High Court dismisses judicial review of Leaving Certificate grading, upholding standardisation model fairness.

APPROVED [2021] IEHC 128

THE HIGH COURT

JUDICIAL REVIEW

[2020 No. 655 JR]

BETWEEN

FREDDY SHERRY

APPLICANT

AND

THE MINISTER FOR EDUCATION AND SKILLS, THE MINISTER FOR

FURTHER EDUCATION AND HIGHER EDUCATION, RESEARCH, INNOVATION

AND SCIENCE, IRELAND, AND THE ATTORNEY GENERAL

RESPONDENTS

Judgment of Mr. Justice Charles Meenan delivered on the day of 2nd March, 2021.

Contents

Introduction ................................................................................................................................ 2 Calculated grades ....................................................................................................................... 4 School historical data (SHD) ..................................................................................................... 5 National historical data (NHD) .................................................................................................. 5 Government decision of 1 September 2020 ............................................................................... 6 Leaving Certificate results of 2020 ............................................................................................ 6 The issue .................................................................................................................................... 8 Evolution of the standardisation model ..................................................................................... 9 Justiciability ............................................................................................................................. 24 Submissions ............................................................................................................................. 24 Evidence ................................................................................................................................... 30 Consideration of submissions .................................................................................................. 32 Legitimate expectation ............................................................................................................. 35 Fairness .................................................................................................................................... 38 Public interest........................................................................................................................... 46

Conclusion ............................................................................................................................... 49 Summary .................................................................................................................................. 49 Consequential orders:............................................................................................................... 53 Appendix A: ............................................................................................................................. 54

Introduction

1. The fight to contain the spread of the Covid-19 virus has involved the imposition by the State of restrictions on almost every aspect of our lives. The young, the old, those with disabilities, and those without have all been affected, some to a greater degree than others. To halt the spread of the virus, it is necessary to limit, if not restrict entirely, the circumstances under which people meet and congregate.

2. In excess of 62,000 students were due to sit the Leaving Certificate in 2020. This obviously involved the congregation of significant numbers of people in close proximity in an indoor setting for prolonged periods of time. As the date for the commencement of the Leaving Certificate 2020 approached, it was apparent that, though the spread of the virus had at that stage been significantly reduced, for reasons of public health, the examination could not take place. For the first time in the history of the State, by a decision of the Government of 8 May 2020, the Leaving Certificate due to start the following month was cancelled.

3. The Leaving Certificate has a central role in the Irish education system. For those who wish to advance to third level education, the results of the Leaving Certificate are, for the most part, a basis for entry to a particular course and a subsequent career. For others, who do not wish to pursue third level education, the Leaving Certificate is a qualification that may be necessary to obtain employment. The importance of the results of the Leaving Certificate cannot be overstated for young people who wish to pursue a particular career or, indeed, for more mature people who may wish to embark on a new and different career. Without a Leaving Certificate, the class of 2020 would have been left stranded, so it was imperative that an alternative system be devised to give an accurate assessment, as far as possible, of the standards that would have been achieved had the exam proceeded in the normal way. The first named respondent (the Minister) did not have the option to simply defer Leaving Certificate 2020 to 2021.

4. The whole Leaving Certificate system has been the subject of comment and criticism over many years. However, it is difficult, if not impossible, to replicate by an alternative system the fairness of the Leaving Certificate exam. Students doing the Leaving Certificate exam come from various and diverse backgrounds. Families of some students have the financial means to send their children to “fee-paying” schools and to provide additional education by way of grinds. Other families who do not have such financial means may, with great sacrifice, enrol their children in “fee-paying” schools. Very many other families simply cannot afford this but want their children to do a good Leaving Certificate and, notwithstanding the inequalities in the system, have equality of opportunity. However, at the end of the day, all students do the same exam. The correction of each subject in the Leaving Certificate is done entirely anonymously and according to guidelines which, prior to their adoption, have been considered in detail by the relevant experts.

5. The task facing the Minister in devising a system that would give to a student a grade that he or she would have obtained had the Leaving Certificate exam been sat in the normal way cannot be underestimated. Further, this task had to be completed to the satisfaction of students, teachers and the wider public in a narrow timeframe. The system devised was the awarding of “calculated grades”.

6. The basis used for the award of a calculated grade raises fundamental issues, not all of which are legal. It is a fact that certain schools, in particular “fee-paying” schools, have historically achieved stronger Leaving Certificate results than other schools. Should the calculated grades being awarded in 2020 reflect this? Also, grades achieved in the Leaving Certificate may vary depending upon the gender and socio-economic background of the candidates. Should this also be reflected in the awarding of calculated grades? It would not be acceptable to many to factor into the system for the awarding of calculated grades inequality, since it must be a basic principle of any for education system to achieve equality of opportunity. Thus, a system that may, statistically, give an accurate result would not be acceptable to the public at large. It was the requirement for public acceptance that was central to the decisions taken by the Minister and the Government that are the subject of these proceedings.

Calculated grades

7. There were two phases to the system for the awarding of calculated grades. Firstly, a schools phase; and, secondly, a phase overseen by the Department of Education and Skills (the Department).

8. The school phase involved the award by the relevant teacher(s) of an estimation of the percentage mark in each subject that each candidate is likely to have achieved if they had sat the Leaving Certificate examination in 2020 under normal conditions. The school then carried out a “class ranking” for each student in each subject, i.e.: a list of all the students in each individual class group for a particular subject in order of their estimated level of achievement. Amongst the matters which informed estimated marks and rankings were the previous results of the school in the particular subject.

9. It was correctly anticipated that the estimated marks coming from the schools were likely to be overestimated and considerably ahead of what the particular school had achieved in past years. This is not an adverse reflection on the professionalism of the teachers involved but, rather, of the difficulties faced by teachers in having to give marks to their students in what, for many, would be the most important examination of their lives. Thus, simply awarding school estimates as calculated grades for the Leaving Certificate of 2020 was not a viable option, as it would undoubtedly lead to hyper “grade inflation”.

10. The Department phase involved the consideration and design of a “standardisation model”, which would be applied to the estimated marks coming from the schools. This was with a view to making the calculated grades that would be awarded statistically accurate and in line with previous years. Hence, the proposed use of “school historical data” and “national historical data”. School historical data (SHD)

11. This was data based on historical Leaving Certificate examination performance for the particular school across three prior years. It is the case that the performance of students in each school does not vary widely from year to year and, so, it was considered statistically reasonable to assume that the cohort of students in any individual school in 2020 would not perform very differently to that of previous years. It should be noted, as I have referred to elsewhere, that historical school performance in each subject was a matter that should be taken into account at school level in the giving of a calculated mark. Also, data concerning prior performance in the Junior Certificate was part of the standardisation model and remained in place. National historical data (NHD)

12. This is data of student results on a subject by subject basis based on historical Leaving Certificate examination performance. The reason for using this data in the standardisation model was to ensure that, overall, the calculated grades awarded in the Leaving Certificate of 2020 were in line with the results of previous years. The purpose of applying this data was to avoid “grade inflation”.

13. In order to avoid “grade inflation”, not only was NHD to be used in the model, but also another mapping exercise at a national level to map the scores from the entire cohort for the examination to the historical national distribution would have to be used. This is referred to as the “mapping tool”. As we shall see, this “mapping tool” was not used.

Government decision of 1 September 2020

14. At its meeting of 1 September 2020, the Government took two decisions on the data that was to be used in the final version of the standardisation model for the awarding of calculated grades: -

(i) School historical data (SHD) would be removed from the range of data being used in the standardisation model, however, data from previous Junior Certificate examinations would still be used; and

(ii) The national historical data (NHD) would be retained in the standardisation model but the “mapping tool” would not be applied, and so the impact of this data would be minimal.

Leaving Certificate results of 2020

15. Results were issued to students entitled to receive calculated grades on 7 September 2020. Students also had available to them the estimated marks from their schools, as were submitted to the Department. Thus, students could compare the calculated grades that had been awarded with the estimated school marks. The media reported widely that students who had attended schools or other educational establishments which historically had good Leaving Certificate results had been awarded calculated grades below the estimated marks that had been submitted. It was maintained that this “downgrading” stemmed from the Government decision to remove SHD from the standardisation model.

16. Some days after the release of the Leaving Certificate results, the Central Applications Office (the CAO) published the points required for entry to various third level courses. It was immediately apparent that there had been a considerable increase in points as a result of “grade inflation”. Though not highlighted at the time, this would appear to have been as a result of the

decision not to use the “mapping tool”.

17. The applicant was amongst those who believed that he had been unfairly downgraded. The applicant attended Belvedere College in Dublin for his secondary education. He stated in his Grounding Affidavit that he studied consistently for the Leaving Certificate up to the time of the announcement of 8 May 2020 that the exam was cancelled and would be replaced by a system of calculated grades. His Leaving Certificate subjects were English, French, Chemistry, Biology, Irish, Latin and Maths. He received his calculated grades on 7 September 2020, and received his teachers’ estimated marks on 14 September 2020. The applicant had apparently been downgraded. He stated that the result of these downgrades was that he did not have the points necessary for his third level course of choice. I will examine, in some detail, the calculated grades that were awarded to the applicant and the influence, if any, which the use of the SHD would have had if it had been applied.

18. An affidavit was filed by Mr. Tom Doyle, Deputy Principal of Belvedere College. In the course of his affidavit, he refers to not only the Leaving Certificate results of the applicant but, also, to those of other students in the school. He maintains that the absence of SHD in the final standardisation model adversely affected the Leaving Certificate results, which had been historically strong, in Belvedere College. Later in the judgment, I will also consider whether, in fact, Belvedere College has been so adversely affected.

