allows us to ask for details about those individuals.

Why would we not do that?

We do not currently do that for support staff, partly because it would be a significant additional burden on the providers of that information—local authorities, in particular, and the grant-aided schools—to extract yet more personal-level detail on those individuals. It also depends on the use to which we will put those data. For example, teacher workforce planning processes require us to have a level of detail about things such as the demographic profile of teachers to enable us to look at the factors that influence future demand for teachers.

We are not involved in a formalised workforce planning process for support staff, and we would need to think carefully about and consult publicly and more broadly on changes to that data, because it would definitely represent a change in the methodology for that type of information. It would require a substantial and formal process to look at the potential changes needed to get that level of detail.

However, as I said in my opening statement, it is not the case that what we have always done is what we will always do. We do change statistical collections, introducing further detail in some things and withdrawing detail when it is no longer appropriate. For example, we are making changes to the early learning and childcare collections to reflect information about qualifications and report on the commitment to have additional graduates and so forth. Changes such as that are plausible, but they cannot simply be made unilaterally by us to reflect a particular demand for information.

I accept that you have made significant changes; all of them have been to give us less information than we had before about the nature of the support that is available in schools. Given the figures that were highlighted by Gordon MacDonald on the scale of need and the reports from families and others who say that the needs of young people are not being met, perhaps now is the time for workforce planning for people who are delivering additional support needs in the classroom and to gain a proper understanding of what those people are able to do.

The issue is about diminishing the important job that is being done in schools by generalising it in a way that means that we do not really know, and we are not even asking, how skilled those people are, how many of them there are and how many of them can be identified as providing individual support for young people and their specific needs. That would be as opposed to the very general categories that have been highlighted elsewhere.

11:30  

As part of our consideration with the advisory group for additional support for learning, we are looking at the collection of data on additional support for learning, but we are considering it in a slightly different way. At the moment, the collection focuses on input rather than outcomes information and, through a range of different measures, we are trying to work through whether we can change the information that we gather to look at the difference that the support has made to the individual child or young person who has received it. To do that would mean looking at different ways of collecting information beyond the statistical collection that we have discussed this morning, and we are currently discussing that with the advisory group for additional support for learning to consider what those different ways are.

Is there a need for workforce planning for additional support needs staff that recognises and values the job that those workers do?

We already value and recognise the job that additional support staff do. That is not in question.

However, you have put them into a broad category in which they might be doing that work, or might be doing something completely different.

I say, with due respect, that the fact that we have made a decision to draw together two categories in a census does not devalue respect for the people who provide support to our children and young people in schools.

I have already listed all the things that we ask our classroom teachers. We do not even ask those groups of people about their qualifications, the appropriateness of their training or their or capacity for supporting young people who have additional support needs. We have just lumped them all together, so it is now a category that does not tell us very much at all.

With partners in the advisory group for additional support for learning, we can consider those matters when we are doing our broader work on collection of data on additional support for learning. However, as part of that process, we also need to take account of the views of COSLA and the Association of Directors of Education in Scotland, which we will do.

I am sorry, but are COSLA and ADES saying that they do not want workforce planning for people who provide support to those with additional support needs?

No, but—with respect—COSLA and ADES are the employers of the people to whom we are referring.

Do you know whether COSLA and ADES think that there should be workforce planning for their work?

The need for workforce planning has never been raised in any of the fora with which I have engaged. I am happy to have that discussion as part of the data collection discussions that we are having with the advisory group for additional support for learning.

I do not think that anybody is entirely fixated on the data; it is about the quality of the support that is offered to young people. The people who are collecting the stats and trying to ensure that the stats are as robust as possible have one job to do, but there is a separate job that is about understanding what the stats tell us. If you do not ask the question, you will not know. The grave concern is that we do not understand the scale of the problem and the challenge for the teaching and support staff in ensuring that young people are properly supported.

As I have already said, as part of our work to support the implementation of additional support for learning, we look at a very wide range of evidence to inform our decisions. It goes beyond the strands of information that are available to us from the statistics. Therefore, we have a good understanding of the implementation position, and the information informs the actions that we choose to take to support implementation further.

Mr Wilson seemed to suggest earlier that he would be willing to look at different questions. Would you consider ideas for new questions that have come from the committee at today’s meeting?

The short answer is yes. We always look at the wide range of users and their interests in our data. The Parliament and committees are part of the user group, so we will, of course, look at those ideas.

However, as I said, some of them are not questions that can simply be tagged on to the current process; they would require a fundamental change to the way in which we collect education data across the system. That would come with an associated process and set of requirements for us, so the decision would have to be balanced with those impacts.

