A student returns from an exchange with a result that looks unfamiliar: a number on a different scale, a letter whose meaning seems obvious, or a percentile printed beside a course. The first temptation is to turn everything into one percentage. That can produce a tidy answer while losing what the original result actually means.
Read an ECTS grade distribution table as a description of how passing grades were awarded within a defined reference group. Compare the relevant distributions and their cumulative ranges, then use the responsible university's published conversion method. Keep the original grade, the reference group and the institutional decision together. A numerical resemblance alone does not establish an equivalent grade, credit award or admission outcome.
Find your next learning step
Read grade frequencies and overlapping percentile ranges, distinguish credit recognition from grade conversion and prepare a precise question for the responsible university.
ROUTE 1 · CHAPTERS 1–3
Identify the comparison
Clarify the purpose, grade scale and reference group.
ROUTE 2 · CHAPTERS 4–6
Work through the distributions
Build frequencies, interpret ties and compare overlapping ranges.
ROUTE 3 · CHAPTERS 7–9
Keep meanings separate
Read local notation, distinguish credits and avoid conversion shortcuts.
ROUTE 4 · CHAPTERS 10–12
Prepare and test the evidence
Check limits, build a university query and try a fresh table.
ROUTE 5 · CHAPTERS 13–15
Make the next decision
Use practical roles, FAQs and current primary documents.
Full chapter index · Independent learning check · Education parent guide
Full chapter index
Identify the comparison · Chapters 1–3
Work through the distributions · Chapters 4–6
Keep meanings separate · Chapters 7–9
Prepare and test the evidence · Chapters 10–12
Make the next decision · Chapters 13–15
CHAPTER 1 OF 15 · Identify the comparison
Identify which question the record needs to answer
The European Credit Transfer and Accumulation System, or ECTS, supports transparency in higher education. Its name can encourage several different questions to be treated as one.
How much learning has been credited? What result was awarded? How was that grading scale used in a particular field? How will a receiving or home institution interpret the result? Each question needs its own evidence.
The European Commission's ECTS overview describes credits through learning outcomes and associated workload. A credit quantity is not a grade, and it does not tell you where a result sits within a grade distribution.
This guide concerns interpreting grade distributions, especially in student mobility. It uses the official ECTS Users' Guide 2015 and current institutional information checked on 23 October 2026.
The named university example is Ghent University in Belgium. The numerical tables and learners used for teaching below are invented. They are not Ghent data, an authorised conversion service or a rule for any particular applicant.
The Education parent guide asks whether learning remains useful when the setting changes. Here the useful capability is reading a result with its conditions attached, rather than carrying an isolated number into a new system.
CHAPTER 2 OF 15 · Identify the comparison
Separate a grade scale from its actual distribution
A scale describes the grades that can be awarded and their order or meaning. A distribution describes how frequently the grades were awarded in a specified group and period.
Two institutions might use numbers from zero to twenty while using the upper end differently. Conversely, different-looking scales might contain grades occupying comparable positions within their respective distributions. The printed maximum alone does not settle either comparison.
The ECTS guide recommends transparent reference groups and distributions of passing grades, with percentages and cumulative percentages. It also explains that overlapping grade ranges require an institutional approach decided in advance.
The guide's distribution method replaces the earlier cross-system ECTS A–E grading scale. That statement does not mean that every local letter grade is obsolete. A university may still use its own letters; the question is which system a letter belongs to and how that system defines it.
A useful first check is to label the object in front of you. Is it the institution's local scale, an actual distribution table, a conversion rule or an unofficial comparison chart? These documents can look similar while doing different jobs.
The existing Singapore guide to comparing grades and qualifications develops the broader distinction between recognition, equivalence and admission. The narrower task here is to read distribution evidence accurately enough to ask a precise conversion question.
CHAPTER 3 OF 15 · Identify the comparison
Read the reference group before calculating
A distribution has a boundary. It may concern a field of study, a programme grouping, a study level and an identified period. The boundary gives the percentages their meaning.
