Translating best practice into real practice: Methods, results and lessons from a project to translate an English sexual health survey into four Asian languages.

BACKGROUND: Migrants are underrepresented in population health surveys. Offering translated survey instruments has been shown to increase migrant representation. While 'team translation' represents current best practice, there are relatively few published examples describing how it has been implemented. The purpose of this paper is to document the process, results and lessons from a project to translate an English-language sexual health and blood-borne virus survey into Khmer, Karen, Vietnamese and Traditional Chinese.
METHODS: The approach to translation was based on the TRAPD (Translation, Review, Adjudication, Pretesting, and Documentation) model. The English-language survey was sent to two accredited, independent translators. At least one bilingual person was chosen to review and compare the translations and preferred translations were selected through consensus. Agreed translations were pretested with small samples of individuals fluent in the survey language and further revisions made.
RESULTS: Of the 51 survey questions, only nine resulted in identical independent translations in at least one language. Material differences between the translations related to: (1) the translation of technical terms and medical terminology (e.g. HIV); (2) variations in dialect; and (3) differences in cultural understandings of survey concepts (e.g. committed relationships).
CONCLUSION: Survey translation is time-consuming and costly and, as a result, deviations from TRAPD 'best practice' occurred. It is not possible to determine whether closer adherence to TRAPD 'best practice' would have improved the quality of the resulting translations. However, our study does demonstrate that even adaptations of the TRAPD method can identify issues that may not have been apparent had non-team-based or single-round translation approaches been adopted. Given the dearth of clear empirical evidence about the most accurate and feasible method of undertaking translations, we encourage future researchers to follow our example of making translation data publicly available to enhance transparency and enable critical appraisal.

Introduction

In 2019, international migrants living in Oceania, North America and Europe accounted for 21 percent, 16 percent and 11 percent of the total population in those regions respectively [1]. Yet, migrants are frequently under-represented in population-level health studies [2-5]. Moradi and colleagues regard the “[s]ystematic under-representation of migrants in epidemiological studies and surveys [as] a serious methodological issue introducing bias and causing lack of generalizability of the results” [6]. Consequently, our ability to use the available evidence to accurately identify priority areas and effectively design policies and programs is compromised [5, 7–9].

Language is one barrier to migrant participation in research, and offering translated survey instruments has been shown to increase migrant representation [6, 9]. However, survey translation is not a straightforward process. As Curtarelli and van Houten note, a good translation must “on the one hand, take into consideration the different social realities, cultural norms, and respondent needs … and, on the other hand, respect the questionnaire design and retain measurement properties” [10].

There are several approaches to translation. Forward-only translation (also known as ‘direct’ or ‘one-for-one’ translation) involves a single individual translating an instrument from one language (the source language) into a second language (the target language) [11]. While forward-only translation has the advantage of saving time and costs, it is considered problematic because it “involves a total dependence on the [single] translator’s skill and knowledge, and often results in low validity and reliability” [12]. Forward-backward translation (also known simply as ‘back translation’) represents an attempt to overcome the risks inherent in relying on a single individual. In forward-backward translation, a second individual translates the target language instrument back into the source language; the original source language instrument and the back-translated source language instrument are then compared, and any discrepancies serve as indications of the need for further refinements of the translation [13]. However, a criticism of forward-backward translation is that it has the potential to focus too narrowly on the task of literal translation at the expense of ensuring that the translation captures the intended meaning of the survey item in a way that is clear and suitable for the intended audience [14]. For example, Behr cites an example in which ‘care services’ was forward translated into German as pflegedienste and back translated as ‘care services’ suggesting no error, when in fact the translated term “did not fit the questionnaire context since it is only used in the context of the ill and/or the elderly and is thus not fitting to general child care services” [15]. Ozolins and colleagues have reported how some forward translators choose literal translations (despite their misgivings as to whether it actually captures the intended meaning) because they do not want their translation to be flagged as an ‘error’ by the back translator [16].

Consequently, there has been a growing call for more nuanced and layered approaches to instrument translation in which accredited translators, other people who speak both the source and target languages, survey researchers, and subject matter experts work together to produce translated surveys which: (1) capture the intended meaning of the source instrument; (2) reflect the cultural and contextual specificities of the target population; and (3) will facilitate meaningful comparisons of data across populations [13, 17–21]. Indeed, the Guidelines for Best Practice in Cross-Cultural Surveys recommend “a team translation approach for survey instrument production” noting that “[o]ther approaches, such as back translation, although recommended in the past, do not comply with the latest translation research” [22]. ‘Team translation’ (also known as ‘committee translation’) is considered preferable to other approaches on the basis that it enables people with complementary knowledge and expertise to work together to arrive at the best translation to ensure that survey items convey what they were intended to convey to the target audience [21, 22].

While team translation can assume a variety of forms, the approach known as TRAPD (Translation, Review, Adjudication, Pretesting, and Documentation) is the version endorsed in the Guidelines for Best Practice in Cross-Cultural Surveys [22]. Under the TRAPD model, two independent translations are produced and are then compared (item-by-item) by bilingual reviewers who possess study design and subject-matter knowledge, and work with an adjudicator to identify the ‘best’ translation for pretesting; each step is documented for transparency [21, 22].

In this article, we apply the TRAPD model to translate an English-language sexual health and blood-borne virus survey into four languages for migrants living in Australia. The aim of the study is to:

document how TRAPD can be applied in practice, including any challenges in its application;

provide examples of issues identified through the processes of team-based ‘review and adjudication’ and pretesting, which are key features of the TRAPD model;

offer guidance to future researchers who seek to use the TRAPD method; and

provide recommendations for further research priorities on the subject of survey translation.

The study makes an important contribution to the literature since there are relatively few published examples describing how TRAPD has been implemented in the context of survey research, despite it being the model endorsed in the Guidelines for Best Practice in Cross-Cultural Surveys [22]. Much of the available TRAPD literature relates to translations carried out as part of large and relatively well-resourced surveys, such as the European Social Survey [23-28]. However, as Sha and Lai have argued “[i]t is important to identify a viable translation process that can be adapted and tailored to the varying level of expertise and resources available” [29].

