We read with great interest a recent article about the translation of the Mastocytosis Quality of Life Questionnaire.1 The main aim was transcultural adaptation of this instrument to achieve equivalence with the German version. The article, in addition to being relevant, has prompted us to discuss some aspects of the concept of equivalence to raise awareness of the complexity and different levels of adaptation to be considered before undertaking the statistical analysis corresponding to different psychometric studies.
The equivalence in questionnaires, as well as in any other psychometric test, is essential for measurement of outcomes in health sciences, as this is closely tied to their validity,2 particularly in transcultural research. However, many studies only report equivalence through multigroup confirmatory factor analysis (MCFA), a statistical technique that consists of progressively restricting parameters to test whether the internal structure of an instrument is equivalent between demographically or culturally diverse groups.3 But this procedure, in itself, is insufficient to guarantee this property. There are different levels or sources of equivalence4,5: conceptual equivalence, item equivalence, semantic equivalence, operational equivalence, outcome equivalence, and functional equivalence; only the latter of these corresponds to MCFA.
Conceptual equivalence refers to the importance, representation, and meaning attributed to a certain construct in both groups or cultures.4 This first level of equivalence requires exploring and analyzing the way in which each group conceptualizes, defines, or evaluates the variable of interest. Thus, translation and statistical analysis of the test is not enough; it is necessary to undertake theoretical and qualitative investigations beforehand,5 and search for instruments already available to compare meaning.
The equivalence of items consists of the importance and acceptability of each of the items in different groups.4 At this level, the aim is to ensure that the items are relevant and not offensive,5 a particularly delicate topic in transcultural studies, as some items may be taboo. Thus, it is necessary to guarantee that the items are relevant, clear, and representative for measuring the construct, as well as retaining coherence with the conceptual framework. This level of equivalence is linked to content validity and input from anthropologists, sociologists, and members of the target populations may be needed.5
Semantic equivalence refers to the fact that the items should mean the same in both groups or cultures.4,5 It is therefore expected that translations and adaptations fit the original meaning of the items. To ensure this, it is important that drafting takes into account dialect, use of technical language, formality, and connotation of a commonly used phrase and even fashionable words in people of the same sex or age.4
Operational equivalence implies that a test can be administered, scored, and interpreted in the same conditions in the target groups in order to provide a fair assessment.3 This type of equivalence could refer to equity and, therefore, impartiality.6
Outcome equivalence means that the psychometric properties of different versions of a test should be the same.4,5 This can be evidenced through statistical methods such as MCFA, in order to confirm that the internal structure of a test is identical in different groups or cultures.
Finally, we now come to functional equivalence, which is defined as the combination of the 5 previous types of equivalence. This is achieved when a test meets its goal, regardless of the target group in which it is administered.
In conclusion, it is important to comply with each of these types of equivalence when developing psychometric tools before performing MCFA, particularly in the case of transcultural studies. Similarly, the importance of qualitative research in the adaptation and validation of tests should be highlighted, as well as semantic analysis of the constructs, because few articles address whether concepts have the same meaning in different cultures or even whether 2 concepts in the same language are interchangeable.
Please cite this article as: Sánchez-Villena AR, de La Fuente-Figuerola V, Ventura-León J. Equivalencia en la medida: la vida antes del análisis factorial confirmatorio multigrupo. Actas Dermosifiliogr. 2021;112:484–485.