Before comparing tables, ask what each counted item represents. The ECTS guide discusses passing grades awarded. That does not necessarily mean one observation per unique student. A person can contribute results from more than one educational component, depending on the table's construction.
Also identify which results are included. A table of passing grades does not, on its own, establish the proportion of all assessment attempts that passed. The denominator has already excluded a different category.
A small original example makes this visible. Suppose a record contains one hundred passing grades and a separate record contains twenty failed attempts. A grade category holding twenty of the passing grades accounts for twenty per cent of that passing-grade distribution.
It does not account for twenty per cent of all one hundred and twenty attempts. That would be twenty divided by one hundred and twenty, or approximately 16.7 per cent. The numbers answer different questions.
Neither denominator should be substituted silently for the other. If the task concerns a published passing-grade distribution, use its stated population. If the task concerns a success rate, obtain the data defining that rate.
The practical habit is to write a small label beside every percentage: percentage of what, in which group, over which period? That label prevents much of the apparent certainty created by a bare figure.
CHAPTER 4 OF 15 · Work through the distributions
Build an original distribution from counts
Consider a fictional institution using four passing grades, ordered from highest to lowest as H, J, K and L. These are invented local labels, not ECTS grades.
Its reference dataset contains one hundred passing grade records. Twenty are H, forty are J, thirty are K and ten are L.
| Fictional local grade, highest first | Passing grade records | Share of passing records | Cumulative share from the highest |
|---|---|---|---|
| H | 20 | 20% | 20% |
| J | 40 | 40% | 60% |
| K | 30 | 30% | 90% |
| L | 10 | 10% | 100% |
The share for J is forty divided by one hundred, giving forty per cent. The cumulative share through J is twenty plus forty, giving sixty per cent. The cumulative column adds the categories from the best end of this explicitly ordered scale.
Check the totals before interpreting the pattern. The counts add to one hundred, the separate shares add to one hundred per cent and the final cumulative share reaches one hundred per cent.
The cumulative column is not a second collection of independent percentages to be added together. Adding twenty, sixty, ninety and one hundred would count earlier categories repeatedly.
A common reading error confuses the forty per cent awarded J with the sixty per cent awarded J or a higher grade. Both are accurate statements, but they concern different groups.
Another error assumes that H must mean the same thing as a familiar high grade elsewhere. Our labels supply an order for a teaching model. They provide no syllabus, assessment criteria or institutional conversion authority.
The table becomes useful when the reader can explain each column and reproduce one row from the underlying counts. Familiarity with the table's appearance is a weaker achievement.
Check rounding before declaring the table inconsistent
Consider another original practice dataset with two hundred passing grade records. The three category counts are thirty-three, sixty-seven and one hundred. Their exact shares are 16.5 per cent, 33.5 per cent and fifty per cent.
If the first two shares are displayed as whole percentages rounded upwards at the half point, the visible entries become seventeen, thirty-four and fifty. They add to one hundred and one, although the underlying counts are consistent.
That discrepancy does not, by itself, establish an incorrect grade or a damaged source dataset. It may concern the display precision. Read the table's notes and, where available, compare the percentages with the counts.
The appropriate repair is not to reduce a student's grade until the displayed percentages total one hundred. A category grade and a rounded frequency display are different objects.
Likewise, recalculating cumulative values from already rounded entries can create a slightly different boundary from the one calculated from the original counts. Keep enough precision for the stated method and use the institution's actual table for an institutional decision.
This exercise shows why a numerical check needs an interpretation. Totals are valuable warning signals, but the reader still has to identify which part of the record produced the difference.
An adviser can be asked a precise question: are the displayed percentages rounded, and which values are used in the authorised conversion? That question is easier to answer than an accusation that the whole table is wrong.
CHAPTER 5 OF 15 · Work through the distributions
Understand why a tied grade occupies a range
In the fictional table, twenty per cent of passing grade records are higher than J. Forty per cent are J, and forty per cent are lower.