Methods

Ethics approval for this study was obtained from the Curtin University Human Research Ethics Committee HRE2019-0395 and participants provided written consent. Data were analysed anonymously. An English-language self-administered paper survey was developed and pretested with South-East Asian, North-East Asian and Sub-Saharan African migrants living in Australia, as described elsewhere [30]. Due to budgetary constraints, the English-language survey could only be translated into a limited number of languages. As this was a feasibility study, four languages were chosen to gain insight into the complexities of multi-language survey development and administration. The selection of languages was informed by: (1) unpublished epidemiological data showing the incidence of sexually transmissible infections (STIs) and blood-borne viruses (BBVs), by country of birth; (2) estimated resident population data, by country of birth; (3) English-language proficiency data, by country of birth; and (4) input from community-based organisations working with migrant communities to improve STI and BBV outcomes. The resulting languages for translation (hereafter referred to as target languages) were Traditional Chinese, Karen, Khmer and Vietnamese.

The approach to translation was based on the TRAPD team translation model depicted in Fig 1, and described in the Guidelines for Best Practice in Cross-Cultural Surveys [22].

Fig 1

Illustration of the TRAPD model adapted from the Guidelines for Best Practice in Cross-Cultural Surveys [22].

Translation

The English-language survey was sent to two accredited translation companies for independent translation. The companies were recommended by project partner organisations with experience working with the target populations. Each company was provided with a detailed brief which included contextual information about the survey, and general principles for translation based on Guidelines for Best Practice in Cross-Cultural Surveys (see S1 Appendix for example) [22]. Translators were asked to enter item-by-item translations into an annotated table comprising over 300 rows of questions, individual response options and instructions. The annotations provided extra information to assist translators to convey the intended meaning of the survey items. Some example annotations are reproduced in Table 1.

Table 1

Example of information provided to translators.

Item	Annotation
Is there non-traditional medication available for people living with HIV so they can live a normal life? (Tick one)	The phrase “normal life” in this question is intended to indicate that the person can function like other people and do everyday things without being impeded by their HIV status.
	The phrase “non-traditional medication” is intended to refer to pharmaceuticals, as opposed to herbal or alternative medicines.
Can you get hepatitis B from swallowing food or water containing the faeces (poo) of an infected person? (Tick one)	There is no need to translate the informal ‘poo’ if people are likely to understand the translated term for ‘faeces’.
If a close friend in Australia told you that they were going to get tested for STIs and BBVs, how would you feel? (Tick any that apply)	Here we are referring to a platonic (non-sexual) friendship
Since January 2018, have you had sex with a sex worker while overseas? (Tick one)	Please chose a neutral, non-stigmatising translation for the term ‘sex worker’.
How do you identify? (Tick all that apply)	N/A
• Woman	Here we mean gender as opposed to biological sex
• Man	Here we mean gender as opposed to biological sex
• Non-binary person	Here we mean people who don’t identify as either a man or woman
• Other (please specify):	N/A
• Transgender	N/A
• Cisgender	Here we mean people who identify with the gender they were assigned at birth
• Prefer not to answer	N/A
Which cultural / ethnic group categories do you identify with?	If Zulu and Hmong are not likely to be meaningful in the translated language please delete
For instance, you might identify with:
• One group of people within your country of birth (e.g. Zulu, Hmong)
• Your ancestral heritage (e.g. Indian-Malaysian or Chinese-Vietnamese)
• People from another place you have lived (e.g. British)

Review

After the independent translations were completed, at least one bilingual person was chosen to review and compare the translations; where differences or errors were observed in the translations, the role of the reviewer was to make recommendations about which translation best reflected the intended meaning of the relevant survey item.

An Excel spreadsheet was developed to assist with the review process (see S2 Appendix for Traditional Chinese example). The spreadsheet contained columns for the English survey item and both target language translations to facilitate side-by-side comparison of each item. Reviewers were instructed to use the spreadsheet to: (1) indicate whether any errors were present in either translation using a drop down menu based on the European Social Survey (ESS) Verifier Intervention Categories (S3 Appendix) [31]; (2) make any relevant additional comments about either translation; and (3) indicate a preferred translation for each item, with reasons.

Reviewers were generally from academic and community-sector organisations responsible for overseeing the design and implementation of the broader study and, as such, had knowledge of the survey context. The exception to this was the Karen reviewer who was recruited externally but had had previous experience with other sexual health research. A summary of the reviewers’ background and experience is contained in Table 2.

Table 2

Summary of reviewer characteristics, by language.

Language	Reviewer number	Place of birth	Years of residence in Australia	Field of occupation
Traditional Chinese	1	Mainland China	24	Community sector (health)
	2	Mainland China	23	Academia (health)
	3	Taiwan	26	Health planning
	4	Hong Kong	6	Academia (health) and clinical sexual health
Khmer	1	Cambodia	10	Community sector (health)
Vietnamese	1	Vietnam	14	Community sector (health)
Karen	1	Myanmar	29	Community sector (social work)

The task of reviewing the translations was provided in-kind and, consequently, only one reviewer was recruited for Khmer, Vietnamese and Karen translations. In the case of Traditional Chinese, two reviewers from mainland China (third- and sixth-named authors) independently reviewed the translations and entered data into the spreadsheet. The reviewers recommended that individuals born in Hong Kong and Taiwan (fourth- and ninth-named authors) should also be involved in the review process as it was known that there were some linguistic variations for important terms.

In practice, only the Traditional Chinese reviewers used the ESS Categories. The Karen reviewer simply indicated a preferred translation for each survey item, usually with reasons. In the case of the Khmer and Vietnamese translations, the reviewers indicated that they preferred one translated version (in its entirety) over another. The implications of these variations are considered in the Discussion section below.

Adjudication

As there was only one reviewer for the Khmer, Vietnamese and Karen translations, adjudication took the form of the first-named author assessing the reviewers’ comments and accepting recommendations unless they did not appropriately convey the intended meaning of the source text (examples of where this occurred are set out below). For the Khmer and Vietnamese translations, no adjudication was possible as reviewers’ did not provide reasons for preferring one translation over another; instead, the reviewers’ recommendations were accepted and the preferred translation was sent for pretesting.

Pretesting

Ethics approval for pretesting was obtained (Curtin University Human Research Ethics Committee 2019–0395). The minimum target sample size for the pilot survey (in any language) was 1,116 respondents equally divided between the three regions of birth (Sub-Saharan Africa, North East Asia and South-East Asia) to detect regional differences at a significance level of 5% and 90% power.