19. In addition to the applicant herein, close to 70 other sets of proceedings were initiated concerning the award of calculated grades. As the issue of SHD was common to many of these actions, the Court directed that the parties involved select one test case so that this issue could be determined. The parties selected the instant case. Both the Court and the parties were conscious that the issues involved had to be resolved as soon as possible in a tight timeframe. This was necessary since the Leaving Certificate of 2021 was then only a matter of months away. It is to the credit of the parties in this case, and their advisors, that this difficult and complex case was ready for hearing in a matter of weeks. The hearing of this application commenced on 8 December 2020 and concluded on 2 February 2021, some five weeks of hearing.

The issue

20. At a hearing towards the end of November, 2020, a number of preliminary matters were decided by the Court, including: the inspection of various standardised models, crossexamination of deponents of affidavits and whether executive privilege was applicable. The Court directed that the following issue be determined: - “That the decision of the first and fifth named respondents of 19 August 2020 and/or the decision of the first named respondent of 21 August 2020 and/or the confirmation of the said decision by the sixth named respondent on 1 September 2020 to alter the standardisation model so as to exclude the use of all school by school historical data (SHD) on the performance of students in past cohorts in each subject was arbitrary, unfair, unreasonable, irrational and unlawful and in breach of the applicant’s legitimate expectations.” It will be seen that the above only referred to SHD and did not refer to the decision not to apply the “mapping tool” to NHD. In the course of the hearing, there were lengthy submissions and considerable evidence on the latter decision. Towards the end of the hearing, objection was taken by the respondents to the inclusion of this decision as being part of the issue before the Court.

21. I will not accede to this objection as, as I have said, evidence was given and submissions were made, without objection, in the course of the hearing. I am satisfied that this second

decision, not to apply the “mapping tool”, was fully dealt with. However, for the reasons I will set out later in this judgment, I am of the view that the same legal principles apply to both decisions. It therefore follows that in considering the issue the Court will give its determination as to the lawfulness of the removal of SHD and the non-application of the “mapping tool” to NHD in the standardisation model that was finally used.

Evolution of the standardisation model

22. The standardisation model that was eventually used, Model 21(a), to process the school data of the applicant evolved over a period of time, undergoing a number of iterations. In the course of its evolution, two decisions were taken by the Minister. Firstly, to remove school historical data (SHD) from the model and, secondly, not to apply the “mapping tool” to the national historical data (NHD). Both of these decisions affected the calculated grades that were awarded. In this section of the judgment, I will consider how and why these decisions came about, the evidence that was given to the Court and my conclusions.

23. The prevailing circumstances that were facing the Minister cannot be discounted. In a matter of weeks, the well tried and trusted Leaving Certificate had to be replaced by an alternative system that would award grades to the Leaving Certificate class of 2020 so that they could go on to third level education or choose other careers. Each student had to have a Leaving Certificate for 2020. Simply postponing everything for one year was not an option, the class of 2020 could not be abandoned.

24. The calculated grades awarded for Leaving Certificate 2020 had to meet two objectives. Firstly, they had to be statistically accurate, and, secondly, they had to be acceptable to universities, future employers and the wider general public. These objectives were not easily met. For example, details of the gender and socio-economic background of a candidate for the Leaving Certificate might make the standardisation model more statistically accurate but would, at the same time, not make it acceptable to the public. It was in this context that the use of SHD and, to a lesser extent, not using the “mapping tool” became problematic. The Minister established a complex structure to advise on and design a system for the awarding of calculated grades. A Calculated Grades Executive Office was established to deliver the system of calculated grades. A decision making forum, the National Standardisation Group (the NSG) was established. The purpose of the NSG was to be “the decision-making group responsible for the implementation of the iterative standardisation process and the application, review, and adjustment of the data in line with the principles, parameters and constraints associated with the model to arrive at fair and just representations of student performance”. (Report from the NSG, 6 September 2020). To achieve its purpose, the NSG would consider the statistical outcomes of various iterations of the standardisation model. This required the NSG to: - i. Interrogate the data-sets emerging from the model at each iteration. ii. Compare outcomes at national level with those of recent years. iii. Consider effects and impacts at school level. iv. Ensure that the appropriate balance was struck between optimising the statistical accuracy and maintaining “face validity”. In addition, an Independent Steering Committee was established by the Minister to provide assurance to the Minister “of the quality and integrity of the outcomes of the calculated grades system …”.

25. The decision to postpone the “traditional” Leaving Certificate of 2020 was set out in a decision of the Government dated 8 May 2020. This decision stated that the Government agreed: - “(i) ---

(ii) to put in place a system to be operated by the Minister … on an administrative basis pursuant to executive powers of Government under Article 28.2 of the Constitution, whereby Leaving Certificate candidates could opt to have calculated grades issued to them by the Minister in order to facilitate their progress to third-level education or the world of work in Autumn 2020, and such system shall include the following elements:

(a) the professional judgment of each of the candidate’s teachers which shall not be subject to appeal;

(b) in-school alignment to ensure fairness amongst candidates at school level;

(c) approval by the school principal of the estimated scores and rankings of students in the school;

(d) a process of standardisation at national level to ensure fairness amongst all candidates; and

(e) --

(iii) to deliver the system through a non-statutory executive office in the Department of Education and Skills and a non-statutory steering committee, made up of relevant experts, who will oversee the quality and independence of the process under the authority of the Minister; …”

The said decision further stated: -

(vi) the calculated grades model would use estimated examination scores from students, teachers and schools and would also involve standardisation at national level to ensure equity of treatment for candidates; …”

26. On 21 May 2020 the Minister published a document entitled: -

“A Guide to Calculated Grades for Leaving Certificate Students 2020”. This document set out in general terms the basis upon which calculated grades would be awarded. In his Grounding Affidavit, the applicant made specific reference to certain sections of this document: - Para. 2, which states: - “A Calculated Grade is a grade that can be provided to students following the combination of school information about a student’s expected performance in an examination and national data available in relation to the performance of students in examinations over a period of time. …”

Para. 10.2: -

“What happens to the school data in this process? The rank order within the class group is preserved in the statistical process. However, the teachers’ estimated marks from each school will be adjusted to bring them into line with the expected distribution for the school. The national standardisation process being used will not impose any predetermined score on any individual in a class or a school. … The relevant Department data sets that support the process include mark data at:

• National level for both Leaving Certificate and Junior Certificate examinations for 2019 and previous years;

• School level for both Leaving Certificate and Junior Certificate examinations for 2019 and previous years;

• Candidate level for both Leaving Certificate and Junior Certificate examinations for 2019 and previous years;

• Candidate level for the Junior Certificate results of the 2020 Leaving Certificate cohort of candidates.”

And 10.3: -

“Processing the school level data

In advance of receiving the estimated marks from schools the information about how the school has done in the past and the information about the strengths and weakness of the current group of students will all be assembled and will be used to predict the level of achievement that this particular group of students would have been expected to reach in that subject if those students had sat the Leaving Certificate examination in the normal way. This information is then combined with the estimates that the school has provided in order to generate the fairest possible result that can be calculated. …”

27. A further document published by the Minister on 20 July 2020 once again set out the relevant information that would be used to support the process. The document further stated: - “By collecting and using a range of different types of information, the different sources of data will complement each other, to provide the most accurate and fair set of results within the limitations of the available data. As the school data is only accurate at school level, the final calculated marks, and so Calculated Grades, provided to students, for any subject and level, may be higher or lower than the estimates provided by their school. It is as a result of this standardisation process that the Calculated Grades will have an equal standing and status with previous and future Leaving Certificate grades. If this is not done, it would undermine the currency and value of Calculated Grades.

---

The maintenance of a national standard during the Calculated Grades process is as important as in previous years in order to ensure that the Leaving Certificate 2020 Calculated Grades are of equal standing to the outcomes from previous years. This is in order to ensure equity and fairness for the 2020 cohort but also for previous and future students who may be competing for college places or in the world of work. …”

28. Candidates for Leaving Certificate 2020 had an option to opt into the calculated grades system and still sit a written Leaving Certificate exam later in the year. Alternatively, candidates could opt not to receive calculated grades and later sit the Leaving Certificate. In any event, the decision had to be made before 4 p.m. on Monday 27 July 2020. This was set out in a document entitled “Getting my Calculated Grades – a Guide for Students”. In the course of the document it was stated: - “Calculated Grades have the same status as the Leaving Certificate results awarded to students in previous years. There is no downside to opting into receiving a Calculated Grade.” The applicant opted to receive calculated grades.

29. Meanwhile, various iterations of the standardisation model were being considered and examined. In the course of a lengthy affidavit, Mr. Dalton Tattan, Assistant Secretary of the Department of Education and Skills, stated: - “The NSG’s role included considering the statistical outcomes within a decisionmaking framework which took account of the commitments, principles, values and constraints which apply to calculated grades and to arrange for the implementation of adaptions in order to tune the model through various iterations. This required the Group, inter alia, to interrogate the data-sets emerging from the model at each iteration from a range of perspectives at national level, at various disaggregated levels, …”

He also stated that: -

“… Those commitments included ensuring that the results of school groups in 2020 were not unduly constrained by the historical performance of the school, ensuring that the results of individual students in 2020 were not unduly constrained by the historical performance of the school and placing a high value on the estimates from schools. It was recognised that the overriding imperatives of fairness and accuracy could require the relaxation or adjustment of some of those commitments in particular circumstances. …”

30. In August, 2020, the commitments that had been given to apply SHD came sharply into focus. This was, primarily, as a result of events that took place in the United Kingdom. The authorities in the UK were faced with a similar problem concerning the cancellation of the A Level examinations, which are, roughly, equivalent to our Leaving Certificate examinations. A calculated grades system was put in place. There was a separate system for Scotland. Like the system in this country, it was envisaged that historical school data could be used to achieve statistical accuracy in the calculated grades given. Mr. Tattan stated in his affidavit that the use of school historic data in the UK was referred to as a “post code lottery” and “school profiling”, particularly in cases where the scale of over estimation of teacher grades was greater in schools serving areas of disadvantage than in schools serving more affluent communities and smaller, private fee-charging schools. This caused a major political crisis in the UK and, indeed, threatened the future of the Government of Scotland. In response, use of a standardisation model was abandoned and candidates were awarded calculated grades based on school estimates, save where the calculated grade was greater.

