Skip to main content

Table 2 Categorization of translatability review criteria

From: Emerging good practices for Translatability Assessment (TA) of Patient-Reported Outcome (PRO) measures

Categories of translatability review criteria* Definition
1. Culture A word or formulation in the original is culturally loaded in the target context due to societal or cultural conventions (e.g.., eating or clothing habits, religious taboos). The usage of certain words or phrases based on the culture of a given society may be improper in the target language and/or culture. For instance, certain foods are not eaten in target countries and should be replaced in the translations.
For example, starchy foods (e.g. potato, bread) becomes starchy foods (e.g. rice, pasta, chapatti).
2. Language
  2a. Meaning (Semantics) Semantics concerns meanings, which are both denotative, i.e. the literal word (lexis), and connotative, namely the set of cultural and/or subjective associations implied by a word in addition to its literal explicit meaning. This category includes lexical differences. For instance, English has a slightly larger lexicon than French. Therefore, some French words have no direct equivalent in English and would need the use of paraphrases.
Example 1: The literal translation of the word “frustrated” in French (i.e., “frustré(e)”) has a very different connotation than the US English meaning. The word “frustré(e)” is often associated to dissatisfaction. In general, the English word “frustrated” covers several feelings (annoyed, powerless, disappointed, and angry) that are not embedded in one word in many other target languages.
Example 2: The meaning of “depressed” can range from “feeling a little sad” to “clinically depressed” in English depending on the context. However, in some languages, only the latter meaning applies, so developer(s) might consider a more specific word instead of “depressed.”
Example 3: “Maladie” in French is a unique word for which English has several terms, i.e., condition (general word), illness (perception, before you go to the doctor), disease (when you come from the doctor, i.e., with a medical diagnosis)
2b. Use (Pragmatics, Idiomatic expressions) The practicalities of how a language is used in its everyday context may be different between the source and target language. For example, one language may have more social registers than another (there are a number of different forms of addressing a person in Japanese, whereas English may only have one) and the idiosyncrasies of one language (repetitions, focus on particular words, use of particular idiomatic expressions, etc.) may not be found in another.
For instance, “I feel downhearted and blue ” translated by an equivalent of “I feel downhearted and sad” or “I feel downhearted and depressed.”
2c. Syntactic (syntax, grammar and punctuation) Grammatical and syntactical possibilities vary across languages and may impact the identification of conceptually equivalent alternatives in a target language. The structure and grammar of the source and target language diverge.
Example 1: The use of a verbal passive form in the original may not be possible in some target languages where active form is a more natural verbal construction.
Example 2: The placement of the recall period might differ in some target languages. In English, it often goes at the beginning or end of the item, but in other languages it might be grammatically necessary to place it in the middle of the item.
3. Item construction
  3a. Item vague, ambiguous The meaning of the item or words within the item are unclear and can be understood in multiple ways in the source text, leading to potential mistranslation in target languages if the wrong nuance is chosen.
3b. Use of double negative A double negative in an item or in conjunction with a negative response choice makes the response and its interpretation difficult because in some languages, the double negative creates a positive meaning, while in other languages, the double negative merely reinforces the negative concept.
3c. Readability issues The language used in the original is too high a reading level for clarity and might impair the understandability of the original and, therefore, impact the future translations. Readability assessment may be needed if not previously conducted by measure developer.
3d. Redundancy between items Two items may express the same concept or close concepts that would be translated the same way
3e. Lack of coherence with concept The terms used in the item do not seem to adequately convey the meaning of the concept to be measured.
3 f. Lack of coherence of response scale with item The response scale does not fit the phrasing of the item.
  3 g. Two concepts within one item The item may express two different concepts that may confuse the respondents.
  1. *See Brislin 12 guidelines for writing translatable English [31]