How Theorio compares driving schools
Start with the registration and reporting period
A school page describes a CBR registration code. Separate codes or branches are not combined because they share a brand, address or Google listing. Results are historical: they do not establish the current teaching area, pickup service or appointment address. Each result names its vehicle category and period. Car figures cover category B, motorcycle figures cover road riding and scooter figures cover AM2; vehicle-control and microcar tests are outside these groups.
First tests, retests and the national comparison
The ordinary first-test pass rate is passes divided by passes plus failures. Retests are counted separately. The national comparison sums the CBR counts for all schools in the same category and period; it is not the average of school percentages. A small number of tests produces a volatile percentage. Student selection and when a school books a test influence the outcome, so a difference is not evidence of a causal effect or a prediction of your result.
Comparing the same test centres
We calculate a separate reference using other CBR registrations at the same centres. We subtract this school and known related registration codes from each centre’s first-test totals, then weight the remaining pass rates by the school’s own distribution of first tests. This is not a pooled national rate or a ranking score. It requires at least 30 first tests at the school, at least 100 other first tests per included centre and coverage of at least 90% of the school’s first tests. The paragraph reports covered tests, total tests, centres and reference-group size. These are editorial minimums, not a significance test. Category, period and centres are aligned; candidate selection and teaching quality cannot be inferred.
Manual and automatic: read the sample size
Where both car transmissions have consistent first-test data, we show passes, pass rates and each group’s share of all first car tests. Sensitivity is 100 divided by the group’s test count: the change in percentage points if one existing outcome were different, with the same total number of tests. It is not the effect of adding a test or a forecast. Any tests outside the two groups are stated separately. Small samples warrant more caution, and past test figures do not confirm current lesson availability.
The comparison score and ranking
The score combines a CBR component (70%) and a Google component (30%) for the same exam centre. The CBR component uses the lower bound of a 95% Wilson interval, not the ordinary pass rate. It maps 30–70% linearly to 0–100. Google ratings are adjusted using a prior of 20 reviews at 4.05, then 3.2–4.9 maps to 0–100. Each component is capped at the ends. These are editorial scale choices. A score needs at least 10 first tests and 5 Google reviews. A rank needs 5 comparable schools; ties in unrounded scores share a position. Rounded displayed scores can look equal. Prices do not affect the score. Shared Google identities that cannot be attributed independently to a branch are excluded. The score is not a CBR endorsement.
History is not a series of independent years
CBR releases can cover overlapping rolling periods. The chart retains their exact dates and does not turn missing values into zero tests. The explanatory comparison selects non-overlapping twelve-month periods with the same start and end months. If no earlier comparable period exists, no trend paragraph is generated. Gaps remain visible; an older source is never labelled as a fresh check.
Facts from a school’s own website
We publish concise structured facts with a source URL and the time the page was observed. An archived page is not current confirmation. A school or branch must be tied to its CBR registration with explicit identity evidence; a matching brand or domain alone is insufficient. Marketing paragraphs are not copied into the school description. A missing fact means it was not verified, not that the service is unavailable.
Prices and package conditions
Offers remain attached to the licence category stated by the source. An unknown category is not assigned by guessing. Hourly rates are only calculated when the duration is explicitly stated and the price buys lesson time alone. Bundles containing exams or other services are not converted into lesson-only hourly rates. English summaries retain source links and explicitly refer readers to original offer conditions. A historic price is not a current quotation; check duration, contents, cancellation and refund terms directly with the school.
Questions for your trial lesson
Ask whether you will keep the same instructor, how often lessons can take place, how your progress is reviewed and how test readiness is decided. Check the lesson duration, package contents and the total expected costs. Discuss cancellation, stopping lessons and refunds before paying.
Sources, corrections and limits
CBR provides the exam results. Contact and school facts can come from the school website, a linked Google listing or a separately identified public source. Sources and observation dates are shown where relevant. Nearby school suggestions use the same category and period and straight-line distances between town centres; they are not recommendations or distances from a teaching address. To report an incorrect match, include the CBR code, the disputed field and a supporting first-party source.