The J category therefore occupies the cumulative range after twenty per cent and through sixty per cent when we accumulate from the highest end. Everyone with J belongs to that grade category, but the table does not rank individuals within it.
A learner may take the middle of the range, forty per cent, and describe it as their exact rank. That would add information the table does not contain.
The midpoint can be a useful mathematical description of an interval in some contexts. It becomes an official individual ranking or conversion device only when the relevant method establishes that use.
Similarly, saying that the learner is in the top sixty per cent when including everyone with the same or a higher grade describes a group boundary. It does not establish that exactly sixty per cent performed better.
The distinction matters because tied results can cover a wide interval. A grade awarded to forty per cent of the reference records cannot, by itself, locate a person at one precise point inside those forty percentage points.
Try a changed distribution. Keep the same local labels but reduce the J category to ten per cent. Even if the student still receives J, the category's position and width may differ. The letter alone has not preserved the distribution.
The useful reading records the higher, equal and lower groups separately, then states the limit. That produces a more faithful account than turning a broad tied band into a precise-looking personal percentile.
CHAPTER 6 OF 15 · Work through the distributions
Compare two tables without inventing an automatic match
Now introduce a second fictional institution. Its passing grades run from highest to lowest as R, S, T and U.
| Fictional receiving scale, highest first | Passing grade records | Share of passing records | Cumulative share from the highest |
|---|---|---|---|
| R | 10 | 10% | 10% |
| S | 30 | 30% | 40% |
| T | 40 | 40% | 80% |
| U | 20 | 20% | 100% |
In this model, S occupies the cumulative range after ten per cent and through forty per cent. T occupies the range after forty per cent and through eighty per cent.
The first institution's J range, after twenty and through sixty per cent, overlaps both S and T. It overlaps S across twenty percentage points and T across twenty percentage points.
That observation is a calculation about our invented tables. It does not establish which receiving grade must be awarded. The institution's actual method is still needed.
A student might prefer S because it sounds higher. Another might use the midpoint of J's interval and choose a category from that point. Neither preference becomes authoritative simply because the arithmetic is transparent.
The ECTS guide makes institutional handling of overlapping ranges an explicit question. Ask which approved method applies before making a prediction about a particular result.
Change the source result to L. Its range is after ninety per cent through one hundred per cent, lying inside U's range in this model. The overlap is simpler, but an actual institutional decision would still require the appropriate tables, scope and rules.
The tables make comparison more informative. They do not replace the university with a personal conversion calculator.
CHAPTER 7 OF 15 · Keep meanings separate
Read Ghent University's percentile notation within its own system
Ghent University's grade conversion and ECTS page, checked on 23 October 2026, explains a transcript notation in the form A(B)C: the percentages with a strictly lower grade, the same grade and a strictly higher grade.
Its published example, 55(10)35, therefore describes fifty-five per cent lower, ten per cent equal and thirty-five per cent higher. The equal-or-higher group is forty-five per cent.
The order differs from our cumulative-from-highest tables. Read the notation before carrying a direction from one display into another.
The page also distinguishes grade interpretation from an institution supplying a conversion for another body's purpose. It places responsibility for transferring study-abroad results into home grades with the home university.
These are statements about the named university's published explanation. They do not create a general right to a GPA conversion, a master's admission offer or a particular grade at another institution.
When reading an actual table, retain the academic period and level printed with it. Do not use an older bachelor's distribution as though it were automatically the current reference for a master's-level result.
A useful question to the institution might identify the course result, the transcript notation and the intended interpretation. That gives the responsible office something more precise than asking whether a number is good.
Use the named example to learn the notation, then return to the documents governing your own record. The direction of the comparison and the authority for the decision both matter.
CHAPTER 8 OF 15 · Keep meanings separate
Keep credit recognition and grade conversion on separate lines
Suppose a fictional exchange plan contains an approved educational component carrying a stated credit value. The student completes it and receives a local grade.