Given the limited availability of resources for pretesting, the exploratory nature of the feasibility study, and the fact that multiple quality checking methods were built into the study design (e.g. two independent translations, review and adjudication), a pragmatic decision was made to only pretest translations on small samples. Pretest participants were recruited using convenience sampling by two members of the research team who had experience working with these migrant communities. The size of the samples varied by community—Chinese (n = 3), Vietnamese (n = 20), Khmer (n = 4) and Karen (n = 3). The larger sample size for Vietnamese pretesting was opportunistic in the sense that a group of 20 participants were gathered for another purpose and expressed willingness to provide feedback on the translated instrument; although this resulted in the Vietnamese pretest sample being larger than those representing other language groups, the opportunity to obtain more feedback with minimal additional resources was recognised as an efficient means of obtaining more data to check instrument validity. There was no intention to engage in any statistical comparison of differences in pretest responses between the communities. While all pretest participants were fluent in the survey language being pretested, no other demographic characteristics were recorded.

Pretesting was conducted in groups in which participants were asked to complete the draft translated surveys and then answer the following questions:

Was there anything that you did not understand? If so, what was it and why do you think you had trouble?

Did you find it difficult to answer any questions? If so, which questions and why?

Were there any errors in the survey that you noticed?

Written notes summarising the participants’ responses were prepared by the pretest facilitators and sent to the adjudicator.

Results

S3 Appendix compares the independent translations for each survey item in each language. As summarised in Table 3, of the 51 survey questions (including five Likert statements), only nine resulted in identical independent translations in at least one language.

Table 3

Identical independent translations of substantive English-language survey questions, by language of translation.

Source text (English)	Whether independent translations were identical (Yes/No)
	Khmer	Traditional Chinese	Karen	Vietnamese
Is an HIV test done whenever someone has a blood test in Australia?	No	No	No	No
Can hepatitis C be passed on by sharing injecting equipment like needles and syringes?	Yes	No	No	No
Did you use a condom the MOST RECENT time you had sex?	No	No	No	No
Why did you NOT use a condom the most recent time you had sex?	No	No	No	No
If a close friend in Australia told you that they were going to get tested for STIs and BBVs, how would you feel?	No	No	No	Yes
How old are you?	No	No	Yes	Yes
What is your religion?	No	No	No	Yes
What are the main languages you speak at home?	No	No	No	No
[Statements for Likert-scale responses]
• I felt upset*	No	No	No	Yes
• I felt embarrassed	Yes	No	No	No
• The survey was too long	Yes	No	No	No
• I found it hard to understand some questions / words	Yes	No	No	No

While identical independent translations were rare, many of the differences in translation were not material in the sense that they did not change the intended meaning of the source text. For instance, the English source item–Which of the following best describes you? (Tick one)–was translated into Traditional Chinese as follows:

Translation 1: 以下哪一項陳述最符合你的情況?(勾選一項) Which one of the following best describes you? (Tick one) (*informal ‘you’)

Translation 2: 以下哪項最符合您的情況?(勾選一項) Which of the following best describes you? (Tick one) (*formal ‘You’)

Similarly, the reviewers considered both Traditional Chinese translations of “How old are you?” to be interchangeable, the main difference being tone (colloquial versus formal):

Translation 1: 你幾歲 (How old are you)?

Translation 2: 您的年齡是 (What is your age)?

However, a number of more material differences between translations were detected during the review process, a sample of which is highlighted in Table 4 below. In most instances, one translator’s version was preferred over another; however, there were instances in which the reviewers and adjudicator considered that neither version suitably captured the intended meaning of the item.

Table 4

Examples of material differences between translations detected during the review process, and outcome of adjudication.

Items / key terms	Language	Translation 1	Translation 2	Preferred translation
Chlamydia	Vietnamese	Bệnh hoa liễu	Bệnh lậu (khuẩn Chlamydia)	Bệnh Chlamydia
		English meaning: Venereal disease	English meaning: Gonorrhoea (chlamydia bacteria)
	Traditional Chinese	衣原體	衣原體病	衣原體 (披衣菌) 感染
		English meaning: Chlamydia	English meaning: Chlamydia infection/sickness	Chlamydia in two dialects + term for infection
Gonorrhoea	Vietnamese	Bệnh lậu	Bệnh da liễu	Translation 1
		English meaning: Gonorrhoea	English meaning: Venereal disease
HIV	Traditional Chinese	愛滋病病毒	HIV	Combine—愛滋病病毒 (HIV)
		English meaning: AIDS virus	English meaning: HIV
	Karen		HIV	Translation 1
		English meaning: Human Immunodeficiency Virus (HIV)	English meaning: HIV
Questions about sexual activities	Karen			Translation 1
		English meaning: Questions about sex	English meaning: Questions about sex among men and women
A casual sex partner	Vietnamese	Một người bạn tình không thường xuyên	Bạn tình bình thường (không có mối quan hệ ràng buộc)	Translation 2
		English meaning: An occasional lover / a lover, not regular	English meaning: Casual partner / normal lover (no relationship tie)
	Traditional Chinese	臨時性伴侶	隨意的性伴侶	非固定性伴(没有稳定關係或臨時性伴, 包括一夜情人
		English meaning: Temporary partner	English meaning: sex partner	Non-regular sexual partners (no stable relationships or temporary sexual partners, including one-night lovers)
	Khmer	ដៃគូរួមភេទដែលមិនមានទំនាក់ទំនងប្តូរផ្តាច់ជាមួយ	ដៃគូររួមភេទម្តងម្តាល	Translation 2
		English meaning: Unprotected sexual partner	English meaning: Occasional sexual partners
Someone you are in a committed relationship with (e.g. husband / wife, boyfriend / girlfriend)	Karen			Translation 2
		English meaning: ‘Committed relationship’ translated as ‘willing—in mind- to live together’	English meaning: ‘Committed relationship’
	Khmer	នរណាម្នាក់ដែលអ្នកមានទំនាក់ទំនងប្តូរផ្តាច់ជាមួយ (ឧទាហរណ៍ ប្តី / ប្រពន្ធ មិត្តប្រុស / មិត្តស្រី)	មនុស្សម្នាក់ដែលអ្នកមានទំនាក់ទំនងជាមួយច្បាស់លាស់ (ឧទាហរណ៍ ប្តី/ប្រពន្ធ, មិត្តប្រុស/មិត្តស្រី)	Translation 2
		English meaning: Someone you have a breakup with (for example, spouse / boyfriend / girlfriend)	English meaning: Someone you have a clear relationship with (e.g. spouse, boyfriend / girlfriend)
Did you use a condom the MOST RECENT time you had sex? (Tick one)	Karen			Translation 2
		English meaning: Condom translated as cover of penis	English meaning: Condom translated as cover of penis
How old are you?	Karen			Translation 2
		Dialect of Irrawaddy Region	Dialect of Karen State

Some differences in translation were the products of clear misunderstandings of the original English survey items. For instance, in the case of Khmer, Translator 1 interpreted ‘committed relationship’ to mean the equivalent of ‘someone you have a breakup with’, and ‘casual sexual partner’ was translated to mean ‘unprotected sexual partner’. Neither of these translations reflected the intended meaning of the original English survey item and, in both cases, Translator 2’s versions were preferred during the review process.