31. Almost immediately, the issues that were convulsing the UK calculated grades system were widely reported by the media here. Attention was immediately focused on the use of SHD. This was taken up by politicians and commentators, criticising the use in this country as to what was being described in the UK as being a “post code lottery” and “school profiling”. The fact that there were a number of fundamental differences between the system for the award of calculated grades in this country and that of the UK (to a lesser extent in Scotland) was completely lost in the ensuing storm.

32. The political storm set in train a number of events that would culminate in a decision of the Government to remove SHD from the standardisation model and not to apply the “mapping tool” to NHD. Evidence of these events was given to the Court by Mr. Tattan, who was cross-examined, at length, on his affidavit. Mr. Tattan stated that the Minister herself had, since taking office at the end of June, 2020, considerable doubts and misgivings on the use of SHD. The Minister had to consider the effects of removing SHD. The removal of SHD would result in taking out of the standardisation model the measure of “school effectiveness”. This means, in simple terms, that schools who have a history of obtaining high grades in the Leaving Certificate are more likely to be accurate in their estimated marks than are schools with a less good history.

33. To illustrate the statistical effects of removing SHD, two models were looked at: Models 10 and 17. For the purposes of these models only two schools were selected. The first school, Mount Anville, had a record of strong results in the Leaving Certificate; and the other school, a DEIS school, did not have such a record. Only one subject was considered, namely: Mathematics, both at ordinary and higher level. In Model 10 SHD was removed, whilst it was retained in Model 17. It should also be noted that Model 17 did not preserve the class rankings as submitted by the schools. This exercise appeared to show that the omission of SHD in Model

10 led “to a substantial decrease in the mean mark (in higher/ordinary level mathematics) and the grade profile in comparison to what the school had seen in the past”. On the other hand, in respect of the DEIS school, the removal of SHD had led to a substantial increase in the mean mark and the grade profile in comparison to what the school had seen in the past. This would, on its face, appear to show that the removal of SHD would adversely affect a traditionally highperforming school, such as Mount Anville. However, as against this it must be noted that these models only looked at two schools and one subject, they were experimental in nature and, in any event, would not have been the final model that would be used for the award of calculated grades.

34. The Minister was informed of the exercise concerning Models 10 and 17.

35. Confusion arose in the course of Mr. Tattan giving his evidence as to inclusion in the standardised model of national historical data (NHD) and the application of a “mapping tool”

to such data. It will be recalled that the purpose of including such data was to bring about an alignment between the results of 2020 and those of previous years. Whereas though NHD was included, its effects on the eventual calculated grades awarded were reduced by the non-use of the “mapping tool”. In the course of his evidence, Mr. Tattan incorrectly indicated that NHD was not used. However, the decision that was, in fact, taken was not to apply the “mapping tool”. Had the mapping tool been used, it would have meant that some 58% of higher level grades would have been reduced by one grade. It was the view of the Minister that such a downward adjustment in the schools’ estimates “could have been fatal to the acceptability of the calculated grades system”. (Per para. 62 of the affidavit of Mr. Tattan). This resulted in “grade inflation” in the order of 4.5% to 5%. Mr. Tattan expressed the view that there was, in effect, a trade-off between grade inflation and restoring public confidence in the calculated grade system.

36. I have already referred to the fact that the standardisation model underwent numerous iterations in the period that led up to the events and the consequent decisions of August and September, 2020. By the middle to the end of August, the Minister was looking at two models, namely: 18(a) and 18(g). In Model 18(a) SHD had been “dialled down to the greatest possible degree”, whereas in Model 18(g) SHD had been removed completely. Further, though in both of these models there was an element of NHD, in neither was the “mapping tool” applied. Before proceeding, two matters should be noted. Firstly, there was a Model 18 which evolved into 18(a) and (g); and, secondly, the model that was ultimately used was Model 21(a), though the differences in the final model and Model 18(g) do not appear to be relevant. As to the circumstances under which Models 18(a) and (g) came about, the Court heard evidence from Mr. Hugh McManus who was cross-examined, at considerable length, on his affidavits.

37. Mr. McManus was Assistant Director of the Calculated Grades Executive Office at the Department of Education and Skills. Mr. McManus was also a member of the NSG. He gave detailed evidence of the various iterations of the standardisation model that led to Model 18 and subsequent models. In particular, he described the effects of using SHD in earlier models. He stated that, paradoxically, over emphasising the role of SHD had actually caused the schools to come closer together, which was not what one might expect. He stated that if you tried to force SHD too strongly into the model that you actually prevent either individuals, or groups of individuals, from deviating from the history of the school to the extent that they ought to if the model was functioning properly. These observations on the effects of the use of SHD had led to Model 18(a), which had reduced SHD to the lowest degree possible. Mr. McManus stated that this had occurred without the intervention of the Minister and that if the Minister had not intervened the SHD that would have been used in the final model was as it had been used in Model 18(a). Mr. McManus was aware of the Minister’s view of SHD. This evidence was the source of considerable controversy, which I will return to shortly.

38. Mr. McManus accepted that removing SHD from the model was removing a significant source of information, being the element of school effectiveness. He gave evidence concerning four tables that were set out in a document entitled “Information Note – Calculated Grades” of 20 August 2020 from the Department of Education and Skills. It would appear that this note informed the decision that was taken to remove SHD and not to use the “mapping tool”. In this document there are four tables which set out the effects of either using Model 18(a) with SHD and Model 18(g) without SHD. In the course of his cross-examination on these tables, Mr. McManus accepted that they give a generalised view of how many grades will be increased or decreased and that they do not give information as to which cohort of students will gain or lose, though he did state that with every change to the standardised model some students would gain and others would lose. Mr. McManus accepted that, from a purely statistical perspective, the results that would have emerged had SHD been left in were likely to be more statistically accurate than the ones that emerged from the standardisation model without SHD.

39. Whilst giving his evidence, Counsel for the applicant challenged Mr. McManus on his evidence that the SHD that was used in Model 18(a) was as a result of the evolution of the standardisation model, rather than by intervention of the Minister. In a later submission, Counsel referred to documents from Ms. Andrea Feeney, Director of the Calculated Grades Executive Office, which, on their face, indicated that the change in the use of SHD was prompted by intervention of the Minister following on from the public controversy, which I have referred to. These documents were not put to Mr. McManus in the course of his crossexamination. This prompted a letter from the Chief State Solicitor’s Office, on behalf of the respondents, suggesting that Mr. McManus be recalled to give evidence on these documents and that Ms. Andrea Feeney also be called to give evidence on this. It should be noted that in the course of a preliminary hearing this Court refused an application by the applicant to crossexamine Ms. Feeney on her affidavits. Counsel for the applicant submitted that the content of the evidence given by Mr. McManus should have been referred to in both the Statement of Opposition and replying affidavits. Arising from this it was maintained that there was a lack of candour on the part of the respondents, and that they had failed to meet the applicant’s case “with all cards face up”, as they ought to have done. The applicant invited the Court to reject the evidence of Mr. McManus on this point. As for the fact that the documents from Ms. Feeney were not put to Mr. McManus, in the course of cross-examination Counsel relied on the rule in Browne v. Dunn, as considered by the Supreme Court in McDonagh v. Sunday Newspapers Ltd [2017] IESC 46.

40. I will not be acceding to this application. I accept the evidence of Mr. McManus for a number of reasons. Firstly, Mr. McManus was cross-examined at length and in detail, commencing on the morning of Wednesday 13 January and concluding on the afternoon of Friday 15 January, in excess of two and a half days. In the course of that time, I had an opportunity to assess Mr. McManus as a witness and I am satisfied that, at all stages, he gave his evidence truthfully and conscientiously. As for the suggestion that this was “new” evidence and ought to have been referred to in both the Statement of Opposition and the various replying affidavits, I do not accept this. All of the documentation in this case makes clear that the standardisation model was undergoing numerous iterations before a final model was arrived at. Various iterations can only mean changes in the model from one iteration to another. Mr. McManus was a member of the NSG and it is clear that changes to the manner in which SHD was being used were within its terms of reference (see para. 24 above). Arising from this, I do not accept that there has been any lack of candour on the part of the respondents. I do not attach very much significance to the fact that Ms. Feeney’s documentation was not put to Mr. McManus in the course of his cross-examination. However, I would have thought that if this evidence from Mr. McManus was as significant as the applicant maintains it was that the offer to recall Mr. McManus and, indeed, the offer to make Ms. Andrea Feeney available to give evidence would both have been accepted. Further, insofar as the evidence of Mr. McManus may have had implications on the statistical aspect of the case, the applicant declined an opportunity to recall their own statistical expert, Professor Cathal Walsh, to deal with this.

41. Having considered the evidence of Mr. McManus, I find, as a fact, that had there been no intervention by the Minister, the SHD that would have been used in the final standardisation model would have been as it was used in Model 18(a). In this context, I also note that the only model in respect of which the applicant submits SHD was used appropriately was Model 17. It was common case that Model 17 was a developmental model and would not have been used as a final model.

42. I will now consider the documentation that was before the Court on which the decision of the Minister was based. The respondent claimed executive privilege over the memorandum to Government. I refer to a document, namely: “Information Note – Calculated Grades” of 20 August 2020. This document refers to the memorandum to Government of 8 May 2020 which sets out the data to be used in the standardisation process so as to ensure the equitable treatment of candidates for Leaving Cert 2020. This data was: -

(1) The estimated marks and ranking of students supplied by schools;

(2) The historical national distribution of student results on a subject-by-subject basis based on historical Leaving Certificate examination performance;

(3) School historical data based on historical Leaving Certificate examination performance at the school level across three prior years; and

(4) A prediction of the likely Leaving Certificate performance of the class of 2020 in each school based on their collective performance when they undertook the junior cycle examinations. The document then sets out why standardisation was required, stating that different schools would take different approaches for generating estimated marks and rank orders. Some schools would be overly optimistic, others very harsh. The document further stated that the element of the standardisation process that had proved to be least acceptable in public discourse had been the use of SHD and reference was made to the events that had unfolded in the UK. However, the safeguards, not present in the UK system, were set out.