Recognition asks how the completed learning fits the home programme under the applicable agreement and rules. Grade conversion asks how the result is represented or interpreted on the home scale. A credit total does not answer the second question automatically.
The existing Education article on Erasmus learning agreements when courses change develops the relationship between the study undertaken abroad and the learning recognised at home. Preserve that agreed relationship when reading the later result.
Similarly, Bologna credit recognition after transfer, withdrawal or a previous degree concerns a particular recognition process. It should not be treated as an ECTS-wide grade-conversion rule.
An original planning record can therefore contain three entries: the original educational component and result, the applicable recognition decision, and the applicable treatment of the grade. If one is unresolved, mark it unresolved.
Do not fill the gap with a plausible calculation. A student can accurately add credits while misunderstanding which degree requirement they satisfy. They can also read a grade distribution correctly while lacking the authority to determine the final home grade.
The learning aim is to explain the route from original evidence to each decision. Keeping the questions separate makes it easier to identify which document or office can answer the remaining one.
CHAPTER 9 OF 15 · Keep meanings separate
Diagnose the common conversion shortcuts
Several shortcuts create false clarity. One divides a local grade by the maximum and calls the result an equivalent percentage. Another matches letters that look alike. A third assumes that the same numerical grade has the same position everywhere.
Each shortcut removes conditions. The maximum does not describe how the scale is used. A shared letter does not establish a shared definition. A shared number does not establish a parallel reference group.
Return to the original record and ask which transformation is being proposed. Is the person rescaling a number, describing a distribution, applying an approved conversion rule or making an admissions judgement?
Those operations can all involve numbers. They do not become the same operation because a spreadsheet can display them in neighbouring columns.
A useful diagnostic exercise asks the learner to explain what would change if the distribution changed while the local scale stayed fixed. If they insist that the original grade must always map to the same foreign result, they have not yet incorporated the distribution evidence.
Another asks what would happen if the same distribution came from a different level or field. A similar pattern of percentages does not, by itself, establish comparable learning or assessment.
The repair is to restore the missing condition. Write the group, period, scale direction and responsible method beside the calculation. The aim is an intelligible comparison with clear limits, rather than an attractive answer produced by an unspecified rule.
Worked variation: a lower number is the better result
Suppose a third fictional local scale uses grades one, two and three, with one the highest passing grade. Its passing-grade counts are thirty, fifty and twenty out of one hundred.
Accumulating from the highest grade gives thirty per cent through grade one, eighty per cent through grade two and one hundred per cent through grade three. The direction is established by the scale's meaning, not by sorting the numbers from largest to smallest.
If a spreadsheet sorts three first because three is numerically greater than one, the cumulative calculation may be arithmetically tidy while its interpretation is reversed. The first repair is to restore the correct grade order.
Compare our source J interval, after twenty and through sixty per cent, with this new receiving scale. It overlaps grade one from twenty to thirty, and grade two from thirty to sixty. Those overlaps cover ten and thirty percentage points respectively.
A reader can describe the different overlap sizes accurately. They still need the responsible institution's method to determine the actual treatment. The larger overlap is not a rule we are entitled to impose privately.
Now imagine the student says that J must correspond to grade two because the midpoint of its interval is forty. That is a possible calculation about the interval. The unresolved question is whether the institution uses that approach for this purpose.
The variation reveals three separate decisions: how the local scale is ordered, how cumulative ranges are calculated and which institutional conversion method applies. Getting one right does not establish that the others are right.
For learning, ask the student to explain the scale without looking at the cumulative column. Which grade is best? Which category comes next? Only then reconstruct the cumulative figures.
This reduces the chance that a familiar spreadsheet convention quietly changes the meaning of the academic record.
CHAPTER 10 OF 15 · Prepare and test the evidence
Examine what the distribution can and cannot establish
A distribution describes recorded outcomes in the reference dataset. It does not independently explain why the pattern occurred.
A high proportion of one grade might raise questions about assessment design, cohort preparation, grading practices or the grouping of results. The table alone does not decide among those explanations.
