Elimu Audit +Add my journey

How we calculate

Every figure on the Elimu Audit dashboard follows the rules on this page. They are published so anyone can check them, and any change is dated here.

Ground rules

These apply to every figure.

  • Privacy first. Any figure based on fewer than 5 people is hidden.
  • Early results from 10, judgements from 30. A course with 10 to 29 answers shows its figures as early results, with a likely range (a 95% range worked out with the Wilson method) and no verdict. Verdicts, rankings and comparisons need at least 30 answers. Below 10, a course shows "Too few answers". Both numbers can be changed in the admin settings.
  • Only graduates in the job market. Graduates doing further study (such as a Master's) or not looking for work (such as caregivers) are left out of outcome charts, so they are never counted as unemployed. Their number is shown under the charts.
  • Check-ins are counted separately. The 1-year, 3-year and 5+ year figures each use the people who answered that check-in. Skipped check-ins are left out.
  • "Not sure" is honest, not a match. These answers count in the total but never as working in the field, so they can't make a course look better.
  • Answers from memory count. They carry the same weight as current answers, but are flagged so the two can be compared.
  • Detail questions attach to the latest check-in answered. Underemployment, "pushed out" and "found their calling" come from the most recent check-in each person answered.
  • Checked before counted. Every answer goes through automatic checks. Clean answers count at the next update. Answers that fail a check don't count until the team reviews them, usually within 72 hours. If many answers for one course and year arrive in a burst, that group's recent answers are held until the batch is checked.
  • Whole numbers. Percentages are rounded to the nearest whole number, so groups may not add up to exactly 100%.

Indicators

Question codes (B1, C3 and so on) refer to the survey question list.

IndicatorFormulaWho is counted
In what they studied

Doing exactly what the course trained them for.

Exactly what I studied ÷ everyone working or looking for work at that check-inAnswers to B1 and B2; further study and not looking are left out
In field or related job (market alignment)

Working in exactly what they studied, or in a different job in the same industry. For example, an accounting graduate working as an accountant, or in a bank. Higher is better. The main measure of whether a course leads somewhere.

(Exactly what I studied + Same industry) ÷ everyone working or looking for workAnswers to B1 and B2; further study and not looking are left out
Different field

Working, but outside both the course and its industry.

Different field ÷ everyone working or looking for workAnswers to B1 and B2
Graduate unemployment

Available for work but unable to find any.

Looking for work ÷ (Employed + Self-employed + Looking for work)Leaves out people studying further or not looking, as international labour statistics do
Pushed out of their field

Working elsewhere because there were no jobs in their field, and would go back if they could.

Working outside their field + chose "couldn't find work in my field" + would take a field job (Yes or Maybe) ÷ everyone workingLatest answered check-in (B2, B5, B6)
Found their calling

Chose a different path and happy to stay there.

Working outside their field + chose "passion or interest" + would not return (No) ÷ everyone workingLatest answered check-in (B2, B5, B6)
Time to first related job

How long it usually takes to get a job linked to the course.

The middle answer (median band) to C1; "Not yet" is shown as its own shareEveryone who answered C1
Would choose again

Satisfaction with the choice of course.

Yes ÷ everyone who answered C3Skipped answers are left out
Cost vs jobs

Unemployment 1 year after graduating, compared across what graduates paid in total fees for their course.

For each fee range from A6 (grouped as under KSh 250,000; KSh 250,000 to 999,999; KSh 1 million or more): unemployed ÷ everyone working or looking, at the 1-year check-inOptional question; "Not sure" is left out; a range is shown only with 10 or more answers
Typical monthly income

The middle income band: half of working graduates earn less, half earn more.

The band where the running total of B8 answers passes 50%, among working graduates, at the latest check-in each one answered; "Prefer not to say" is left outShown only for groups of 10 or more; income is never shown for one person
Placed, not chosen

Courses people end up in without choosing them.

"KUCCPS placed me" ÷ everyone who answered C2Skipped answers are left out
Trajectory

Whether things get better or worse with time.

