Measuring physical attributes like weight and height is reasonably simple. These are all physical and observable traits that cab be assessed easily and objectively. However, traits such as emotions, personality, intelligence, attitude and beliefs are important characteristics to measure but difficult to do so. These are important variable to measures when it comes to hiring and developing people.
We all make intuitive assessment of people consciously or unconsciously. We find some people having a good personality and some bad. And some no personality! Unlike intuition based judgements that are subjective in nature, psychometrics use standardized questionnaires based on a well-researched statistical approach. By using standard models, it is then possible to highlight the differences between individuals or between groups sharing the same or similar characteristics.
Differences between individuals belonging to the same group, e.g. the differences in educational qualification among members of a team. These are call Inter-Individual Differences.
Differences which can be seen in one individual e.g. variations in a subject’s degree of stress depending on changes in situation. These are called Intra-Individual Differences.
Differences between different groups of people e.g. divergent personality traits between those who work in the factory and those who work on the field, supporting clients. These are called Intra-Group Differences.
Today, these psychometric tools are applied not just in recruitment but also in internal assessments. Psychometric assessments become handy in predicting behavior and performance more accurately than ever before. They allow companies to appraise candidates’ personality traits, priorities, stress factors, motivation, reasoning capacity and emotional intelligence.
Psychometric tools are constructed by applying statistical analysis. They are standardized and calibrated against standardized norms. They are then subjected to validity studies. The calibration is necessary to allow users to position each respondent within their reference group in terms of level of education, age, gender, in the same way that some of the traditional intelligence tests are calibrated in terms of age and gender for children.
Standardization – The tests are based on a sample of population that is not just sufficiently large for statistical analysis but is also a true representative of the entire group. Realistically, it is not possible to test every working person in a group / population. By analyzing the traits of a representative sample group, we can apply the results to the entire group.
Reliability – The test must produce consistent results. “Reliability” refers to the way in which a test measures a given dimension (when considering the stability of results or the consistency of the measurement). A test which provides similar scores when a person takes it several times (test-retest) is considered a reliable test.
- Internal consistency that indicates the extent different items measure the same dimension or trait. A statistical measure called Cronbach’s Alpha is regarded as the best method of evaluating internal reliability. Cronbach’s Alpha expresses the degree of correlation as a specific number, which typically varies between 0.0 and 1.0.
- Test-retest reliability indicates the stability of results obtained by the same individuals for each measured trait. Everything DiSC assessment are quite stable over time. Test takers and test administrators should expect only small changes when the test is taken at different times. As the period between administrations increases, however, the divergent results of these attempts will become more noticeable.
Validity – A valid test must measure what it's intended to measure. If a test is supposed to measure a person's orientation (to people for example), then it must clearly demonstrate that it does measure orientation to people, and not something else that's related to it.
Content validity which ascertains how well a test measures characteristic it has been conceived to measure;
Criterion validity which looks at the correlation between a respondent’s performance in a test and his/her performance according to an external criterion;
Theoretical validity where the test results must conform to the theoretical model the instrument is based on.
Some test also provides a score of Social Desirability. It is a bias which results from a respondent’s willingness to adapt his/her responses so that they will be seen favorably by another people / stakeholder. e.g. a person answers all questions that portray him as an assertive person as the respondent believes that his/her manager likes assertive people.
No psychometric tool is perfectly reliable or perfectly valid. Psychological instruments are subject to various sources of error. Reliability and validity are matters of degree on a continuum, rather than a binary reliable/unreliable and valid/invalid on dichotomous scales.
Everything DiSC assessments are quite stable over time. Test takers and test administrators should expect only small changes when the test is taken at different times. As the period between administrations increases, however, the divergent results of these attempts will become more noticeable.