Nor does a rarer high grade prove that every student receiving it has more knowledge than a student in another system. The learning outcomes, assessment and conditions remain relevant.
This is a valuable evidence boundary. Grade distributions can improve transparency while leaving important educational questions unanswered. Their usefulness does not require pretending that they measure every aspect of learning.
A student's own learning record can address a different question. What can they now explain or do? Which work demonstrates that capability? What support was present? Those questions may help preparation for the next course, but they do not privately rewrite the official grade.
Consider a learner whose converted record meets an administrative requirement but who has not recently used an important prerequisite. The conversion question may be settled while a readiness question remains.
The next action is to investigate the prerequisite through appropriate work. A formal record and a fresh learning task can each be useful without being interchangeable measures.
This distinction keeps an institutional comparison from becoming an unsupported claim about the whole person.
CHAPTER 11 OF 15 · Prepare and test the evidence
Build a concise record for the responsible university
Start with the original transcript and the institution's explanation of its grading scale. Add the relevant distribution table if supplied, retaining its reference group, academic level and period.
Then identify the actual purpose: returning exchange credit, interpreting a course result, applying for further study or another specified decision. The purpose determines which instructions and authority you need.
For a conversion question, ask which table and method the responsible institution uses. If two ranges overlap, ask how that situation is handled under its published approach. Do not assume that a method used for mobility also governs a different admissions process.
A short enquiry can identify the original course, local grade, relevant table and unresolved point. Avoid attaching a self-created equivalence as if it had already been approved.
Keep the response with the documents it interprets. A later reader should be able to distinguish the original result from a conversion decision and from the student's own explanatory notes.
If a document is unavailable, say which one. Missing evidence is easier to investigate when it has a name. A general statement that the university has not recognised my work can conceal several different unresolved questions.
The aim is a record whose meaning survives a handover. It should help the next adviser understand what was awarded, what has been decided and what still requires a responsible answer.
CHAPTER 12 OF 15 · Prepare and test the evidence
Try an independent distribution-reading exercise
Use a new fictional table with three passing categories, ordered highest to lowest as X, Y and Z. Their counts are fifteen, twenty-five and sixty out of one hundred passing grade records.
The category shares are fifteen, twenty-five and sixty per cent. The cumulative shares from the highest end are fifteen, forty and one hundred per cent.
For a Y result, fifteen per cent of records are higher, twenty-five per cent are equal and sixty per cent are lower. Explain why forty per cent means Y-or-higher rather than strictly higher.
Now double every count while preserving the categories: thirty X, fifty Y and one hundred and twenty Z. The total becomes two hundred, while the percentages remain unchanged.
The exercise separates a frequency from its share. A table can describe a larger reference dataset without changing the percentage pattern.
Change the definition instead. Suppose the new table concerns a different level of study. Matching percentages no longer establishes that the groups are appropriate for the same conversion purpose. The reference-group question must be revisited.
Finally, introduce a receiving table whose best category occupies the top twenty per cent and whose second category extends from twenty to seventy per cent. Y's source range, after fifteen through forty per cent, overlaps both.
A strong answer explains the arithmetic and identifies the institutional question left unresolved. It does not need to manufacture a single equivalent grade.
Keep a different purpose from silently changing the answer
Ella is an invented student who has received a home-university conversion after an exchange. Later, an unrelated institution asks for her original transcript when considering a new programme application.
She is tempted to send only the converted home result because it looks more familiar. The useful response is to read what the new institution actually requests and preserve the original evidence it needs.
An exchange conversion and a new admission process may involve different authorities, documents and purposes. The fact that one decision is complete does not establish that its converted number replaces the original record everywhere.
Ella can organise the documents into a short chain: the host transcript, its grading explanation, the home conversion decision and the new institution's request. The labels make it possible to see which document answers which question.
If the requested evidence remains unclear, she should ask the receiving institution rather than delete an unfamiliar part of the record. This fictional case supplies no admission rule for a named programme.