Before you designed the questions, how much discussion did you have with individual local authorities about the data that they were collecting? My local authority collects data on, for example, how many hours of one-to-one support it provides to young people, because that has to be agreed at regional level. Do you ask local authorities what data they collect?

Yes. We have an extensive and on-going process of engagement with local authorities. At the beginning of the session, I outlined a list of the data collections that we bring in. They require us to have pretty close relationships with a range of people in local authorities, from directors of education and children’s services down to management information specialists and data providers.

We get extremely useful feedback from them about how they capture the data from pupils, parents and teachers, how they store it in their management information systems, how the data does or does not match the way in which we seek to extract it, and the terminology and guidance that we issue. We fairly continuously update the guidance to reflect changes—some of which have come as part of the process on support staff.

We know that a lot more information is held in the management information systems—data that is held by schools and local authorities so that they can run the education system that they provide. We do not uplift all the data that they hold because that would be an unfeasibly huge exercise that would leave us with data for which we would have no practical use.

We have an on-going conversation with local authorities, and there is a specific network that we bring together at least once a year to talk about current or forthcoming issues and to explore options, as we did in the case of the pupil support assistant category.

In your professional experience, is there more useful information out there that you are not picking up at the moment? I am not being critical: it might be that, in the past, those avenues were not explored fully because the matter was not a priority. Is that a possibility?

Absolutely—I would be surprised if that was not the case. The question remains whether we need to source that information regularly and routinely, as we do with the information that we are discussing, which we collect annually.

If local authorities or any other administrative data holders have information that we have a particular need or use for, there are mechanisms that we can use to extract it on a one-off basis to give us a snapshot of what is going on, rather than establishing a reasonably onerous annual process of getting the data.

There are options. I would be surprised if there is not more data that would be of use.

I have a final question that follows on from that. You mentioned the possibility of doing one-off work. Given the concern of colleagues on the committee about the change, could you do a one-off bit of work to probe the change, so that the committee and Parliament could understand what the merging of the categories has meant?

It is possible to do such work. I am not currently in a position to think through all the various options that might be available and the correct mechanism for doing it. We would have to think about whether it would be done through working with the advisory group—as Laura Meikle described—through qualitative information from a select group of local authorities that have already given us information through this process, or through a one-off data uplift of some sort. If there is demand and need for such information, there are broader analytical questions that we could ask, rather than the statistical questions on which data comes to us regularly.

In theory, there is nothing to stop you from doing, for example, a one-off survey to get a snapshot of who is doing what in ASN on one day in September?

Philosophically, there is nothing against that idea, although there might be practical barriers and other limits to how we would conduct and fund it.

I will stick with the issue of the individual versus the aggregate method of collecting data. Like my committee colleagues, I am not a statistician, so I may be misunderstanding the issue.

My understanding is that the collection of individual teacher data allows for distinctions to be made in relation to, for example, teachers who deliver multiple subjects—mathematics and physics, or history and modern studies. That collection method means that the teacher is not double counted as both a maths teacher and a physics teacher, but is also not counted as just a single full-time maths teacher. That distinction is made, which in turn informs, for example, the committee and Parliament about what the full-time equivalent capacity is for delivering subjects such as maths and physics.

I understand that the shift of collecting the data on support staff from the aggregate level to the individual level would be a considerable one. Will you clarify the process for me? Who would make the decision about whether to seriously explore that option?

There are two things in that. I will clarify a matter for you on the teacher side. You are right to say that we collect information about the subjects that individual teachers provide. We have information on their main subject, but we also have information on other subjects that they can teach. However, we do not assign proportions of their time to those subjects. We may know that a teacher who is working full-time in a local authority is primarily a physics teacher, but can also teach maths, but we do not know from that information whether they teach any maths, or whether they spend half their time teaching one subject and the other half teaching the other. We cannot measure, in the way that you described, exactly what teaching resource is provided in practice to individual subjects.

On how we would decide whether to move to a different collection process, that is, in many ways, an organic exercise. Need and demand would arise from users, and that would be discussed—certainly, internally first, in the way that we have described. In respect of the current change, we would at least discuss with the data providers the feasibility of asking them for different types of data. It is good to have a starting position of knowing whether it would even be feasible for them to do that.

There would then be broader discussions with Roger Halliday, in his capacity as chief statistician. If that idea looked like something that would be worth exploring, we would develop options. A formal public consultation on such a change would be needed, because it would be regarded as a significant change to the collection methods for official statistics. As Laura Meikle said, in terms of ASN in particular, other legislation requires us to consult on such changes before any such decision can be taken.