43. The document stated that two versions of the standardisation model were now available, namely: Model 18(a) and Model 18(g). It was stated that one of the models, Model 18(g), involved a change in the data sources used and that the Government would need to be made aware of this change.

44. The document considered the estimated marks that had been submitted by the various schools, stating that when aggregated across all subjects the percentage of grade ones at higher level in 2019 was 5.8%, while in the 2020 school based estimates it was 13.2%. Thus, the percentage of grade ones had more than doubled in teachers’ estimates in many subjects. The document noted that “such a change in standards within one calendar year is simply not credible”.

45. The document, in its four tables, sets out the effects of retaining SHD, albeit at a low level, in Model 18(a), comparing the outcome to Model 18(g) where SHD had been removed. This exercise was carried out for DEIS schools and non-DEIS schools. In summary, it was found that Model 18(g) appeared to show that over 75% of all teachers’ estimated grades would remain unchanged, about 5% would be raised and 18.6% of the grades would be lowered. Marginally fewer grades would be lowered in DEIS schools than in non-DEIS schools. The recommendation was made to use Model 18(g).

46. In the course of the hearing, the applicant emphasised that the NSG do not appear to have been involved in the decision to remove SHD and not to apply the “mapping tool”. These matters were also the source of considerable debate amongst officials in the Department, with various emails and commentaries therein being produced to the Court. However, the role of the NSG was to develop a standardisation model based on the four data sets that had been identified in various documentation that emanated from the Department. The NSG did not have authority to remove a data set from the standardisation model as this was a matter for the Minister. In any event, the NSG in its comprehensive report of 6 September 2020 supported the decisions taken, as did the Independent Steering Committee.

47. Prior to being considered by the cabinet, the decisions on the final standardisation model were discussed by the Minister with the Taoiseach, and, separately, leading members of the parties that make up the present coalition government were briefed. The matter came before the Government on 1 September 2020. The written decision of the Government recorded the following had been agreed: - “(i) that historical school distribution data, based on historical Leaving Certificate Examination performance of past cohorts of students at the school level across 2017, 2018 and 2019, will be removed from the range of data being used in the standardisation model in response to concerns that have been raised; and

(ii) that the reliance in the standardisation model on historical national distribution of students’ results on a subject-by-subject basis, and therefore the impact of this data, will be minimal.”

48. Having considered the affidavits filed, the exhibits and the evidence given to the Court by Mr. Tattan and Mr. McManus, I am satisfied that the data that was used in the standardisation model for the awarding of calculated grades did not meet the commitments that had been given by the Minister. Though there was an element of SHD from the basis on which teachers’ awarded estimated marks and the use of prior Junior Certificate data, it clearly was not used in the form that had been communicated to the applicant in the documentation I have already referred to.

49. There was also a commitment by the Minister that the results of the Leaving Certificate of 2020 would be in line with the results of previous years. This commitment was also not honoured in that by deciding not to apply the “mapping tool” to NHD it, inevitably, resulted in “grade inflation” . As far as the applicant is concerned, this “grade inflation” has to be seen alongside the effects on him of the removal of SHD.

50. In the course of the hearing, the respondents supplied to the Court an aide-mémoire illustrating how the standardisation model worked. There was no disagreement as to its contents. I have therefore included it in Appendix A to this judgment. For clarity: -

- “A” represents how NHD and prior Junior Certificate results were used;

- “B” represents SHD, which was not used; and

- “C” represents the “mapping tool”, which was also not used.

51. The failure of the Minister to honour these commitments has to be seen in the context that the results of Leaving Certificate 2020 had to be acceptable to the general public. I will now consider the legal principles that apply. I will also consider whether, in fact, the applicant suffered any unfairness.

Justiciability

52. The first legal issue which I will consider is whether the decisions, be they of the Minister or the Government, are justiciable.

53. Article 28.2 of the Constitution provides: -

“The executive power of the State shall, subject to the provisions of this Constitution, be exercised by or on the authority of the Government.”

Submissions

54. Mr. Feichín McDonagh SC (with Mr. Micheál P. O’Higgins SC and Mr. Brendan Hennessy BL), on behalf of the applicant, submitted that the system providing for the award of calculated grades does not have a legislative basis, nor, indeed, does the Leaving Certificate examination itself. Rather, the system was clearly administrative and voluntary. The system was the result of negotiations entered into by the various interested parties. Therefore, the decisions of August/September 2020 could not, and did not, transform the voluntary system into a compulsory one. It followed, according to the submission, that the executive power of the State was not invoked, save in respect of funding.

55. As for the public controversy that arose following the events in the UK arising from its system of calculated grades, Counsel submitted that the controversy was based on a complete misunderstanding of the differences between the Irish system and that operating in the UK. In the applicant’s view, the pressure that built up on the Minister, which resulted in the impugned decisions, was a result of ill-informed commentators and politicians. Thus, there was no reasonable or rational basis for the decisions.

56. As part of their submission, Counsel relied on a number of authorities. In State (C) v.

Minister for Justice [1967] I.R. 106, Walsh J. stated: -

“With regard to the last point, it is my opinion that the fact that a statutory power is conferred upon a member of the executive or a representative of the executive, as was the Lord Lieutenant, does not make that power an executive power within the meaning of that expression in the Constitution as the statute might just as easily have conferred the power on anybody else. …” Reliance was also placed on Gutrani v. Minister for Justice [1993] 2 I.R. 427. This case concerned a letter that had been written on behalf of the respondent concerning procedures that would be implemented for applications for refugee status and asylum, as had been suggested by the representative of the United Nations High Commissioner for Refugees. The applicant maintained that humanitarian considerations had not been taken into account in accordance with the terms of the said letter. In the course of giving the judgment of the Supreme Court, McCarthy J. stated: - “The Minister does not contest that he is obliged to consider the application within the framework of the letter of the 13th December, 1985. Having established such a scheme, however informally so, he would appear to be bound to apply it to appropriate cases, and his decision would be subject to judicial review. It does not appear to me to depend upon any principle of legitimate or reasonable expectation; it is, simply, the procedure which the Minister has undertaken to enforce.” The applicant also relied on a number of English authorities, in particular: The Secretary of State for Foreign and Commonwealth Affairs v. Quark Fishing Ltd [2002] EWCA Civ. 1409. This case concerned a judicial review arising out of a refusal to grant a licence to catch what was a very commercially valuable fish, the Patagonian toothfish.

57. Ms. Eileen Barrington SC (with Mr. Brian Kennedy SC, Mr. Francis Kieran BL and Mr. Joseph O’Sullivan BL) submitted that in making the impugned decisions the executive power of the State was engaged. “Executive power” is not defined, but reliance was placed on the following passage from the judgment of O’Donnell J. in Barlow v. Minister for Agriculture [2017] 2 I.R. 440 where he stated: - “… It appears that the executive power in Irish law to date is, as Professor Casey observed, the residue which is left when the judicial and legislative powers are subtracted: Casey, Constitutional Law in Ireland, 3rd Ed., (Dublin, 2000), pp. 230-231. …”

58. The respondents submit that the calculated grade system was put in place by executive

decision to be operated by the Minister. This was deposed to in the affidavit of Mr. Tattan. Mr. Tattan further deposed that the Government decision of 8 May 2020 was based on a memorandum to the Government which included reference to using SHD as part of the standardisation process. He maintained that in having to reconsider the matter a further decision of the Government would be required.

59. More particularly, the respondents submitted that the decisions to remove SHD and not to apply the “mapping tool” to NHD involved matters of policy which, given the separation of powers, fell outside the jurisdiction and competence of the courts to review. A number of authorities were relied upon. As a general statement of the law, reliance was placed on the

decision of the Supreme Court in T.D. v. Minister for Education [2001] 4 I.R. 259. This was an appeal to the Supreme Court from a decision of the High Court which had granted a mandatory injunction directing the respondent to implement forthwith a policy that had been formulated regarding the accommodation and treatment of children with special needs, which category included the applicant. In his judgment, Murray J. (as he then was) stated: - “Adopting a policy or a programme and deciding to implement it is a core function of the Executive. It is not for the courts to decide policy or to implement it. It may determine whether such policy or actions to implement such policy are compatible with the law or the Constitution to fulfil obligations. That is not deciding policy.

Judicial review in a democracy

Thus the powers of the court include judicial review of acts of the executive and the legislature. It is a feature common to many democracies, particularly with a written constitution. Judicial review permits the court to set aside executive actions or legislative measures which offend against the law or the Constitution. Judicial review does not in such democracies give the courts jurisdiction to exercise rather than review executive or legislative functions. Judicial review permits the courts to place limits on the exercise of executive or legislative power, not to exercise it themselves. It deals with the limits of policy, not its substance. That is why judicial review by the courts, which are not answerable to any constituency other than the law and the Constitution, is democratic. …” In considering the granting of mandatory orders, Murray J. stated: - “In so far as McKenna v. An Taoiseach (No. 2) [1995] 2 I.R. 10, Crotty v. An Taoiseach [1987] I.R. 713 and District Judge McMenamin v. Ireland [1996] 3 I.R. 100 might be said to be authority for the making of some form of mandatory order where there is ‘a clear disregard’ by the State of its constitutional obligations, it must be borne in mind that in none of those cases was a mandatory order granted. I have already made the distinction between ‘interfering’ in the actions of other organs of State in order to ensure compliance with the Constitution and taking over their core functions so that they are exercised by the courts. … In my view the phrase ‘clear disregard’ can only be understood to mean a conscious and deliberate decision by the organ of state to act in breach of its constitutional obligation to other parties, accompanied by bad faith or recklessness. A court would also have to be satisfied that the absence of good faith or the reckless disregard of rights would impinge on the observance by the State party concerned of any declaratory order made by the court. …”

In the instant case, were the issue before the Court to be decided in the applicant’s favour, a declaratory order would follow.