Other differences related to variations in dialect. A reviewer noted that one Karen translator used the dialect associated with the people of Karen State while the other used the dialect of the Irrawaddy delta region. These differences related to only a few survey response items and the adjudicator chose to adopt the dialect associated with the people of Karen State based on what was known about the profile of Karen migrants in Australia. In the context of Traditional Chinese, both translations adopted the dialect commonly spoken in mainland China and Hong Kong but the reviewers recommended that, in some cases, other dialects should be incorporated into the survey. For instance the reviewers noted that while the term 衣原體 (chlamydia) was familiar to people living in mainland China and Hong Kong, a different term was used in Taiwan—namely, 披衣菌. Similarly, reviewers noted that hepatitis B was written as 乙型肝炎 in mainland China and Hong Kong but B型肝炎 in Taiwan. Given that participants from both regions were expected to complete the Traditional Chinese version of the survey, the review panel determined that both translations should be included for completeness, e.g. in the case of chlamydia 衣原體(披衣菌).

A third category of differences related to technical terms. Vietnamese Translator 1 translated chlamydia as bệnh hoa liễu (a generic term for venereal diseases), and Translator 2 translated it as bệnh lậu (khuẩn Chlamydia) which literally means “gonorrhoea (Chlamydia bacteria)”. Neither translation was considered suitable, leading the reviewer to suggest a third term (Bệnh Chlamydia) which was adopted. However, there were instances in which reviewers determined that a balance needed to be struck between technical accuracy and comprehensibility for lay audiences. An example of this related to the translation of Human Immunodeficiency Virus (HIV) into Traditional Chinese. In one translation HIV was translated to 愛滋病病毒 (AIDS virus), while in the other it was translated using the more clinically accurate 人類免疫缺陷病毒 (HIV). A reviewer noted that:

HIV is not AIDS virus. However, many people don’t know what 人類免疫缺陷病毒(HIV’s correct [or technical] name in Chinese) is but all know 愛滋病病毒 (translation of AIDS virus). In addition, not all people know HIV if we use the term in English instead of the translation. This translation 愛滋病病毒 is well known and used by the media and the community in their hometown. For the survey’s purpose I think it is better to use 愛滋病病毒.

Through the review and adjudication process, a compromise was reached whereby the more widely understood translation (AIDS virus) was used with the more technically accurate English abbreviation (HIV) in parentheses.

Finally, some differences in translations reflected different cultural understandings. In relation to the English section heading “Questions about sexual activities and relationships”, one Karen translator interpreted “sexual activities” as “sex among men and women”. Notwithstanding the reviewer’s preference for this translation, the adjudicator adopted the alternative translation (which was not limited to heterosexual activities) given that the survey was intended to capture the full spectrum of sexual relationships. Additionally, the Karen reviewer preferred to translate ‘condom’ as ‘the cover of this’ rather than ‘the cover of the penis’ on the basis that it was less direct and offensive. However, after consultation with the reviewer, the adjudicator opted for the latter on the basis that it was less euphemistic and thus more easily understood. Both the adjudicator and the reviewer agreed not to use the Karen translation which equated the concept of “committed relationship” with a term meaning “willing—in mind—to live together” on the basis that it did not capture the meaning of the source item which was intended to include partners who were not co-habiting.

Through pretesting, two further revisions to the Vietnamese translation were proposed and adopted. The first related to the translation for “non-traditional medicine” (thuốc phi truyền thống). Participants expressed the view that thuốc Tây (Western medicine) would be easier to understand. The second revision related to the translation for the question “What is the postcode in which you live?” Participants felt that the Vietnamese term for postcode (mã bưu điện) was unfamiliar and recommended that the English term should be added in parentheses to improve understanding: Mã bưu điện (postcode) nơi bạn sống là gì?

No significant issues were identified when pretesting the Traditional Chinese translation. Similarly, pretesting of the Khmer survey did not yield any substantive changes; the changes were limited to some grammatical amendments to the introductory text and the detection of minor formatting errors.

Two of the Karen pretesters said that they did not have any comments to make. A third Karen-speaker with experience with translations detected some minor errors around formatting (e.g. some lower portions of the script was not visible for some words) and identified some typographical errors. The main substantive change related to the translation of HIV: “It is translated as human immunodeficiency virus, so many people would find it hard to understand. It will be better if the pronunciation of HIV is written in Karen (as it was done for the country names, gonorrhoea, syphilis, chlamydia)”. Additionally, this pretesting participant noted that the character signifying ‘no/not’ had been omitted from one translation which fundamentally changed the intended meaning. All suggested changes were adopted.

Discussion

Whereas much of the existing TRAPD literature reports on the use of the method in the context of large and relatively well-resourced, international surveys, this study documents the experience of applying the TRAPD method to a modestly-resourced survey in one country. The purpose of this study is not to provide evidence that TRAPD is the final word in effective translation methods. As has been noted elsewhere, empirical data on “the contribution of the TRAPD process … to the overall quality of the translated survey instrument” [18] remains limited, and more studies comparing the results of different translation methods are needed. Instead, this paper shows that the TRAPD method has value in identifying errors and exposing nuances in translation, but that it can be difficult to implement in practice, particularly where resources are low and translations into many languages are required. As Knight and colleagues acknowledge, the reality of studying minority populations is that “research context often makes full compliance with … best practice recommendations impossible and … the best one can do is to make incremental approximations of these recommendations” [32].

As was reported in the Methods, deviations from TRAPD ‘best practice’ occurred. Significantly, it was not possible to employ or compensate translation reviewers; instead, the task of reviewing was undertaken in-kind which made it difficult both to recruit multiple reviewers for each language, and to rigidly proscribe the manner in which the translation reviews were undertaken. For instance, while reviewers were instructed to code translation errors using the European Social Survey (ESS) Verifier Intervention Categories, few reviewers used the ESS categories, and some did not document any reasons for preferring one translation over another which made the process of adjudication difficult. Problems around documentation in the TRAPD review process have been noted in other studies, including the European Social Survey where it was noted that “maintaining documentation can be burdensome … The documentation provided by the country teams on the development of ESS translations was at times meagre” [21].

