This is the third of my articles written for Intelligent Instructor considering ADI training. In my first article I looked at the history and in the second I considered whether the qualification rates and/or ORDIT could help me answer the question ‘Is ADI Training fit for purpose?’.

In this article I consider typical training approaches in my quest for an answer.

Three tests or a three-part examination?

The ADI qualifying exam tests a candidate’s knowledge of the theory of driving and training, driving skill and teaching skills. It does this in three separate tests, all of which must be passed in order to gain the qualification. OK – you knew that already!  But given that there is only one qualification, it puzzles me why many training providers still treat each of the tests as seemingly unrelated entities.


A typical approach to Part-One is to provide the ADI Student with a DVD or web link to the Part-One questions and then pretty much leave them alone to pass the test (checking to see how they are getting on occasionally).  Some trainers do this in parallel with Part-Two training and allocate some of the in-car time for the discussion of theory material. More progressive companies/trainers are perhaps now offering webinars covering Part-One topics.

By its very nature, Part-One theory learning lends itself to self-study. This is because the learning needs, styles and time available will differ greatly from person-to-person. In the past I have seen companies offering classroom study for Part-One but I’ve never seen it done effectively – I estimate that a ‘good’ Part-One classroom programme would take (at least) two or three weeks full-time study plus additional home-learning.

From the above you can see that some approaches are more ‘hands-on’ than others, but my experience suggests that however Part-One is delivered the content is ‘learned in isolation’ – that is there is no mechanism to help the students to engage more deeply with the material or to relate it to the later parts of the exam or the instructors job. Because of this I believe that many students feel that they are simply ‘learning for the sake of learning’ – they read the questions and learn the answers by rote from the brief explanations given with the questions. It’s possible to pass Part-One in this way but, if you read a few papers about the psychology of learning and recall, you will realise that much of the learning will be forgotten as soon as the exam is passed and definitely before Parts Two and Three are taken.

My rather pessimistic view above suggests that although the Part-One exam might be fit for purpose in terms of the information it tests it might only offer a snapshot of knowledge and does little to explore understanding. If the preparation for a multiple-choice exam is simply about learning the answers, those answers are soon forgotten and the value of the test is limited.

On a more general note a little research into academic studies will give mixed results about the effectiveness of multiple-choice as an examination method, however, I think it has a valuable place in the

ADI qualification – or at least it could have if trainers (and many students) approached the preparation differently.


Part-Two is fairly straightforward – demonstrating driving skill and a knowledge of rules and regulations.

Part-Two training has not changed much over the years and more-often than not is delivered one-to-one in the car. It’s difficult to tell how effective this approach is because the pass-rate statistics vary wildly in different areas ranging from (2021 figures) 87.5% in Kendal to 30% in Chertsey. Different areas have different numbers of test and there is no data about pass rates for individual trainers.

Experience suggests that some trainers have a very high pass rate, perhaps 90% plus first time – but again there is no reliable way to verify this ‘gut feel’ figure, nor is there any data on the average number of training hours taken in order to pass.

Is Part-Two training fit for purpose? It’s probably comes nearer to ‘fitness’ than the other two parts of the exam but I’m absolutely sure that it could be done more efficiently.


Why has Part-Three always been touted as the most difficult part of the exam?  It should be the easiest! I base my assertion on the fact that most people who choose to become ADIs are by their very nature ‘people-people’ who have good/reasonable communication skills or a clear potential to develop those skills.

If test results are a good measure of whether training is fit for purpose  things seem to be getting better. From a low pass rate of 26.9% in 2007/8 the latest figures stand at 38%. However, as always we can’t take raw statics as a measure of success. In 2007/8 there were a staggering 11287 Part-Three tests taken as opposed to just 2055 in 2020/21. I’m not a statistician but I’m sure that there is a deeper story in the numbers. But numbers apart, is 38% at Part-Three an acceptable pass rate given that those who ‘make it that far’ through the exam process are probably the most diligent students?

Part-Three pass rates have improved since the new exam was introduced. Is this because there is an element of cheating? (Rehearsed lessons) Is it because trainers are ‘getting their act together’ and preparing students better? Or perhaps it’s the fact that more students are taking up Trainee Licences and getting hands-on practise.

No conclusion!

Once again, as with my previous article, I’ve failed to answer my own question about whether or not ADI training is fit for purpose and seem to be raising more questions than answers.

I have an opinion, but no hard data to back it up – and so I suspect that my ‘fit for purpose?’ and related questions will remain unanswered (or my answers will be unproven). But, in my next article I’ll offer my own ideas about how things might be tweaked to make training more effective and efficient.

John Farlam

John Farlam, Founder and Training Director, DRIVE.

To learn more about DRIVE and how you can grow your career with them, please see:

You can check out John’s previous article by clicking here