60. The respondents also submitted that not only are the impugned decisions a matter of policy, but also that there were no legal standards of competency on the part of the Court which could guide it were it to review the said decisions. The respondents relied on the decision of the Court of Appeal in Moore v. Minister for Arts, Heritage and the Gaeltacht [2018] 3 I.R.

265. This case concerned a judicial review of a decision by the respondent arising out of the refusal to declare Moore St. and its environs a “national monument” for the purposes of the relevant statute. The High Court had granted the relief sought, including a declaration that certain sites constituted national monuments for the purpose of the statute. In allowing the appeal, Hogan J. stated: - “(41) The reason why, however, the matter would be regarded as executive (or, possibly in some instances, legislative) in character is, however, readily apparent, because the designation of a monument as a national monument ultimately calls for political and, in some instances, perhaps, administrative judgment …”

and: -

“(42) The judicial branch quite obviously lacks the institutional competence, capacity and, most of all, democratic legitimacy to determine policy matters of this kind. Article

34.1 of the Constitution instead requires the judiciary to administer justice, thus typically requiring the judges to apply conventional legal materials - such as the corpus of common law rules and principles, rules of statutory interpretation, precedent and reasoning by analogy – in a detached and principled fashion, regardless of the consequences. …”

On the issue of separation of powers, Hogan J. stated: -

“(44) The other branches of government bring with them the strength of democratic accountability and the constitutionally assigned role of policy making. The Government brings with it the policy insights of its members and the wider civil service. It can give a lead as to what is likely to be effective in practical policy terms and it is likewise dispensed from the necessity to rationalise its actions by reference to conventional legal principles.

(45) The reason for this different approach is, of course, because, generally speaking, the two other branches of government are engaged in the business of policy formulation as distinct from the administration of justice. In contrast to judicial decision making, the policy makers of the legislative and executive branches are not required to be consistent or to have regard to established precedent or to proceed from legal principle or to give detailed reasons in writing for their decisions. Nor are they required to be detached and impartial in the same manner as is required of the judiciary by Article 34.6.1° of the Constitution. Critically, however, the two other branches of government are democratically accountable in a way that the judiciary are not.”

61. Similar views were expressed by Hogan J. in Garda Representative Association v. Minister for Public Expenditure and Reform [2016] IECA 18, which case concerned an application for judicial review of a decision by the respondent to include An Garda Síochána in “Sick Leave” Regulations. This inclusion was made notwithstanding a commitment that the Gardaí would be excluded. It was not in dispute that the respondent Minister suddenly changed his mind and decided that the Regulations would apply to the Gardaí, with immediate effect. It subsequently emerged that the reason for the volte-face was that a senior trade union official stated that the public sector trade unions would not accept any exemption or derogation for the Gardaí. In upholding the decision of the High Court to refuse the applicants the reliefs they sought by way of judicial review, Hogan J. stated: - “(42) In the present case the Minister was engaged in the practical politics of policy formation by piloting the 2013 Act through the Oireachtas and by subsequently promulgating [the 2014 Regulations] … It is true that the GRA might legitimately consider that they had been let down by the manner in which that decision was arrived. The GRA are also entitled to feel disappointed given that the critical intervention of Mr. Cody was not disclosed to them at the time and this only came to light subsequently in the course of the discovery process after these proceedings had been commenced. Yet, from the Minister's perspective, the greater prize of securing the reform of a very expensive feature of public service pay and conditions while avoiding the threat of industrial action from other public sector unions made it imperative in the circumstances that a snap decision of this kind (i.e., to include An Garda Siochána in the new regime) be taken immediately, regardless of any assurances in relation to consultation which the Minister might previously have given to the GRA.

(43) This conclusion finds expression in the case-law which has consistently rejected the suggestion that legislative, or quasi-legislative decisions, attracts the principles of fair procedures, even though such generally applicable rules might have significant implications for the livelihood, well-being and general welfare of those affected by such decisions. …”

Evidence

62. Dr. William Maxwell filed an affidavit exhibiting an expert report on behalf of the respondents. Dr. Maxwell is an Educational Consultant and was Her Majesty’s Chief Inspector of Education for Scotland from 2010 to 2017. He led the creation and development of Education Scotland, a new quality improvement agency established in 2011 to integrate both inspection and improvement support functions. As HM Chief Inspector, he was Chief Professional Advisor to Scottish Ministers on matters relating to education. From 2008 till 2010, he was Her Majesty’s Chief Inspector for Education for Wales. Dr. Maxwell was in a good position to give evidence on the system that was adopted in Scotland and the remainder of the UK for the awarding of calculated grades due to the cancellation of state examinations, the circumstances and nature of the political storm that blew up and the steps taken by politicians to address the situation.

63. In his report Dr. Maxwell stated: -

“In conclusion, on the basis of the evidence I have seen in relation to the development of the Irish ‘calculated grades’ approach through to its conclusion, it is my view that the Minister acted rationally and reasonably in ordering the changes which were set out in the memorandum of the 24th August. These were a reasonable response to ensuring that the overarching policy objective of maintaining confidence in the fairness and integrity of the awarding system was secured, in the context of heightened risk arising from a series of high-profile problems emerging around equivalent approaches in the UK and the resulting focus of public disclosure in Ireland throughout August.

Furthermore, having considered evidence of the outcomes of the standardisation process in general, and its impact on Belvedere College in particular, I see no evidence that the changes which the Minister ordered, including the removal of school-by-school performance data, resulted in any unfair disadvantage to students at high-performing schools, and consequently to him personally, in the way that the Applicant contends. …”

64. Dr. Maxwell was cross-examined on his report. It was put to Dr. Maxwell that the standardisation model was likely to be more accurate if SHD was included. In response, Dr. Maxwell stated that, from a statistically purist perspective, it would make the model more accurate, as indeed would including “—the social postcode of the child and the family –”. This would not be considered to be ethical or credible in the public domain, even though it might enhance the statistical technical accuracy of the model. He gave details of the controversy in calculated grades that arose in the UK. Scotland was first to give its results first and the initial public reaction was not concerning the treatment of disadvantaged schools, but, rather, “the sheer scale and number of students who were affected by downgrading effectively …”. A narrative developed that this downgrading had happened more to the pupils in certain disadvantaged schools, and hence the use of a term such as “postcode lottery”.

65. Dr. Maxwell gave evidence that a feature of the criticism of the calculated grade system in the UK was that by including SHD in the standardisation model, not only did it favour wellresourced schools with a history of good exam results, but it also “pegged back” less wellresourced schools whose examination results were improving year on year. He expressed the view that when perception of the potential for unfairness takes hold, then the credibility of the system suffers.

66. Dr. Maxwell outlined what would have been the appropriate response to the public controversy on calculated grades. He stated in evidence: - “I think the role of public administration is to adjust and adapt the way systems are operated to guard against criticisms that are, whether justified or unjustified, as far as is reasonably possible, and it is, therefore if some adjustment can be made to a system to protect it against allegations or the perception, right or wrong, without doing detriment, significant detriment to the system, when it should be - - it makes sense to adjust.” In the UK the response was, in effect, to abandon the standardisation process.

Consideration of submissions

67. The first matter which I wish to consider is whether the impugned decisions were an exercise of executive power. Applying the passage I cited from O’Donnell J. in Barlow v.

Minister for Agriculture, it would seem to me that this was an exercise of executive power. The decisions taken were clearly not an exercise of judicial or legislative powers. The fact that the calculated grade system was voluntary does not alter this. (See Ryanair DAC v. An Taoiseach [2020] IEHC 461).

68. In these proceedings, the applicant has not invoked any constitutional or legal right (save for legitimate expectation, which I will consider later) that has been infringed. Nor has the applicant identified any “clear disregard” of the respondents of their powers and duties conferred on them under the Constitution, as per Murray J. in T.D. v. Minister for Education (see para. 58 above). In my view, the decision of Walsh J. in State (C) v. Minister for Justice is not of assistance. The passage of the judgment cited earlier (para. 56) is clearly referable to “statutory power”, which is not the case here. This also applies to the English Court of Appeal

decision in The Secretary of State for Foreign and Commonwealth Affairs v. Quark Fishing Ltd. What was involved there was also statutory provision. Further, the issue in Gutrani v. Minister for Justice was a scheme set out in a letter from the respondent to the United Nations High Commissioner for Refugees. This arose by reason of Ireland being a signatory to the United Nations Convention on the Status of Refugees and Stateless Persons, 1951 and the United Nations Protocol on the Status of Refugees and Stateless Persons, 1967, though these were not part of domestic law of the State, such a letter is very different from the proposed system of calculated grades in these proceedings. Further, in Gutrani the respondent did not contest that he was obliged to follow the terms of the letter. That is clearly not the case here.

69. Whatever issues there may be as to whether or not executive power was engaged in the making of the impugned decisions, to my mind, these were policy decisions which, on the authorities of T.D. v. Minister for Education; Garda Representative Association v. Minister for Public Expenditure and Reform; and Moore v. Minister for Arts, Heritage and the Gaeltacht are not reviewable by the Court. To explain how I reach this conclusion, it is necessary to retrace some steps.

70. The purpose and aim of the calculated grades system was to award each candidate a calculated grade that would represent the grade that he or she would have obtained had the particular exam been sat in the normal way. There were always two fundamental aspects to the system. Firstly, that the system had to be statistically accurate, and, secondly, that there had to be public acceptance of the system. The importance of public acceptance cannot be minimised. Calculated grades awarded for Leaving Certificate 2020 would have to be accepted by those responsible for admissions to third level education and present and future employers of the class of 2020. However, what was not in dispute was that the use of certain data, e.g.: gender and economic background, which might make the system more statistically accurate would not be acceptable to the public. Statistical accuracy had to be sacrificed to maintain public acceptance. This was the case even before the controversy arose in the UK.

