What is psychometric testing

The term 'psychometric' comes from two Greek words and, literally translated, means 'measurement of the mind'. The term 'measurement', as used here, is not meant to imply that mental abilities or qualities can be measured in the same way as physical quantities. Rather, it is meant to convey the fact that psychometrics involves a 'scientific' approach to the assessment of these abilities and qualities. The best way to illustrate this is by an example from the assessment of personality.

The most commonly used method of assessing personality in employment situations is subjective judgement. Thus, when a job candidate is being interviewed, the interviewer tries to make a judgement of what the person is like as a person and how he or she might behave in a variety of different situations. This judgement will of course be influenced by many factors - for example:

• How experienced the interviewer is - not just in interviewing applicants but in interacting with people and making predictions about how they are likely to behave
  
  
• The extent to which the candidate is projecting his/her true self or projecting an image that he/she feels will go down well with the interviewer
  
  
• The range of situations covered by the interview and the extent to which these are relevant to behaviour in a work situation
  
  
• Who precisely the interviewer is comparing the candidate with - with the interviewer him/herself, with the interviewer's friends and associates, with people who work in the same company, with people who have attended for similar interviews in the past?
  
  
• What particular prejudices the interviewer may have - for example, does he/she believe that women are not well suited to jobs involving authority over others or that men are incapable of working in a caring role?
  
  
• What particular internal 'theories of personality' does the interviewer hold dear - for example, does he/she believe that if someone gives a firm handshake, they must be a strong character?
  
  
• What sort of mood and general state of health the interviewer happens to be in that day
  
  
• What sort of mood and general state of health the candidate happens to be in that day

This is not to say that some interviewers may not be able to make very accurate judgements of personality. Probably some are. The problem is that we don't really know who these people are and, even worse, they probably don't even know themselves.

In contrast, the psychometric approach to the assessment of personality tries as far as possible to overcome the sources of error and likely misjudgement mentioned above. There are several different psychometric approaches to the measurement of personality, but they all have the following elements in common.

• They are based on scientific research (at least in so far as it is possible to be scientific in relation to human behaviour)
  
  
• They used a standardised approach to obtaining the relevant information - i.e. every person who takes a particular personality questionnaire will be asked precisely the same set of questions. This differs markedly from the interview situation where each candidate may well be asked very different questions (even when the interview is a 'structured' interview). Effectively this means that the psychometric approach gives each person equal opportunity to show what they are like across the same range of behaviour, attitudes and interests as everyone else taking the questionnaire.
  
  
• They use a standardised approach to judging the information provided by the candidate. Unlike in an interview where a given statement by the candidate might be judged differently by different interviewers (and even judged differently by the same interviewer on different occasions), the psychometric approach evaluates the responses of all candidates according to a fixed set of rules. For each personality questionnaire, there is a scoring key - and the same scoring key is used for each person taking the questionnaire. There can therefore be no argument, at least at this stage in the process, in interpreting particular responses the candidate may have given.
  
  
• They use a standardised and quantified approach to expressing the conclusions reached by the assessment. In the case of most psychometric tests, this is done by comparing the scores obtained by a candidate with the scores obtained by a large sample of people who have taken the test on a previous occasion. Thus, in contrast to the interviewer who might at best be able to say "I think this chap is pretty sociable and should be able to get on well with people", the psychometric questionnaire would conclude that, for example, the candidate has scored "at the 75th percentile on the scale of Sociability". What this means is that 75 percent of people in the comparison sample scored at a lower level than this (were less sociable) and only 25 percent of people scored higher (were more sociable). In this way, the psychometric approach is able to make a quantitative assessment of just how sociable the person is, when compared to other people who have taken the test in the past.
  
  
• The effectiveness of psychometric tests can be evaluated objectively. For example, studies can be conducted to see whether people vary in their responses to a questionnaire on different occasions. If it is found that their responses vary greatly from occasion to occasion, this means that the test may be unreliable and may need to be improved. Studies can also be conducted to see whether the predictions made from the a test or questionnaire are actually born out in practice. For example, the scores on a test of Sales Ability can be correlated mathematically with actual sales figures. If the people who score highly on the test do not turn out to have higher sales figures than people who score low on the test, this suggests that the test is not assessing what it is supposed to assess and should probably be abandoned.

Although most of the above example has been taken from the assessment of personality, these general principles are applicable to all other areas of psychometric testing - testing of ability, of aptitude, of interests, of motivation - and also to other forms of assessment which, although not making use of questionnaires or tests, nevertheless are based on objective and standardised methods of assessing people.

Next