In addition to being time-consuming, the practice of coding translation errors may not be familiar to bilingual reviewers who do not necessarily possess a research background. Many of the distinctions between the categories—for instance, register/wording issues, grammar/syntax issues and minor linguistic defects (which can include grammar)–are subtle or require a high level of linguistic awareness. This suggests that the use of alternative coding protocols should be considered and tested, and efforts should be made to simplify the process of documenting reasons for decision at each stage of the translation process. In an effort to understand the difficulties they experienced in obtaining adequate documentation from translation partners, Harkness, Villar and Edwards surmised “[t]his may be because those involved were not familiar with how and what to document, but it is also likely that the effort involved in manual documentation played a role” [21]. One solution may be for a member of the research team to assume responsibility for documentation by sitting with each panel member (either physically or virtually) while they independently review the translations. The researcher could record the preferred translation for each item and the reviewer’s reasons, using probing questions where necessary; this was the approach taken for the Karen translation in our study. An alternative would be to investigate the development and use of specialised software to manage the documentation process. It is noted that an online, interactive Translation Management Tool was developed for the Survey of Health, Ageing and Retirement in Europe (SHARE) and tested in round 8 of the European Social Survey in 2016, although a number of areas for improvement were identified [21]. Currently, the tool is not publicly available but can be tailored to support specific survey requirements at a cost.

It is not possible to determine whether closer adherence to TRAPD ‘best practice’ in this study would have improved the quality of the resulting translations. A literature review of studies using multi-step, team-based translations of health quality of life questionnaires found that “[a]lthough there is some evidence that different methods (i.e., ‘light/simpler’ vs. ‘heavy/complex’) yield similar results, this has been tested empirically only on a very limited scale” [12]. However, our study does demonstrate that even ‘light/simpler’ adaptations of the TRAPD method can successfully identify issues that may not have been apparent had non-team-based or single-round translation approaches been adopted. For instance:

relying on only one translator to conduct a forwards-only translation without further review would have resulted in important errors being overlooked (e.g. the Khmer translation of ‘casual sexual partner’ as ‘unprotected sexual partner’, and the Karen translation of ‘committed relationship’ as ‘willing in mind to live together’);

relying on forward-backward translation only may not have revealed importance nuances (e.g. the translation of hepatitis B as 乙型肝炎 would likely have been translated back into English as ‘hepatitis B’ suggesting no problem with the translation, when in fact the review process highlighted that it would not be inclusive of/familiar to Taiwanese-Chinese speakers);

not including researchers as partners in the translation process may have resulted in translations which did not reflect the intended meaning of the survey items (e.g. a bilingual reviewer without a sexual health and blood-borne virus background favoured the narrow, heteronormative translation of Karen translation of ‘sexual activities’ which meant ‘sex among men and women’);

pretesting assisted in identifying typographical/formatting errors made when incorporating the reviewed translations into the survey tools (e.g. the omission for the term ‘no/not’ in one item in the Karen survey, and parts of the Karen script not being displayed correctly);

the benefits of assembling a team comprising people with complementary expertise was particularly evident when attempting to translate technical, medical terminology in a way that, on the one hand, is meaningful to the target audience and, on the other hand, guards against perpetuating terms which are stigmatising or give rise to inaccurate assumptions (e.g. the difference between ‘clinical’ and ‘lay’ translations of HIV into Traditional Chinese).

It is, of course, possible that some errors in translation were not detected as a result of the approach we adopted. For this reason, we have taken the rare step of making all of our translation data publicly available (S3 Appendix). Given the dearth of clear empirical evidence about the most accurate and feasible method of undertaking translations, we encourage future researchers to follow our example. Greater transparency around survey translation research processes and results will: (1) allow the accuracy of different approaches to be studied in greater detail; (2) provide practical insight into effective ways of adapting established processes in response to different needs, settings and levels of resourcing; and (3) enable the validity of the survey results to be assessed with reference to the quality of the underlying translations.

Unless a culture of transparent trial-and-error in translation is fostered, negative consequences will follow. First, survey researchers may feel daunted by the complexity (and uncertainty) of translation and may therefore avoid attempts to engage with linguistically diverse populations; this would either result in: (a) underrepresented sections of the population continuing to be overlooked in studies geared towards collecting data to improve health outcomes, or (b) respondents having no option but to complete surveys in a language other than their preferred language, thus increasing the potential for survey items to be misunderstood and data quality to be compromised. Second, survey researchers who have attempted to undertake translations but whose methods have deviated from ‘best practice’ (e.g. due to circumstantial challenges like budget limitations) may hesitate to publish their translation processes and results for fear of being criticised; the consequence of this would be a continued dearth of information about the most effective and efficient means of undertaking survey translations. As a recent editorial in Nature Human Behaviour makes plain, “[s]cience is messy, and the results of research rarely conform fully to plan or expectation. ‘Clean’ narratives are an artefact of inappropriate pressures and the culture they have generated” [33]. It is only through more detailed accounts of the processes and results of survey translation that we can come closer to understanding which methods (and variations of methods) are best suited to produce the most accurate results in specific contexts. Given the increase in the scale of international (pre-pandemic) migration [34], and the growing appetite for multinational, multicultural and multiregional surveys [35], the importance of continuing to build the evidence-base for survey translation is both apparent and urgent.

Conclusions

Survey translation is time-consuming and costly and, as a result, deviations from TRAPD ‘best practice’ occurred. It is not possible to determine whether closer adherence to TRAPD ‘best practice’ would have improved the quality of the resulting translations. However, our study does demonstrate that even adaptations of the TRAPD method can identify issues that may not have been apparent had non-team-based or single-round translation approaches been adopted. Given the dearth of clear empirical evidence about the most accurate and feasible method of undertaking translations, we encourage future researchers to follow our example of making translation data publicly available to enhance transparency and enable critical appraisal.

Brief for Translators.

(DOCX)

Click here for additional data file.

Reviewers’ spreadsheet (Chinese).

(XLSX)

Click here for additional data file.

TRAPD results.

(XLSX)

Click here for additional data file.

2 Jul 2021

PONE-D-21-17384

Translating best practice into real practice: methods, results and lessons from a project to translate an English sexual health survey into four Asian languages

PLOS ONE

Dear Dr. Vujcich,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

==============================

ACADEMIC EDITOR COMMENTS:

Thank you for inviting me to review this manuscript. The manuscript is well written. Actually, it's challenging to translate survey or questionnaire into different languages with different context and culture. I have a few questions and comments for the authors which are listed in the following:

1. The team approach TRAPD for translation is systematic, but no statistical method conducted to assess the equivalence such as content validity test (CVI). Is it possible to add or combine statistical method to increase the validity of the translation?