71. The effect of the controversy in the UK was that the inclusion of SHD and the downgrading, on a large scale, of teachers’ estimated grades, though statistically justifiable, was not acceptable to the public. This left the Minister with three options, as per the evidence of Dr. Maxwell: -

(i) To carry on regardless, stick to the original plan, highlight the differences between the Irish model and that in the UK, and make all the arguments to persuade the public as to the fairness of the standardised model as it stood;

(ii) To make a number of adjustments to the standardised model so as to reduce the likelihood of a narrative gaining ground which could undermine the whole system;

(iii) To abandon the standardised model, as was done in all parts of the UK, the Netherlands and France.

What is common to these options is an appreciation and understanding of what is necessary to gain public acceptance and to restore confidence in the system for the awarding of calculated grades. Being democratically elected politicians, the Minister and her Government colleagues are best placed to decide what is, or is not, acceptable to the public. This puts the matter firmly in the area of policy, which is a matter for elected representatives and not the courts. I refer to the authorities, which the respondents relied upon, set out at paras. 56-60 above, arising from the separation of powers, the courts have neither the competence nor legal authority to choose one of the above options over another.

72. Those, including the applicant, who feel aggrieved by the decisions taken by the respondents are not without redress. I refer to the following passage from Hogan J. in Garda Representative Association v. The Minister for Public Expenditure and Reform: - “41. This democratic accountability has the important consequence that the electorate expect their politicians to achieve practical results. Politicians who are perceived by the electorate as having failed to deliver such results will potentially suffer the electoral consequences. For these reasons, these politicians must have regard to the practical consequences of their decisions – and the wishes of the electorate – in a manner which would not be appropriate to judicial decision-making.”

73. In view of the foregoing, I am satisfied that the impugned decisions, notwithstanding that they had the effect of reducing the statistical accuracy of the calculated grades that were awarded and caused “grade inflation”, were policy decisions taken by the respondents to ensure public acceptance of the calculated grades system and are not justiciable by the Court.

Legitimate expectation

74. It is clear from the terms of the issue before the Court that the applicant relies upon “legitimate expectation”.

75. In Glencar Exploration Plc v. Mayo County Council (No. 2) [2002] 1 I.R. 84, Fennelly J., having reviewed a number of authorities on legitimate expectation, drew the following conclusions, which he described as provisional: - “… Firstly, the public authority must have made a statement or adopted a position amounting to a promise or representation, express or implied as to how it will act in respect of an identifiable area of its activity. I will call this the representation. Secondly, the representation must be addressed or conveyed either directly or indirectly to an identifiable person or group of persons, affected actually or potentially, in such a way that it forms part of a transaction definitively entered into or a relationship between that person or group and the public authority or that the person or group has acted on the faith of the representation. Thirdly, it must be such as to create an expectation reasonably entertained by the person or group that the public authority will abide by the representation to the extent that it would be unjust to permit the public authority to resile from it. Refinements or extensions of these propositions are obviously possible. Equally they are qualified by considerations of the public interest including the principle that freedom to exercise properly a statutory power is to be respected. However, the propositions I have endeavoured to formulate seem to me to be preconditions for the right to invoke the doctrine.” I will now apply these conclusions to the case the applicant is making.

76. Earlier in the judgment I quoted extensively from various documentation that emanated from the Minister as to the data that would be used in the standardisation model for the awarding of calculated grades. It was clearly the case that SHD would be used to achieve statistical accuracy. It was also the case that the results of Leaving Certificate 2020 would be aligned with those of previous years. This would necessitate the application of the “mapping tool” to NHD. This is clearly a “public authority” making a statement “amounting to a promise or representation, express or implied as to how it will act in respect of an identifiable area of its activity”.

77. These representations were made directly “to an identifiable person or group of persons”, this being the applicant and the Leaving Certificate class of 2020.

78. The third condition is somewhat more problematic. The respondents submitted that the applicant failed to satisfy this condition in that in the document inviting the applicant to opt into the calculated grades system stated: - “… there is no downside to opting into receiving a calculated grade.” This statement was made in circumstances where the applicant still had the option to sit all or any subjects in a “traditional” Leaving Certificate exam that was to be held in November that year. In any event, the academic year in third level institutions would begin before that date and, so, any points requirement had to be fulfilled on the basis of calculated grades so there was no other option. In his affidavit, the applicant stated: - “… As I thought the process was going to be fair, I didn’t keep on studying for the possible sitting of an exam in November.” Thus, the applicant maintains he acted to his detriment. In any event, there is authority for the proposition that “detrimental reliance” is not a condition precedent to a successful claim of legitimate expectation. I refer to the following passage from Administrative Law in Ireland by Gerard Hogan, David Gwynn Morgan and Paul Daly (5th ed.) p. 1261 where, the authors having quoted Fennelly J. in Daly v. Minister for the Marine [2001] IESC 77, state: - “21-68 To put this important point another way: unfairness in the context of legitimate expectations is not to be equated with detrimental reliance in the context of estoppel. Whilst they draw inspiration from similar springs, they are distinct concepts which play different roles in public and private law respectively. Where an applicant has detrimentally relied on an assurance, policy or practice, his or her case may be stronger (and/or may be needed to justify a public body resiling from its previous position), but the absence of detrimental reliance will not be fatal to a legitimate expectations claim. However, what has to be established is that it would be ‘unjust to permit the public authority to resile…’ ” It follows from this that the applicant must establish some unfairness that he has suffered as a result of the impugned decisions.

Fairness

79. In the following paragraphs, I will consider the effects on both the applicant and his school, Belvedere College, of the removal of SHD and the non-application of the “mapping tool” to NHD. I will also consider the evidence given by the statistical experts on behalf of both the applicant and the respondents.

80. Given the number of candidates, possibly in excess of 62,000, being awarded calculated grades, the unfairness would have to be of an order to render the system unlawful. It was clearly the case that, from the start of the process, the standardisation model was going to undergo a number of iterations. Each iteration was going to produce different results. Thus, it was inevitable that any change to the model would result in some candidates gaining and others losing. Those who lost would, undoubtedly, consider it to be unfair. However, in the absence of evidence that the unfairness suffered by an individual was widespread across the cohort of those receiving calculated grades and of a serious nature, it is difficult to see how such unfairness could amount to being unlawful.

81. In his Grounding Affidavit, the applicant maintains that, due to the absence of SHD in the final standardisation model, he was unfairly downgraded from the estimated marks which were submitted to the Department by his school. The following was his situation.

Estimated grades from school: -

(i) English H3

(ii) Irish H2

(iii) Maths H2

(iv) Chemistry H2

(v) Biology H2

(vi) French H3

(vii) Latin H2

Calculated grades received: -

(i) English H3

(ii) Irish H3

(iii) Maths H3

(iv) Chemistry H3

(v) Biology H3 (Subsequently, upgraded to H2 due to a coding error).

(vi) French H3

(vii) Latin H3

The applicant considered that the estimated marks submitted by his teachers and school were a fair and accurate prediction of the grades he would have achieved had he sat the Leaving Certificate 2020 in the usual way. The applicant also criticises the grade inflation, which was of the order of 4.5% to 5%, which put the points requirements for his third level courses of choice out of his reach. The applicant was supported in an affidavit of Mr. Tom Doyle, Deputy Principal of Belvedere College. He stated: - “9. I say that Belvedere College students have a proven history of high performance across Leaving Certificate subjects and that inevitably appropriate clusters of students with high grades have occurred in the applicant’s classes. The College could easily have explained these clusters of students with high grades had we been given an opportunity to do so relying on our historical data.”

82. In a replying affidavit, Mr. Hugh McManus exhibits a series of tables showing historic Leaving Certificate results (2017-2019) and the 2020 estimated grades from Belvedere College in respect of each of the subjects taken by the applicant. I will set out these figures as they apply to H1s and H2s, and the percentage of Belvedere students awarded these grades: - Subject Average: 2017-2019 2020 Estimate 2020 Calculated Grade Irish H1 8.70 20.70 12.20 H2 26.20 31.70 26.80 English H1 5.80 13.70 7.50 H2 13.60 19.90 16.40 Mathematics H1 6.90 18.90 13.50 H2 18.60 27.00 22.50 Latin H1 11.10 37.50 37.50 H2 22.20 50.00 37.50 French H1 12.10 15.70 12.00 H2 22.50 25.00 20.40 Biology H1 12.80 25.80 18.20 H2 21.70 30.30 24.20 Chemistry H1 19.20 43.60 25.60 H2 18.40 33.30 25.60

83. It is readily apparent from the above that the 2020 estimates from the school for H1s/H2s were well ahead of what had been achieved in the previous three years, in some cases a multiple of such. Lest it be suggested that the class of 2020 had more ability and application than the previous years, this is not borne out by the results which the same cohort of students received in their Junior Certificate examinations. Overestimation by teachers in schools was anticipated, but probably not on this scale. As mentioned earlier, amongst the evidence to be considered in awarding an estimated mark by a teacher was “previous results in the school in this subject”. It is difficult to see that this was done. Belvedere College was certainly not alone in providing overestimates, as such was prevalent in other schools, particularly at the higher levels.

84. I should say that I do not intend this to be critical of the teachers or the school involved. The awarding of estimated marks by teachers to their pupils, whom they would have known well over many years, for an examination as important as the Leaving Certificate put both the school and the teachers in a difficult and invidious position. This underlines what is one of the most important features of the traditional Leaving Certificate, namely: its anonymity.

85. Given the scale of the teacher and school overestimates, it is not at all surprising that the applicant was going to be downgraded. As Professor Van der Linden, Statistical Expert on behalf of the respondents, stated in his report: - “Unless the Applicant’s school is able to explain the sudden increase in the percentage of its estimated H1 and H2 grades for 2020, it is quite unlikely that the Applicant was disadvantaged by the downward adjustment of his school grades. Also, as already noted, the adjustments applied to them by the standardisation model appear to be relatively mild.”