The share in their field or a related job, and unemployment, at 1, 3 and 5+ years, and the change in percentage points between themResponse counts are shown for each check-in, since different people answer each one

Outcomes over time

The stacked bars and outcome mix show graduates who are working or looking for work at each check-in, split into five groups: in what they studied, same industry, different field, working but not sure how it relates, and unemployed. Working people who answered "Not sure" are shown as their own group and are never counted as aligned.

Share in a groupPeople in the group ÷ everyone working or looking for work at that check-in

Each check-in shows its own response count, because older graduates answer more check-ins than recent ones.

Underemployment

Underemployment measures people who work but below their potential. It is calculated among those working (employed or self-employed) at their latest answered check-in, and is separate from unemployment.

MeasureFormula
Job doesn't need their qualification

Skills underemployment

B4a = Yes ÷ everyone working
Role below their level

Level underemployment, e.g. a Master's holder working as an assistant

B4d = Below my level ÷ everyone working
Too few hours

Time underemployment

B4c = "I want more hours or work, and could take it on" ÷ everyone working
Underemployed (overall)

In at least one of the three groups above; each person is counted once

Any of B4a Yes, B4d Below, or wanting more hours in B4c ÷ everyone working
Stuck

Below their level with no promotion

B4d = Below my level and B4e = No ÷ everyone working
Qualified above their roleHighest qualification (A5) above the course level (A2) and B4d = Below my level

Shows people who studied further but whose role hasn't caught up. Reported alongside the level measure.

Course verdicts

Verdicts describe how well a course's graduates are absorbed by the job market. They are based on the 3-year check-in, which gives slow-starting careers time to settle. Thresholds are reviewed once a year and any change is dated on this page.

VerdictRule
Strong job matchAt the 3-year check-in: 60% or more work in their field or a related job and unemployment is 15% or less.
MixedAt the 3-year check-in: anything between Strong and Weak.
Weak job matchAt the 3-year check-in: under 30% work in their field or a related job, or unemployment is 40% or more.
Slow starterWeak at the 1-year check-in, but Mixed or Strong by the 3-year check-in. Common where graduates first need internships or licensing, such as medicine or law.
Early signalFewer than 30 responses at 3 years but at least 30 at 1 year. The same rules are applied to the 1-year check-in and labelled as early, since slow starters can look weak at this point.
Early results10 to 29 answers at both the 1-year and 3-year check-ins. Figures are shown with a likely range, but no verdict is given yet.
Too few answersFewer than 10 answers. Figures are not shown.

Institution rankings

Institutions are ranked by graduate unemployment at the selected check-in. Only institutions with at least 30 responses are ranked; the rest are listed last as "Not enough data".

Compared with the same courses nationallyInstitution's unemployment − expected unemployment

Expected unemployment is the national rate for each course the institution's respondents studied, averaged using the institution's own mix of responses. A result within 2 points either way is shown as "Similar". This separates an institution that mostly offers courses with few jobs from one whose graduates do worse than others taking the same courses.

Public vs private

Every response is grouped by the institution's ownership: public (owned by government) or private (owned by a church, company or individual). Answers come from A1b, filled in from the institution list where possible.

How the comparison worksSame indicators, calculated separately for public and private

The comparison follows the level and field filters but ignores the public or private filter, so both sides always show. The summary line appears only when each side has at least 30 responses. Public and private institutions offer different courses, so a same-courses comparison, like the one used for institutions, will be added.

Senior school pathways

Each course is linked to the CBC senior school pathway that usually leads to it: STEM, Social Sciences, or Arts and Sports Science. Pathway figures pool the responses of all courses linked to that pathway. The full course-to-pathway list will be published with the course list.

Roles board

Graduates who are working can describe their job or main activity in up to 5 words (B7). Each week the team exports these answers, groups similar ones (for example "poultry farming" and "chicken farmer" become Poultry farmer) and publishes roles given by at least 5 graduates of the same course. Employer names and anything that could identify a person are removed. The board shows roles from graduates working outside their field or in the same industry but a different role.