Keywords
Dear Editor,
Linguistic validity is vital when the major instrument is not in native language of the country. When an instrument is to make (develop, adopt, or adapt), some different evidence of validity and reliability should be considered. To make the instrument for the undergraduate medical education accreditation, it should be made including the set of standards. WFME standards were determined to develop standards for medical schools in Iran. These standards were contextualized and translated from English to Persian language for easy understanding of the readers. However, the content validity and the germane indices were observed in Gandomkar et al. (1), linguistic validity was not considered to check the language clarity and the same understanding between and among involved in internal and external assessment processes. In this regard, the collection of evidence for response process validity appears to be critical (2).
As a flash back, when the first round of the accreditation was done, the accreditation committee were asked to reflect on the findings of the first round. In reflection sessions, one of the agendas was reading critically the WFME certificate. One of the results of that critical reading was the necessity of post-accreditation monitoring. Detailed literature search was conducted to find an acceptable and feasible protocol for post-accreditation monitoring. WFME does not provide guidelines for post-accreditation monitoring and each country or agent is free to choose its own methodology according to the local context. For the accreditation system in Iran, post-accreditation monitoring protocol was designed in three phases. The protocol was approved by the National Commission of Evaluation and Accreditation (NCEA) for implementation. During the phases, according to the mixed-method approach of the protocol, agreement of the external and internal assessors regarding their decision on the available status in a medicine school in accordance with the standard was evaluated. In some cases, inter-rater agreement between and among internal and external assessors was convincing; however, in some cases, it was critical. To find the reasons behind, hermeneutic phenomenology of Gadamer as he depicted in Truth and Method (3) was applied to reconstruct the scenario.
To investigate the reasons behind the occurrence of this error, fifteen focused group sessions were conducted with the participation of NCEA members. At the end of these sessions, consideration lack for linguistic validity in the development and translation process was identified as the fundamental cause of this error. To promote the linguistic validity and then the instrument refinement in particular and the accreditation in general, post-accreditation monitoring committee proposed some suggestions as follows:
(1) The number of standards should be reduced as much as possible.
• Logic: To decrease the number of external assessors and to select more qualified ones.
(2) All standards and sub-standards should be started with the focal term.
• Logic: Some standards are too long, and the focal term cannot be recognized easily by the assessor. Theme and rheme analysis of the standards should be conducted.
(3) The wording minimum and maximum of each standard and sub-standard should be set.
• Logic: To raise the standards understanding
(4) For each area, vital terminologies should be defined based on context.
• Logic: To control the standard bias understanding
(5) Key questions should be written for standards and sub-standards for the coherent interpretation by most involved.
• Logic: To direct the assessment process
(6) For the gathering of the best evidence for response process validity, it is recommended to utilize two methods—cognitive interviewing or cognitive pre-testing—during the development of assessment tools.
• Logic: Cognitive interviewing helps us assess the extent to which assessors' understanding of the questions and phrases in the questionnaires aligns with the understanding of the tool developers, so that in case of any ambiguity, steps can be taken to address it.
To sum, to make an instrument which its language source is different from the native language of the users, in developing process, linguistic validity is paramount of importance. If linguistic validity was not considered appropriately, the coherent understanding of the standards between and among the internal and external assessors is not guaranteed.
Acknowledgements
References
-
1.
Gandomkar R, Changiz T, Omid A, Alizadeh M, Khazaei M, Heidarzadah A, et al. Developing and validating a national set of standards for undergraduate medical education using the WFME framework: the experience of an accreditation system in Iran. BMC Med Educ. 2023;23(1):379. [PubMed ID: 37226139]. [PubMed Central ID: PMC10210375]. https://doi.org/10.1186/s12909-023-04343-9.
-
2.
Willis GB, Artino AJ. What Do Our Respondents Think We're Asking? Using Cognitive Interviewing to Improve Medical Education Surveys. J Grad Med Educ. 2013;5(3):353-6. [PubMed ID: 24404294]. [PubMed Central ID: PMC3771159]. https://doi.org/10.4300/JGME-D-13-00154.1.
-
3.
Gadamer HG. Weinsheimer J, Marshall DG, translators. Truth and Method. 2nd ed. London: Continuum Publishing Group; 1975.