86. The basis for using SHD was that if the estimated grades came from a school that had a record of achieving high grades, it was more likely that the estimates would be accurate and not downgraded. In the course of the hearing, SHD was referred to in the following terms: - “(If) School historical data (is) still working in the model, students who receive high estimates and deserved them tend to keep them; those who received low estimates that they deserved tended to keep them while students who were marked low deserved better tended to go up and students who were marked low correctly tended to stay there…”

The issue is, of course, what is meant by “deserved”? The answer is, on the rational of SHD, that a student with a high estimate “deserves” to keep it given the school’s past record.

87. It is not at all surprising that the Minister was not a supporter of SHD. Her view was, as stated by Mr. Tattan in evidence, that: - “And the way she expressed that was when, say, a candidate goes into an exam hall, whatever they do or don’t do it is their effort on the exam paper and nobody else’s.” This view was also supported by Professor Van der Linden in the course of his evidence.

88. In considering the matter of fairness, it is necessary to look at what calculated grades would have been awarded to the applicant had SHD been retained in the standardisation model. On the evidence of Mr. McManus, I have found, as a fact, that the SHD would have been used as it was in Model 18(a). This was a dialled down version of SHD. The comparator is Model 18(g), where SHD was removed completely. (The final model was Model 21(a), but the changes in this model do not appear to be relevant to the issue in these proceedings). The exercise of comparing the applicant’s grades under Models 18(a) and 18(g) was carried out by Ms. Andrea Feeney, Director of the Calculated Grades Executive Office. (This exercise was not carried out for the subject of Latin). The findings were as follows: - Model 18(a): - Irish – H2 English – H3 Mathematics – H2 French – H3 Chemistry – H2 Biology – H2 Model 18(g): - Irish – H2 English – H3 Mathematics – H2 French – H3 Chemistry – H2 Biology – H2 It will be immediately seen that the removal of SHD did not affect the applicant’s calculated grades.

89. The other decision taken was not to apply the “mapping tool” to the NHD. Had this been done, some 58% of grades (higher level) would have had to be downgraded. This was the source of some of the grade inflation. If the grade inflation was even across the board, then it cannot be said there was any unfairness to the applicant. However, it was maintained by the applicant that the grade inflation was not even and, effectively, that it favoured those schools that had a less strong history of achievement in the Leaving Certificate. Mr. McManus compared the results of Belvedere College over the last three years with those of 2020. The comparison showed that the school’s standing relative to other schools in respect of its results did not suffer. This can be summarised as follows: - Table 1h: scores on composite scale – Belvedere College 2020 2020 calc 2017 2018 2019 sch. est. grades Mean grade 98.5 101.1 100.4 108.5 105.9 Std. dev. 19.6 16.8 15.8 17.8 17.0 School rank by mean grade 86 64 75 52 71

90. Professor Cathal Walsh, Professor of Statistics, University of Limerick, gave evidence on behalf of the applicant. Professor Walsh questioned whether you could compare the Leaving Certificate data of 2020 with previous years, given that the data of previous years had been collected on an entirely different system, being actual exam results. Professor Walsh is clearly correct in saying that the results of Leaving Certificate 2020 were arrived at in an entirely different way to those of previous years. However, in my view, the point of calculated grades for 2020 was to enable a comparison with previous years. In my view, it is a legitimate exercise to compare 2020 with previous years for the purposes of identifying anomalies. Professor Walsh placed emphasis on the outcomes of Models 10 and 17, which I have already referred to at para. 33 above. It is the case that these Models only looked at two schools and one subject, were experimental in nature and would not have been the final model that would be used for the award of calculated grades. Also, Model 17 did not preserve the class rankings as submitted by the schools. In these circumstances, I am of the view that it is not of assistance for the applicant to rely upon SHD, as used in Model 17, to demonstrate unfairness to him.

91. In his affidavit, Professor Walsh stated that the difference between the output of Model 18(a) to 18(g) varied “but amounted to of the order of up to a few percentage points, depending on the subject”. Although, he did say that the movement from 18(a) to 18(g) “had a much larger impact on some schools”. However, these schools were not identified, nor, indeed, was the degree of movement. Overall, Professor Walsh accepted that the effect of the removal of SHD was to move the calculated grades awarded closer to the school estimates. Given the fact that there was a considerable amount of overestimation on the part of the schools, this would point towards reduced downgrading.

92. In comparing the performance of Belvedere College between 2020 and previous years, Professor Walsh accepted that there were no changes that would be “irrational, capricious, averse or completely disproportionate”. Expert evidence was given on behalf of the respondents by Professor Wim van der Linden, Professor Emeritus of Measurement and Data Analysis of the Faculty of Behavioural, Management and Social Sciences, University of Twente Drienerlolaan, Overijssel, Netherlands. As has already been referred to, Professor Van der Linden had serious doubts as to the predictive value of SHD, saying that “historic distributions are for different students – the same school but different students ..” and “—in the case of the applicant, that if (he) is at a 36 percentile in 2020, according to the teachers, that he should be at the same percentile in the past. – we’ll never know –”. Professor Van der Linden did accept that SHD did remain part of the model in the form of the use of Junior Certificate grades achieved in the past.

93. Professor Van der Linden emphasised that the standardisation model was a statistical exercise. He accepted that the changes made in the standardisation model could mean that some 2020 candidates had “bad luck”. When it was put to him that the “unlucky students” could have been the group with the highest marks, he did not accept that this could have been corrected by retaining SHD. He stated “…using historical data does not tell me anything about an individual student in the 2020 cohort”. Professor Van der Linden accepted that downgrading was a function both of teacher estimates and what the standardisation model predicted. A student could be downgraded because of teacher overestimates or because of “bad luck” which is a statistical consequence.

94. In considering the evidence, both on affidavit and the cross examination of a number of deponents, a number of conclusions can be reached. It is a fact that in order to preserve public acceptance of the system of calculated grades, significant “grade inflation” was permitted by the respondents. However, the initial significant “grade inflation” commenced with the estimated marks submitted by the schools. In some cases, these estimated marks were a multiple of what had been achieved in the past. In order to bring the results of Leaving Certificate 2020 back into line with previous years, downgrades of the order of 58% would have had to have been applied. Downgrades on this scale were not, as a matter of policy, acceptable to the respondents. On a statistical analysis, Belvedere College does not appear to have been adversely affected.

95. In the course of the hearing, it became apparent that the use of SHD was questionable on two grounds. Firstly, statistically and, secondly, as a matter of policy. On the statistical side, the standardisation model had undergone various iterations, and by Model 18(a) SHD had been “dialled down to the greatest possible degree”. If the Minister had never intervened on the use of SHD, it would have been used as it was in Model 18(a) and would not have improved the applicant’s calculated grades. Further, Professor Van der Linden expressed doubts as to the predictive powers of SHD.

96. On the policy side, the use of SHD meant incorporating “inequality” into the standardisation model. This may not be acceptable in that the thrust of education policy must be to reduce inequality. Further, it is difficult to see how there could be an entitlement to an inequality so as to amount to a legal obligation.

97. By reason of the foregoing, I am not satisfied that it has been demonstrated that either the applicant, or his school, have been the subject of an unfairness arising from the removal of SHD and the minimising of NHD in the standardisation model that was applied.

Public interest

98. Even were the applicant to satisfy the various conditions identified by Fennelly J. in Glencar Exploration Plc v. Mayo County Council (No. 2) to establish a claim for legitimate expectation, he would still be faced with the issue of “public interest”. As Fennelly J. stated at the conclusion of the passage cited above (para. 75): - “Equally they are qualified by considerations of the public interest including the principle that freedom to exercise properly a statutory power is to be respected. …” Though a statutory power is not involved here, matters of public interest are.

99. In Glenkerrin Homes v. Dun Laoghaire Rathdown County Council [2011] 1 I.R. 417, Clarke J. (as he then was) stated: - “It is clear from the passage from Glencar Exploration p.l.c. v. Mayo County Council (No. 2) 1 I.R. 84 that the promise or representation may be expressed or implied. I am satisfied that an implied representation can derive from the universal following of a particular practice for a prolonged period of time. It is, of course, important to note that the executive enjoys a constitutional entitlement to change policy. ...”

100. An example of a change of policy defeating a claim of legitimate expectation was well illustrated in Curran v. Minister for Education [2009] 4 I.R. 300. In this case, the applicants were post primary school teachers who wished to apply for early retirement to take effect at the end of the school year 2008/09, pursuant to an early retirement scheme of the Department of Education and Science which had been in operation since May, 1997. A circular from the said Department invited applications for early retirement in respect of the school year 2007/08. The circular stated expressly that the scheme would operate again in the school year 2008/09. Due to the gathering financial crisis in November, 2008, the Minister for Finance announced that the scheme was being withdrawn. Thus, the applicants were denied the opportunity of submitting applications under the scheme, which they had intended to do. The applicants claimed that they had a legitimate expectation that the scheme would be available to them, and that they had acted significantly and to their detriment in the reliance on that expectation.

101. In the course of her judgment, Dunne J. reviewed the law on legitimate expectation, including the distinction, if any, between a procedural or a substantive legitimate expectation. She referred to the decision of Laws L.J. in R. v. North and East Devon Health Authority; ex parte Coughlan [2001] Q.B. 213 and stated: - “However, Laws L.J. concluded that distinction between procedural and substantive legitimate expectation was not helpful. He held that the doctrine was founded upon the constitutional principle of good administration under which public bodies ought to deal straightforwardly and consistently with the public.”