2. In the pretesting phase, the sample size of Vietnamese was 20, but other migrant communities were 3 or 4. Why was there a great variation in the sample size in different groups?

3. In line 141-142, "While identical independent translations were rare, many of the differences in translation were not material." What's the meaning of "material"?

4. In the Discussion, it is stated that "Brislin's approach to back-translation was regarded as the gold standard". What's the justification for adopting other approach e.g. TRAPD for translation but not adopting Brislin's approach? What's the advantages of TRAPD over Brislin's approach?

==============================

Please submit your revised manuscript by 28 July 2021. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ka Ming CHOW

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. In your Methods section, please provide a justification for the sample size used in your study during the pre-testing of the translated questionnaire, including any relevant power calculations (if applicable).

3. Please provide additional details regarding participant consent. In the ethics statement in the Methods and online submission information, please ensure that you have specified whether consent was written or verbal/oral. If consent was verbal/oral, please specify: 1) whether the ethics committee approved the verbal/oral consent procedure, 2) why written consent could not be obtained, and 3) how verbal/oral consent was recorded. If your study included minors, please state whether you obtained consent from parents or guardians in these cases. If the need for consent was waived by the ethics committee, please include this information.

4. Thank you for stating the following financial disclosure:

"This project was funded by the Australian Research Council (https://www.arc.gov.au/), Curtin University (https://www.curtin.edu.au/), ShineSA (https://shinesa.org.au/), the Queensland Department of Health (https://www.health.qld.gov.au/), the Western Australian Department of Health (https://ww2.health.wa.gov.au/), the South Australian Department of Health (https://www.sahealth.sa.gov.au) and the Victorian Department of Health and Human Services (https://www.dhhs.vic.gov.au/).

The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript."

We note that one or more of the authors is affiliated with the funding organization, indicating the funder may have had some role in the design, data collection, analysis or preparation of your manuscript for publication; in other words, the funder played an indirect role through the participation of the co-authors. If the funding organization did not play a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript and only provided financial support in the form of authors' salaries and/or research materials, please do the following:

a. Review your statements relating to the author contributions, and ensure you have specifically and accurately indicated the role(s) that these authors had in your study. These amendments should be made in the online form.

b. Confirm in your cover letter that you agree with the following statement, and we will change the online submission form on your behalf:

“The funder provided support in the form of salaries for authors [insert relevant initials], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.

5. Please include your full ethics statement in the ‘Methods’ section of your manuscript file. In your statement, please include the full name of the IRB or ethics committee who approved or waived your study, as well as whether or not you obtained informed written or verbal consent. If consent was waived for your study, please include this information in your statement as well.

6. Please remove your figures from within your manuscript file, leaving only the individual TIFF/EPS image files, uploaded separately.  These will be automatically included in the reviewers’ PDF.

19 Jul 2021

Dear Editors,

Translating best practice into real practice: methods, results and lessons from a project to translate an English sexual health survey into four Asian languages

Thank you for your considered and helpful feedback on our manuscript. We have reproduced and responded to each item of feedback below.

1. ASSESSING EQUIVALENCE

Feedback:

• The team approach TRAPD for translation is systematic, but no statistical method conducted to assess the equivalence such as content validity test (CVI). Is it possible to add or combine statistical method to increase the validity of the translation?

Response:

While statistical techniques such as exploratory and confirmatory factor analysis are commonly used to test equivalence of clinical assessment instruments and screening tools, they are less common in the context of population surveys geared towards measuring knowledge, attitudes and behaviours. The Cross-Cultural Survey Guidelines state that: “When multiple languages are used in the same survey, pretesting the different language versions is an essential part of ensuring measurement equivalence and cultural and cross-cultural equivalence” (emphasis added). Pretesting data are typically collected using qualitative methods such as cognitive interviews, focus group discussions and behaviour coding.

Assessing equivalence through the use of qualitative methods is the approach taken in a number of population surveys, including:

• The National Health Interview Survey (Cancer Control Module) and the California Health Interview Survey (Willis G, Stapleton M.S., Leven K., Norberg A., Stark D., Forsyth B., Brick P.D., Berrigan D., Thompson F.E., Lawrence D., Hartman A. Evaluation of a multistep survey translation process. In Harkness J., et al (eds). Survey methods in multinational, multiregional, and multicultural contexts. London, John Wiley & Sons. 2010);

• The Tobacco Use Survey component of the Current Population Survey (Forsyth B., Kudela M.S., Levin K., Lawrence D., Willis G. Methods for translating an English-language survey questionnaire on tobacco use in Mandarin, Cantonese, Korean, and Vietnamese. Field Methods. 2007;19(3):264-283);

• US Household Food Security Survey (Kwan C.M., Napoles A.M., Chou J., Seligman H.K. Development of a conceptually equivalent Chinese-language translation of the US Household Food Security Survey Module for Chinese immigrants to the USA. Public Health Nutr. 2015;18(2):242-250);

• An 11-language survey conducted by the European School on Interdisciplinary Tinnitus Research (Biswas R., Lugo A., Gallus S., Akeroyd M.A., Hall D.A. Standardized questions in English for estimating tinnitus prevalence and severity, hearing difficulty and usage of healthcare resources, and their translation into 11 European languages. Hear. Res. 2019;377:330-338); and

• The Parent Attitudes about Childhood Vaccines Survey (Cunningham R.M., Kerr G.B., Orobio J., et al Development of a Spanish version of the parent attitudes about childhood vaccines survey. Hum. Vaccines Immunother. 2019;15(5):1106-1110).

In our Discussion we note that:

… [A]dditional and innovative techniques for quality appraisal are prudent to check the quality of translations. For instance, in the context of our study, we plan to compare results obtained from English-language and target-language surveys, controlling for factors such as country of birth, length of time in Australia and gender. Any differences in responses may serve as an indication that the translated version did not convey the same meaning as the English version of the survey, prompting further investigation into the adequacy of the translation.

However, we consider that actually undertaking this analysis lies beyond the scope of this manuscript, the explicit aim of which is to document the process and results of applying the conventional TRAPD method.

2. SAMPLE SIZE

Feedback:

• In the pretesting phase, the sample size of Vietnamese was 20, but other migrant communities were 3 or 4. Why was there a great variation in the sample size in different groups?

