Readers like you keep news free for everyone.
More than 5,000 readers have already pitched in to keep free access to The Journal.
For the price of one cup of coffee each week you can help keep paywalls away.
Shutterstock/marekuliasz
# Leaving Cert
One in five Higher Level grades reduced and a tweaking of the gender gap: How 'standardisation' changed teacher-estimated grades
10% of Leaving Cert Higher Level grades were reduced by up to 10 marks.
ONE IN FIVE Higher Level Leaving Cert Calculated Grades were downgraded from their teacher-assessed mark by one grade due to standardisation, according to the Department of Education.
At Ordinary Level, one in ten grades were downgraded by one grade.
The figures are contained in over 250 pages worth of documents published by the Department of Education, which explain the process used to standardise results and produce this year’s Leaving Cert grades.
The system – which opposition TDs had called on the government to publish to ensure transparency in the Calculated Grades process – uses sets of data to bring teacher-assessed percentages closer in line with previous Leaving Cert results.
Among the information that had already been known about the system is that it uses Leaving Cert students’ own Junior Cert results, as well as the average grade in each subject in the Leaving Cert the past three years, to standardise results.
The Minister for Education Norma Foley has repeatedly said that the system is “blind to gender” and does not use the location of students’ schools in standardising grades.
The link to the standardisation information can be found here.
1. Gender gap
The school-estimated grades resulted in a wide gender-breakdown: there was a gap of 5.7, 5.9 and 6.5 points respectively in 2017, 2018 and 2019 between female students scores and males (with females ahead).
While the gap had widened in successive years over the period 2017 to 2019, the increase to 7.9 points is too great to be considered a continuation of a trend.
The teacher-estimated marks put the gap this year at 7.9: standardisation reduced this to 7.6 points, which means that female students were downgraded more than male students.
Interestingly, the gender gap in exam scores tends to be wider among students attending mixed-sex schools than in single-sex ones. This trend remained in 2020 both in teacher-based assessments and in the Calculated Grade results.
2. A fifth of Higher Level grades lowered by one grade
Last week the Department of Education released detail about what percentage of Calculated Grades had been awarded to students. In the documents published today, more detail was given about what degree grades were lowered during standardisation.
One out of five Higher Level Calculated Grades were lowered by one grade, according to the Department’s documents:
- Although 77% remained unchanged (215,815 grades), 20% were reduced by one grade (55,892 grades), while 347 grades were reduced by two grades and 2 grades were lowered by three.
- 3% of Higher-Level grades were increased by one grade (8,964), 80 were increased by two grades, and 6 Calculated Grades were increased by three grades.
- In Ordinary Level, 85% of grades were unchanged in the standardisation process (85%), while 9% were lowered by one grade (10,253 grades), and 6% were increased by one grade (6,069).
- At Foundation Level, 94% of Calculated Grades remained unchanged (3,821), 3% were lowered by one grade (117) and 3% were increased by one grade (114).
“Most of the mark adjustments did not lead to changes of grade,” the report says.
A report compiled by the National Standardisation Group, which includes experts that oversee the implementation of the standardisation process, gives a breakdown of how dramatically grades were altered, giving a mark-breakdown per Leaving Cert level.
A mark refers to a point given when correcting exams, which is then converted to a percentage (eg, ten marks awarded out of 20 is 50%).
It states that:
- 0.5% of final Calculated Grades at Higher Level were reduced by more than 10 marks (1,761), 9.7% reduced by 6-10 marks (27,239), and 52.6% reduced by 1-5 marks (147,896). 16.8% remained unchanged (47,324), while 19.3% were increased by 1-5 marks (54,155).
- At Ordinary Level, 28.9% remained unchanged (31,691), 33.2% were increased by 1-5 marks (36,494), and 32.9% were decreased by 1-5 marks (36,041).
- At Foundation Level, 62.4% remained unchanged (2,528), 23.8% increased by 1-5 marks (962), and 11.6% were reduced by 1-5 marks (474).
3. Clustering
The appendices of the National Standardisation Group’s report notes that teachers were prone to ‘clustering’ marks when assessing their own students, giving marks close to “known locations of grade boundaries”.
This means they tended to give marks in multiples of five, and a graph of this year’s Leaving Cert scores tallies with that expected trend:
Although the Department of Education warned against this in giving guidance to teachers, it appeared anyway, as was expected.
Clustering was less apparent in leaving Cert Applied subjects, which is likely to do with the different numbers of credits associated with different exams and tasks, meaning that teachers are less sure of where the threshold mark for a certain grade is.
To tackle the clustering issue, the Department of Education said that school estimates were “combined and smoothed to produce a broadly supported discrete distribution for the entire school”.
The documents published today were the Discussion Paper for SEC-DES Technical Working Group on Calculated Results (39 pages); the Report from the National Standardisation Group (205 pages); the opinion of the Independent Steering Committee (12 pages); and the External Reviewer’s statement (5 pages).
