We can weigh ourselves on a scale, or measure our height with a meter, or assess our body temperature with a thermometer. The data that we obtain should be objective and reliable, as well as referring specifically to what we want to measure (weight, height or temperature). But what if they also reflect other things such as volume or colour or are influenced by atmospheric pressure or humidity? Our results would not be totally valid, since we would not be looking only at the characteristics we wanted to evaluate.
In psychology, a science whose object of study is not directly observable and in which different constructs are analyzed, validity is something that must be carefully considered in order to ensure that we are evaluating what we should be evaluating. It is something that is essential, for example, to evaluate the mental state of a subject or to assess the effectiveness of a treatment. And it should be borne in mind that, depending on what is being analysed, we may find different types of validity . In this article we will review what they are.
What is validity?
Before seeing what the different types of validity are, it is advisable to do a small review of what this term refers to.
Validity is understood to be the property or capacity of a test or other measuring instrument to adequately measure that for which said instrument has been generated , independently of the theory or model of reality from which it has been elaborated. It is linked to what is measured and how it is done, assessing whether the measurement is carried out correctly. That is: that the measurement data correspond to the real data.
Validity can be calculated on the basis of the coefficient of validity, based on the degree of correlation between the variable measured and the one studied.
The different types of validity
Validity is a fundamental property when carrying out measurements of any kind. As we have commented in the introduction, in sciences such as psychology it is essential to take this aspect into account in order to generate valid measuring instruments to evaluate the state of the people analysed. However, validity can be considered from different perspectives, and different types of validity can be found that focus on different aspects.
1. Construction validity
This type of validity refers to the accuracy with which the measuring instrument measures what it is intended to measure. In other words, it assesses the extent to which the responses or results of the assessment method used have a specific meaning, with a relationship between the observed and the construct of interest .
2. Content validity
It is the degree to which a measuring instrument contains items that are representative of the construct or content that it is intended to evaluate. It is valued that aspects of interest that represent the attribute to be evaluated are included in the items that form part of the measurement. Two main types of validity can be assessed.
3. Apparent validity
Although it is not really a type of validity, it refers to the degree to which a test appears to value a certain attribute. In other words, it is the appearance of validity that an instrument can give to anyone who looks at it, without any kind of analysis. It has no real significance.
4. Logical validity
This is the type of validity used to generate an instrument and measurement items, according to the representativeness of the analyzed in the valued content .
5. Criteria validity
It refers to the degree to which a test correlates with scales and external variables , and can relate the results of the measurement to a specific criterion. It also allows for the establishment of predictions.
6. Predictive validity
Criterion validity type that allows to establish predictions with respect to behaviour , from the comparison between the values of the instrument and the criterion. Generally, a time elapses between the moment of measurement and that of the criterion used.
7. Concurrent validity
Both the measurement and the verification of the criterion are carried out at the same time, allowing both elements to be related and the current state of the subject to be assessed.
8. Retrospective validity
An unusual type of validity in which the item or method of evaluation assesses the existence of a certain value or feature in the past. The criterion is taken before the test measurement .
9. Convergent validity
This type of validity refers to the validity obtained from the relationship of two measuring instruments. Convergent validity indicates the existence of a relationship between two tests that assess the same thing , i.e. it indicates the existence of an interrelation or correspondence between the two measuring instruments.
10. Discriminant or divergent validity
Divergent validity is the other side of the coin of convergent validity. We are talking here about the degree to which two tests or instruments differ, reflecting that two tests are associated with different constructs or elements. That is, it reflects that two instruments referring to two constructs that should be different have different results .
- Antequera, J. and Hernángomez, L. (2012). Experimental psychology. Manual CEDE de Preparación PIR, 09. CEDE: Madrid
- Prieto, G.; Delgado, A.R. (2010). Reliability and validity. Papers of the psychologist, 31 (1): 67-74.