Mechanical comprehension and fault finding tests for technical roles

Mechanical comprehension and fault finding tests are the workhorses of technical selection, from engineering apprentices to maintenance technicians and rail staff. They measure different things, and the research shows where each adds value and where it needs support from other evidence.

What a mechanical comprehension test measures

A mechanical comprehension test presents everyday physical situations, usually as diagrams of levers, pulleys, gears, springs, fluids and loads, and asks the candidate to predict what will happen: which way a gear turns, which beam carries more weight, where a pendulum moves fastest. The questions assume no specialist training. What they sample is the ability to reason about how physical systems behave, which is why mechanical comprehension tests are used for engineering, maintenance, production and skilled trades roles and for apprentice intakes, where candidates often have little direct experience yet.

Cognitive research explains why these items work as they do. Hegarty's review shows that people often solve mechanical problems by mentally simulating the system, working step by step along the causal chain, and that performance on such problems is closely tied to spatial ability rather than verbal ability.^[1] A mechanical comprehension test is therefore partly a test of spatial reasoning applied to physical content, and it pairs naturally with a spatial reasoning measure for roles that involve reading drawings or assembling equipment.

What the evidence says

The best-known instrument in this family, the Bennett Mechanical Comprehension Test, has a long research record. In one concurrent validation study with 193 employees making electro-mechanical components, Muchinsky found it was the strongest single predictor of supervisors' ratings of performance (uncorrected r = .38), with a general mental ability test and a second mechanical test adding no significant prediction beyond it.^[2] The same study also carried a caution: the author argued that conventional mechanical tests may not fully capture how manufacturing work was changing, a reminder to check content against current job demands rather than assume a test written for one era fits the next.^[2]

The wider selection literature puts single studies in context. Schmidt and Hunter's review of 85 years of research established that general reasoning ability predicts performance and, especially, success in training across occupations.^[3] Sackett and colleagues' 2022 re-analysis lowered most validity estimates by correcting for statistical overcorrection, yet found that most procedures that ranked high before still rank high, and emphasised weighing validity against subgroup differences when designing a process.^[4] Mechanical comprehension tests should be read in that light: useful, job-related evidence of reasoning about physical systems, not a stand-alone verdict. For the broader picture on reasoning tests, see Ability tests in recruitment: what the research shows.

What a fault finding test measures

Fault finding (or fault diagnosis) tests ask candidates to trace why a system is not producing the expected output. A typical item shows a network of switches, components or logical rules with inputs and outputs, and the candidate works out which element must be faulty. Unlike mechanical comprehension, the content is usually abstract, so prior knowledge of particular machinery helps little. What is sampled is systematic diagnostic reasoning: forming hypotheses, testing them against the evidence and eliminating possibilities in an efficient order.

The research on real troubleshooting supports measuring this skill, with two qualifications. Morris and Rouse's review found that troubleshooting performance falls as systems become more complex and as time pressure rises, and that explicit guidance on strategy improves performance.^[5] They also concluded that task-related knowledge matters more than general aptitude for on-the-job troubleshooting, although more able people are better at some diagnostic strategies.^[5] A fault finding test is therefore best understood as a measure of the reasoning that underpins diagnosis, which training and experience then build on. That makes it especially informative for trainees and apprentices, and a complement to technical knowledge checks for experienced hires.

Choosing and using the tests

Match the test to the work. Mechanical comprehension suits roles where people install, operate or maintain physical equipment. Fault finding suits roles built around diagnosis: maintenance technicians, electrical and control engineers, signalling and rolling stock staff in rail, and IT support. Many technical roles need both. Our mechanical comprehension test and fault finding test pages describe the formats.

Use job analysis to decide the weight. Ask what proportion of the job involves reasoning about physical systems versus diagnosing faults, and weight the evidence accordingly rather than giving every test equal say.

Interpret in bands, alongside other evidence. Report scores in bands from Very Low to Very High against a relevant comparison group, and combine them with structured interviews, work samples and technical knowledge where those are available. A single test score should never act as a gate on its own.^[3,4]

Monitor group differences. Mechanical content can favour candidates with more hobby or school exposure to tools and machinery. Track outcomes by group, explain the test format in advance and offer practice items so everyone starts from the same understanding of what is asked.^[4]

Mind the clock. Because diagnostic performance degrades under time pressure,^[5] set time limits that measure reasoning rather than panic, unless speed of diagnosis is genuinely critical in the role.