• In your Methods section, please provide a justification for the sample size used in your study during the pre-testing of the translated questionnaire, including any relevant power calculations (if applicable).

Response:

Decisions around sample size for pretesting must be read in context. In the Introduction, we state: “The project formed one part of study to ascertain the feasibility of conducting a periodic national survey in Australia to measure migrants’ knowledge, attitudes and practices in relation to sexual health and blood-borne viruses.” We have further clarified the study’s emphasis on investigating feasibility in the first paragraph of the Methods section:

“Due to budgetary constraints, the English-language survey could only be translated into a limited number of languages. As this was a feasibility study, four languages were chosen to gain insight into the complexities of multi-language survey development and administration”.

Additionally, we have redrafted the pretesting sub-section of the Methods to make it clear that the sampling strategy for pretesting was pragmatic and primarily guided by the availability of resources:

The minimum target sample size for the pilot survey (in any language) was 1,116 respondents equally divided between the three regions of birth (Sub-Saharan Africa, North East Asia and South-East Asia) to detect regional differences at a significance level of 5% and 90% power. Based on available Census data on country of birth and English-language proficiency, it was estimated that:

• 82.00% (n=305) of North East Asian-born respondents would be from a country in which Chinese was the dominant language and that, assuming all spoke some Chinese, 25.9% (n=79) would be likely to not speak English well or at all;

• 25.21% (n=94) of South East Asian-born respondents would be from Vietnam and that, assuming all spoke Vietnamese, 31.50% (n=30) would be likely to not speak English well or at all;

• 3.81% (n=14) of South East Asian-born respondents would be from Cambodia and that, assuming all spoke Khmer, 33.50% (n=5) would be likely to not speak English well or at all;

• 3.75% (n=14) of South-East Asian-born respondents would be from Myanmar and less than 10% would speak Karen (proportion of those who would be likely to speak English not well or not at all is unknown) [20,21].

Given these small numbers, the limited availability of resources for pretesting, the exploratory nature of the feasibility study, and the fact that multiple quality checking methods were built into the study design (e.g. two independent translations, review and adjudication), a pragmatic decision was made to only pretest translations on small samples. Pretest participants were recruited using convenience sampling by two members of the research team who had experience working with these migrant communities. The size of the samples varied by community – Chinese (n=3), Vietnamese (n=20), Khmer (n=4) and Karen (n=3). The larger sample size for Vietnamese pretesting was opportunistic. While all pretest participants were fluent in the survey language being pretested, no other demographic characteristics were recorded.

In the Discussion section we have also now expressly noted our approach to pretesting as a limitation (see line 400 of revised transcript).

3. ETHICS

Feedback:

• Please provide additional details regarding participant consent. In the ethics statement in the Methods and online submission information, please ensure that you have specified whether consent was written or verbal/oral. If consent was verbal/oral, please specify: 1) whether the ethics committee approved the verbal/oral consent procedure, 2) why written consent could not be obtained, and 3) how verbal/oral consent was recorded. If your study included minors, please state whether you obtained consent from parents or guardians in these cases. If the need for consent was waived by the ethics committee, please include this information.

• Please include your full ethics statement in the ‘Methods’ section of your manuscript file. In your statement, please include the full name of the IRB or ethics committee who approved or waived your study, as well as whether or not you obtained informed written or verbal consent. If consent was waived for your study, please include this information in your statement as well.

Response:

The Methods section has been amended as shown below:

Ethics approval for pretesting was obtained (CUHREC Curtin University Human Research Ethics Committee 0395/2019-0395) and participants provided written consent.

4. RESULTS

Feedback

• In line 141-142, "While identical independent translations were rare, many of the differences in translation were not material." What's the meaning of "material"?

Response:

We have amended the line as follows: “While identical independent translations were rare, many of the differences in translation were not material in the sense that they did not change the intended meaning of the source text.”

5. DISCUSSION

Feedback:

• In the Discussion, it is stated that "Brislin's approach to back-translation was regarded as the gold standard". What's the justification for adopting other approach e.g. TRAPD for translation but not adopting Brislin's approach? What's the advantages of TRAPD over Brislin's approach?

Response:

The Introduction has been revised as follows to address this earlier in the manuscript:

It is generally recognised that ‘team translation’ represents the best current method for translating surveys.[11] The Guidelines for Best Practice in Cross-Culture Surveys recommend “a team translation approach for survey instrument production” noting that “[o]ther approaches, such as back translation, although recommended in the past, do not comply with the latest translation research” [11]. ‘Team translation’ is considered preferable to other approaches on the basis that it enables people with complementary knowledge and expertise (e.g. language and cultural experts, survey researchers, and people with expert knowledge relevant to the particular survey topic) to work together to arrive at the best translation to ensure that survey items convey what they were intended to convey.[12] In team translation, two independent translations are produced and are then compared (item-by-item) by bilingual reviewers who work with an adjudicator to identify the ‘best’ translation for pretesting.[11] However, there are relatively few published examples describing how team translation has been implemented in practice, and much of the available literature relates to translations carried out as part of large and relatively well-resourced surveys, such as the European Social Survey.[12-1613-17]

The Discussion has also been revised as follows:

Finally, it is worth remembering that our understanding of what constitutes best practice changes over time. It was not long ago that For decades, Brislin’s approach to back-translation was regarded as the gold standard.[32 12] However, an over-reliance on back-translation is now considered prone to producing more literal translations which can overlook important conceptual or cultural nuances.[12, 34, 35] As Bretschneider and colleagues note, “[t]he term ‘best practice’ implies that it is best when compared to any alternative course of action”.[33] While the in-depth discussions that occur in team translation methods such as TRAPD are considered useful in detecting nuances in meaning and achieving equivalence between languages, However it has been noted empirical data on “the contribution of the TRAPD process … to the overall quality of the translated survey instrument”[24] remains limited, and studies comparing the results of different translation methods are even more rare. Consequently, it makes sense for researchers to remain open to exploring and gathering data on a variety of approaches to translation provided they are transparently reported and critically appraised.

6. STYLE

Feedback:

• Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming.

Response:

All files have been re-uploaded to meet the requirements for file naming. All other style requirements have been reviewed and the following changes have been made:

• Line numbers added to title page;

• Title on title page centred and font size increased;

• Corresponding authors initials added in parentheses after email address;

• Abstract reformatted to appear as a single paragraph with first-line indentation;

• Figure 1 now cited as ‘Fig 1’;

• Full stops added to the end of Table titles.

