It is hard to carefully study whether passing the theory test and/or practical driving test has an effect on crash risk. After all, the candidate failing the licence acquisition test for a particular vehicle is not allowed to enter traffic driving that vehicle independently. Thus, it is impossible to compare crash risk of candidates that passed to candidates that failed the test. The few studies of the effectiveness of both theory and practical driving tests that do exist, do not show a marked relation between driving test performance and crash risk. Moreover the test effect cannot be dissociated from the driver training effect, since candidates will prepare for the test through training.
A theory test assessing knowledge of traffic rules appears to have no effect on crash risk, whereas a hazard perception test does appear to have a positive effect.
In their meta-analysis of ten (mostly older) studies Elvik et al.  did not find any crash risk effect of incorporating a theory test into the licence acquisition test. Simpson et al.  verified whether the introduction of the theory test in the United Kingdom in 1997 resulted in crash decrease for novice drivers. They showed that attitude and behaviour had somewhat improved but that crash risk had not decreased. After introducing a hazard perception test with moving images, crash risk of novice drivers in the United Kingdom did slightly decrease . Australian research showed a 25% decrease in crash involvement for candidates passing a hazard perception test, compared to those not passing this test .
Practical driving test
Although the effect of the practical driving test is hard to study carefully (see above), researchers in the United Kingdom did try to remark on the responsiveness of the practical driving test to safe driving. No marked relation between driving test performance and crash risk emerged.
Unlike the Dutch examiners, British examiners meticulously assess candidates’ behaviour. Every small or significant imperfection in the execution of traffic actions is scored on a scoresheet. Successful candidates who have made minor mistakes and successful candidates who have not made any mistakes at all may be compared for crash risk. Differences in crash risk for these two groups would indicate how responsive to safe driving the practical driving test is. Baughan en Sexton  did, however, not find any difference in crash involvement between the two groups in the first six months after licence acquisition. Baughan et al.  did report that successful candidates who made a lot of minor mistakes during the practical driving test test were slightly more often involved in crashes than those who made few minor mistakes, but explain this larger involvement by the observation that the first group drives more and is therefore more exposed to traffic risks.