Dunne J. further stated: -

“I am of the view that the applicants herein seek to avail of a scheme which confers a substantive rather than a procedural benefit. …”

102. On the issue of “the public interest” Dunne J. stated: - “There is no doubt that it is open to the executive to change their policies. Fennelly J. in Glencar Exploration p.l.c. v. Mayo County Council (No. 2) [2002] 1 I.R. 84 recognised that the doctrine of legitimate expectation could be qualified by public interest considerations. … This particular qualification of the doctrine was endorsed by MacMenamin J. in Power v. Minister for Social and Family Affairs [2006] IEHC 170, [2007] 1 I.R. 543 at p.566:- ‘(28) … the statement that a legitimate expectation will arise only if the court thinks that there is no good reason of public policy why it should not is certainly applicable in the instance of a discretion or power made pursuant to a statue or statutory instrument which is exercisable for the good of the public or a specific section thereof. Thirdly, it is clear that the court must ultimately carry out a balancing exercise between the interest to the applicant and the public interest in the unfettered exercise of the decision maker's discretion.’

(37) That a legitimate expectation may be overridden by virtue of the public interest has also been recognised by the English courts in, for example, R (Nadarajah) v. Secretary of State for the Home Department [2005] EWCA Civ. 1363, [2005] All E.R. (D) 283 (Nov) and R v. North and East Devon Health Authority, ex parte Coughlan [2001] Q.B. 213.” Applying these principles to the facts of the case before her, Dunne J. stated: - “(39) I am satisfied that declining economic circumstances were such that the overriding public interest in taking the decision to suspend the scheme must outweigh any legitimate expectation the applicants had to pursue their applications under the scheme.”

103. Earlier in this judgment, I set out the circumstances that arose in this country following the controversy in the awarding of calculated grades in the UK. I set out the options that were facing the respondents to maintain public acceptance of the calculated grades system. The respondents were fully entitled, despite earlier commitments, to make changes to the standardisation model which they considered to be in the public interest. As stated earlier, this Court has neither the competence nor the jurisdiction to review such a decision.

Conclusion

104. By reason of the foregoing, I conclude that the decision of the respondents not to apply school historic data (SHD) and not to apply the “mapping tool” to national historical data (NHD) in the standardisation model for the award of calculated grades was not arbitrary, unfair, unreasonable, irrational and unlawful and in breach of the applicant’s legitimate expectations.

Summary

105. The cancellation of the traditional Leaving Certificate for 2020 required the Minister to put in place an alternative system for grading students and awarding Leaving Certificates. Simply to postpone Leaving Certificate 2020 for a year was not an option. The class of 2020 could not be left stranded. The system for the awarding of calculated grades was a difficult and complex exercise that had to be put in place in a matter of weeks.

106. There were two fundamental requirements for the calculated grades system. Firstly, it had to be statistically accurate; and, secondly, it had to have the support of those involved in third level education, future employers of the class of 2020 and the public in general. These requirements are not always compatible. Certain data which could lead to a more statistically accurate result may not be acceptable to the public in general, e.g.: details of gender and the socio economic backgrounds of students. In the course of August, 2020, school historical data (SHD) and a widespread downgrading of teacher/school estimates fell into that category. In the end, a degree of statistical accuracy had to be sacrificed to gain public acceptance.

107. In the lead up to 27 July 2020, the date for candidates, including the applicant, to opt into the calculated grades system all the documentation and information that came from the Department of Education and Skills clearly stated that the data that would be used in the standardisation model for the award of calculated grades would include SHD. It was also clearly stated that the Leaving Certificate of 2020 would be comparable with, and have the same standing as, the Leaving Certificates of previous years.

108. In August, 2020, a widespread and serious controversy arose in the United Kingdom concerning the awarding of calculated grades. This controversy focussed on what was described as “school profiling” and a “post code lottery”. Almost immediately, politicians and commentators in this country drew parallels between the use of SHD in the Irish system and its use in the UK system. There then followed a controversy, similar to that in the UK, centred on the use of SHD. The fact that the UK system (to a lesser extent the system in Scotland) was very different and, indeed, in some respects less sophisticated than the Irish system was lost in the controversy. All this required a response from the Minister as it was believed that public acceptance of the calculated grades system, as envisaged, was in serious jeopardy.

109. There was a serious debate amongst officials in the Department as to the effects of the removal of SHD from the standardisation model. The Minister herself, well before the eruption of the controversy, had serious misgivings on the use of SHD. Having considered a number of iterations of the standardisation model on the use of SHD to varying degrees, a decision was taken by the Minister and the respondents to remove it completely from the model. Given the potential number of downgrades from teacher/school estimated marks it was further decided to minimise the effects of national historical data (NHD) by not applying the “mapping tool” in the model. This led to significant “grade inflation” in the results of Leaving Certificate 2020 with consequent effects on the points requirements for certain third level courses.

110. The decision to remove SHD and minimise the effects of NHD were clearly breaches of the commitments that had been given to the applicant and other candidates for Leaving Certificate 2020. This formed the basis of the issue which the court has to decide namely whether these decisions were “arbitrary, unfair, unreasonable, irrational and unlawful and in breach of the applicant’s legitimate expectations”.

111. I am of the view that the decisions to remove SHD minimise the effects of NHD were an exercise of executive power under Article 28.2 of the Constitution. Further, these decisions involved an area of policy, namely how to maintain public acceptance of the calculated grades system. In following the various legal authorities which I have referred to, I am of the view that this Court, given the Constitutional provisions on the separation of powers, does not have the jurisdiction to review these decisions. Further, I am satisfied that the Court does not have the competence to make a finding as to which was the appropriate course for the Minister and the other respondents to take when faced with the public controversy, as described.

112. The applicant has not sought to base his case on a failure to observe or vindicate his rights under the Constitution nor has he identified any breach of a statutory duty by the respondents. No case has been made that the Minister and other respondents making the decisions they did were in “clear disregard” of the duties imposed upon them by the Constitution and law. Rather the applicant has sought to make his case as being one of breach of his legitimate expectation.

113. Though the applicant has met some of the conditions necessary to establish a breach of his legitimate expectation, he has not satisfied others: -

(a) The applicant has not established that he suffered any material unfairness as a consequence of the said decisions. Certainly he has not established an unfairness as would lead the Court to the conclusion that the system, following the said decisions, for the awarding of calculated grades was unlawful;

(b) The applicant’s complaint is that he was unfairly downgraded from his teacher(s)/school estimated marks. I have set out the relevant estimates as were submitted by Belvedere College. It is very clear that these estimated marks were significantly inflated being considerably ahead of what had been achieved by the school in past years. It is not at all surprising that the applicant was downgraded from these inflated estimates. Inflated school estimates were not unique to Belvedere College and were common amongst other schools particularly at the higher levels. These findings should not be taken as criticism of the teachers or school involved. Teachers were placed in a very difficult and invidious position in having to give estimated marks to their students, whom they would have known and worked with for many years, in one of the most important examinations of their lives;

(c) Belvedere College maintained its position relative to other schools following the calculated grades of Leaving Certificate 2020. This position was maintained notwithstanding the “grade inflation”. It should be noted that the starting point for grade inflation was the estimated marks submitted by the schools; and

(d) Had SHD been retained in the standardisation model it would have been as it was in Model 18(a) namely “dialled down to the greatest possible degree”. An exercise in calculating the applicant’s grades with Model 18(a) being applied showed little difference from the calculated grades which the applicant did, in fact, receive.

114. Even if the applicant had satisfied the conditions necessary to maintain a claim for breach of his legitimate expectations he would still be met with the defence of “public interest”. I am satisfied that the respondents are entitled to maintain that the decisions to remove SHD and minimise the effects of NHD (not using the “mapping tool”) were ones taken in the public interest to maintain public acceptance of the calculated grades system

115. I am therefore satisfied that the removal of SHD and the minimising of NHD in the standardisation model for the awarding of calculated grades was not “arbitrary, unfair, unreasonable, irrational and unlawful and in breach of the applicant’s legitimate expectations”, as claimed by the applicant.

Consequential orders

116. Given the ruling that has been made, the Court will have to consider the issue of costs. I would ask the parties to make written submissions on this, and other consequential orders, within fourteen days.

117. On the matter of costs, I would ask the parties to consider that this was the “lead case” for the purposes of determining a central issue that is common to numerous other applications.

APPROVED [2021] IEHC 128

Appendix A: Use the school estimated mark for each student to create a Respondents’ Aide-mémoire distribution describing the likelihood that this student will get any particular score in the examination (with the most likely mark being the one the school predicted, and other marks getting progressively less likely as you move farther from this mark). Combine these individual likelihood functions to give an overall expected distribution of the LC results of this group School Estimates of students in this examination.

Distribution No. 1

Map the scores from the original school-

estimate distribution to this overall

Regression Distribution combined (‘conditioned’) school

No. 2 distribution. Use historical national data sets of JC The (“equipercentile”) mapping exercise and LC results, linked at a candidate Combine these three (later, two) expected identified and preserved gaps between level, to build a set of equations that distributions into one overall expected students with uncharacteristically low or describes how likely it is that a Apply the JC regression model to high school estimates compared to other distribution. Note that this overall person with a given set of JC results every student in the school group JC students in the same school.

expected distribution already incorporates

(&TY completion & repeat status) for the LC examination the school’s estimates as a strongly will get any particular score in the concerned. Combine these influential component. (The other examination you are trying to predict. individual likelihood functions to component distributions(s) i.e. regression This is the ‘JC regression model’. give an overall expected and school historical are referred to as distribution of the LC results of ‘conditioning’ distributions.)

“A” this group of students in this

examination.

Option to use another mapping

exercise at a national level to map the

scores from the entire cohort for the

School Historical examination to the historical national

Distribution No. 3 distribution (not availed of). “C” Based on the results from this school in this subject in the past, and on the assumption that the current students are similar to those who have attended this school in the past, create an overall expected distribution of the LC results of this group of students in this examination.

“B” Convert the calculated mark to a

(Note that in some earlier versions of the model, this distribution was calculated grade. a blend of the school-historical and national-historical distributions.)

[(validated by reference to Gender, DEIS/Non-DEIS etc)]