The transfer from the table-reading lesson is practical. Preserve the object, purpose and authority whenever a result moves into a new decision. A result can remain valid while needing a different explanation for a different reader.
Do not quietly turn frequencies into credit weights
An original two-course example shows another possible change of meaning. A student has one H result in a six-credit component and one J result in a three-credit component.
Counting grade records gives one H and one J: half of the two records in each category. Weighting the same results by the stated credits gives six of nine credits attached to H and three of nine attached to J.
Those are different calculations. Neither should be substituted for the other without identifying the method and purpose. A published frequency table of passing grades should be read according to its definition.
This example also does not create a GPA. We have supplied no authorised grade-point values or institutional averaging rule. Inventing numerical points for H and J would add another unsupported transformation.
Ask the learner to identify exactly what is being counted before opening a spreadsheet formula. Grade records, credits and unique students can all be legitimate quantities in an educational record. They answer different questions.
The small example helps preserve that distinction when a larger transcript contains many components, different credit values and several summary columns.
CHAPTER 13 OF 15 · Make the next decision
Give students, families and university staff workable roles
The student can preserve the original record, identify the intended purpose and ask a specific question. They can also learn enough of the table's arithmetic to recognise when a proposed explanation changes the denominator or direction.
A family member can help organise documents and dates. They do not need to turn an unfamiliar result into a familiar number immediately. A calm statement of what remains uncertain can be more useful than premature reassurance.
University staff can explain the applicable institutional process within their responsibilities. Students should use the office or academic authority identified by the institution, rather than collecting several informal answers and selecting the most favourable one.
A teacher can use fictional data to develop the reading skill without pretending to make the official decision. The learner should be able to reconstruct a category share, a cumulative range and an overlap.
Keep assistance focused on understanding. If an adult completes every calculation and writes the whole enquiry, the record may look polished while the student's interpretation remains insecure.
An appropriate next check changes one feature: a count, the order of a scale, the reference level or the purpose of the comparison. Ask the learner to explain which part of the earlier reasoning survives.
The shared aim is accurate interpretation and a manageable next action, supported by the actual institution's documents.
CHAPTER 14 OF 15 · Make the next decision
Questions students and families often ask
Is an ECTS credit the same thing as a grade?
No. Credits and grades describe different aspects of the record. Identify which question you are answering before making a calculation.
Does a cumulative percentage mean that proportion scored strictly higher?
Read the direction and inclusion rule. In our examples, the cumulative share includes the stated category and all higher categories.
Can a tied grade give me an exact individual rank?
The category can occupy a range. Without further information or an authorised method, the table does not rank individuals within the tie.
Should I always choose the higher receiving grade when ranges overlap?
The responsible institution's method governs the decision. The overlap calculation does not authorise a personal choice.
Are H, J, K and L official ECTS grades?
No. They are invented local labels used to make the original example easy to follow.
Does a matching percentile guarantee admission?
This guide establishes no such entitlement. Admission involves the actual institution, programme, qualification and applicant requirements.
What if the reference table is missing?
Identify the missing document and ask the responsible university how the result will be interpreted. Avoid silently replacing it with an unrelated table.
CHAPTER 15 OF 15 · Make the next decision
Use the current documents to make the next comparison clearer
Begin with the European Commission's ECTS overview and the official Users' Guide, particularly its grade-distribution and conversion sections. Use the actual universities' current documents for an individual decision.
Ghent University's linked explanation provides a named example of transcript notation and institutional responsibility. It is not an authority for every receiving university or every purpose.
The official sources in this guide were checked on 23 October 2026. All H–L, R–U and X–Z tables and student situations are original illustrations, not published institutional statistics.
Choose one result and identify its scale, group, period and purpose. Reconstruct one percentage and explain one range. Then name the question that only the responsible institution can settle.
A useful comparison preserves the meaning of the original evidence while making the next decision easier to understand. That is a stronger educational achievement than making two different numbers look the same.