There is an appetite from the users here in the committee for that possibility to be at least explored.

That is clear.

On the point about consultation of other users and stakeholders, and to return to an earlier discussion, Laura Meikle mentioned to Johann Lamont that the advisory group for additional support for learning was being consulted. Was it consulted on the change in how statistics are published?

No. When we had the discussion with the advisory group, it was considering a particular piece of business on statistical information. We were reporting to the group the fact that the issue had arisen and that additional quality assurance was needed in order to make sure that the information that we provided was robust. As part of that, I explained that we were considering a number of ways to resolve that, but we did not consult the group specifically on the matter that we discussed. I laid out to the group a number of approaches that we might take, but—

If you do not mind me asking, why not? Surely the advisory group for additional support for learning should be sounded out on something like this.

Absolutely.

Once you have a proposal, you can take it to the group of people whom you have assembled as being the relevant experts and people who have an interest and ask them what the implications would be.

We did that as part of the discussion of the different options. I was aware of their views. As I said, we did not put to the group a specific question along the lines of: “Do you think that this approach is what should be done?” Rather, it was asked, “These are the things that we might do, so which of those feel appropriate”—

Just to be clear, are you saying that the approach that was ultimately taken was one of the specific proposals laid out to—

Yes, it was one of the things that we discussed.

What was the feedback on that proposal?

There was no concern about it; it was felt to be appropriate.

We regularly discuss statistical information and other evidence as part of our work. It was not a non-routine discussion, if I can put it like that. The discussion was about the type of business that we carry out in that arena, and there was the discussion that I talked about earlier on potential changes to the data collection in terms of moving to outcomes. It was a routine discussion—it was not a stand-out discussion for the group. That is the only way that I can describe it to you.

The specific question was not asked was partly because I was already aware of the group’s position on the issue and partly because the group did not happen to meet at that time.

11:45  

I will move to a different but relevant area. Do you know what is causing the quite significant discrepancies between the data that you are collecting through the census and the data that are being issued by local authorities in response to freedom-of-information requests from external organisations for the same information?

That issue crops up from time to time when people compare official national statistics with information that is issued under freedom-of-information requests. Although FOI requests are often for specific information that, on the face of it, would seem to match what the official statistics report on, in fact there is no exactly clear and equitable match.

I will give an example. When you ask a local authority how many teachers it has, it might return to you the full-time equivalents or it might return to you the head count. Obviously, those are two different numbers. You could, having asked the same question, have been given one of two different answers.

That is reasonably common across a range of areas. Sometimes, FOI requests that have been sourced from 32 local authorities come to us for review, and sometimes we see the numbers in the media and try to reconcile them with the numbers that we hold centrally from similar sources. From my now fairly considerable experience across a range of areas of government, I say that it is rare that the numbers match. The statistical collection comes with guidance on how to complete the return, and what should be included and what should not; an FOI request does not, and nor should it. That, however, will necessarily lead to discrepancies between numbers that are returned.

I am aware of some of the inconsistencies. My office is trying to compete with certain journalists in Scotland in the number of FOI requests that we send out—

Good luck.

—and we are trying to resolve the inconsistencies in what comes back. An example that has been raised with the committee, which you might be familiar with, relates to work that was done by the Scottish Secondary Teachers Association, in which there are some considerable inconsistencies. The example is additional support needs teachers in Dumfries and Galloway on a particular day. The SSTA was informed that there were 92 ASN teachers, whereas the census indicated that there were 38. That is quite a considerable difference, and is more than any difference that I have encountered previously in relation to FOI data and census data.

I have not had a chance to investigate that issue but—off the top of my head—a number of things could have contributed to the discrepancy. I have already mentioned the difference between FTE and head count. Also, information will have been gathered at different points in time and there will be definitional differences. We see, too, a bit of inconsistency in how the term “ASN” is used. Some local authorities will talk about ASN schools, whereas we would refer to them as special schools. You might hear us talk about the number of teachers in special schools, rather than the number of teachers whose main subject is additional support needs.

I am aware that I asked about a specific example and that if you do not have the information in front of you, you cannot provide the full context. In any further work that the committee undertakes on this, we would benefit from a written response from you about why you think there is an inconsistency in the example.

Is there a statutory duty on local authorities to return the census information?

The Education (Scotland) Act 1980 requires local authorities to return data to the Scottish Government. There is a broader duty to return education data such as we require; the return of the census itself is not specifically described in legislation.

Okay. Thank you very much for your attendance at the committee this morning.

11:50 Meeting continued in private until 12:11.