7. FINANCIAL DISCLOSURE

Feedback:

• We note that one or more of the authors is affiliated with the funding organization, indicating the funder may have had some role in the design, data collection, analysis or preparation of your manuscript for publication; in other words, the funder played an indirect role through the participation of the co-authors. If the funding organization did not play a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript and only provided financial support in the form of authors' salaries and/or research materials, please do the following:

a) Review your statements relating to the author contributions, and ensure you have specifically and accurately indicated the role(s) that these authors had in your study. These amendments should be made in the online form.

b) Confirm in your cover letter that you agree with the following statement, and we will change the online submission form on your behalf:

“The funder provided support in the form of salaries for authors [insert relevant initials], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.

Response:

We confirm the following statement:

The funder Curtin University provided support in the form of salaries for authors DV, MR, AR, and RL and scholarship support to NP, but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.

8. FIGURES

Feedback:

• Please remove your figures from within your manuscript file, leaving only the individual TIFF/EPS image files, uploaded separately. These will be automatically included in the reviewers’ PDF.

Response:

• The figure has been deleted in the revised manuscript.

We are grateful for the contributions you have made to help improve this paper. We hope that our revisions have adequately addressed the issues you have identified.

Yours sincerely,

The Authors

Submitted filename: Response to reviewers.docx

Click here for additional data file.

4 Aug 2021

PONE-D-21-17384R1

Translating best practice into real practice: methods, results and lessons from a project to translate an English sexual health survey into four Asian languages

PLOS ONE

Dear Dr. Vujcich,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

ACADEMIC EDITOR:

Thank you for resubmitting the manuscript. The authors have attempted to address the comments, but the justification for some issues are not well supported, including the disproportion sample size for pre-testing, and the strengths of adopting TRAPD model for instrument translation. Suggest to discuss further on these two issues. Also discuss the impact and significance of this study on future research.

Please submit your revised manuscript by 25 Aug 2021. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ka Ming Chow

Academic Editor

PLOS ONE

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: N/A

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Thank you for revising and resubmitting the paper. The authors addressed the reviewer’s comments and provided supplementary information in the text to clarify several methodological issues. I understand that the study was limited by various methodological and budgetary constraints. However, some of the decisions in the study need to have a stronger justification. Please find my specific comments on the responses below.

1. Sample size: The authors have further explained the reason for recruiting a relatively small sample for pretesting, which is understandable. While budget is limited, I notice that the number of respondents is disproportionate among different communities. Only 3 Chinese (3.8% of the estimated population size) but 20 Vietnamese (66.7%) were recruited. This difference cannot be solely explained by budgetary constraint. Were there any practical reasons?

2. Discussion: The authors suggested additional research on the application of TRAPD model in translating survey instruments. Why kind of research or data do they recommend? Meanwhile, after revisiting the discussion, I have questions about the advantages of TRAPD model – (i) In this study, the ‘leaner’ TRAPD model could save time and resources for translation. However, is there evidence that the outcome of TRAPD model (e.g., culture adaptability and language) is comparable to the original TRAPD model and the conventional approach? (ii) Regarding the rigour, I notice that the TRAPD model was not adhered thoroughly. For example, some reviewers did not provide their comments for adjudication and documentation was simplified. Would the failure to adhere suggest the difficulty maintaining the rigour of the model in practice? (As you argued that the TRAPD model is the ‘real practice’)

Additional comments:

1. P.9: Please explain why simplified Chinese does not need to be reviewed separately? In fact, you mentioned that differences existed in Chinese used in Mainland China, Taiwan, and Hong Kong (in fact, there are other major communities using Chinese).

2. P.9-13: Were findings of all languages presented in the Results section? I notice that much of the results focuses on the Traditional Chinese version.

3. P.14-18: I think the numbering of paragraphs is not necessary, especially when some points are actually related (e.g., #6 and #7).

Overall, I would suggest the authors to review the aim of this article – Is it a methodological discussion or research paper? If it focuses on the methodology, a more critical discussion is warranted to compare the TRAPD model with other translation approaches in terms of its reliability and practicality, with sharing of practical experience and insightful recommendations. If it is more research-focused, findings need to be presented in an analytical manner in respective to a well-defined research aim.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

14 Nov 2021

November 2021

Dear Editors,

Translating best practice into real practice: methods, results and lessons from a project to translate an English sexual health survey into four Asian languages

Thank you for providing additional feedback on our manuscript. We have reproduced and responded to each item of feedback below (see attached version for proper formatting). Red font indicates new material, and ‘strike through’ shows where material from the previous draft has been deleted.

Introduction

1. The authors have attempted to address the comments, but the justification for some issues are not well supported, including … the strengths of adopting TRAPD model for instrument translation. [Editor]

We amended the Introduction to further emphasise the strengths of team translation, one version of which is TRAPD:

There are several approaches to translation. Forward-only translation (also known as ‘direct’ or ‘one-for-one’ translation) involves a single individual translating an instrument from one language (the source language) into a second language (the target language).[11] While forward-only translation has the advantage of saving time and costs, it is considered problematic because it “involves a total dependence on the [single] translator’s skill and knowledge, and often results in low validity and reliability”.[12] Forward-backward translation (also known simply as ‘back translation’) represents an attempt to overcome the risks inherent in relying on a single individual. In forward-backward translation, a second individual translates the target language instrument back into the source language; the original source language instrument and the back-translated source language instrument are then compared, and any discrepancies serve as indications of the need for further refinements of the translation.[13] However, a criticism of forward-backward translation is that it has the potential to focus too narrowly on the task of literal translation at the expense of ensuring that the translation captures the intended meaning of the survey item in a way that is clear and suitable for the intended audience.[14] Similarly, Behr cites an example in which ‘care services’ was forward translated into German as pflegedienste and back translated as ‘care services’ suggesting no error, when in fact the translated term “did not fit the questionnaire context since it is only used in the context of the ill and/or the elderly and is thus not fitting to general child care services”.[15] Ozolins and colleagues have reported how some forward translators choose literal translations (despite their misgivings as to whether it actually captures the intended meaning) because they do not want their translation to be flagged as an ‘error’ by the back translator.[16]

Consequently, there has been a growing call for more nuanced and layered approaches to instrument translation in which accredited translators, other people who speak both the source and target languages, survey researchers, and subject matter experts work together to produce translated surveys which: (1) capture the intended meaning of the source instrument; (2) reflect the cultural and contextual specificities of the target population; and (3) will facilitate meaningful comparisons of data across populations.[13, 17-20] Indeed, the Guidelines for Best Practice in Cross-Cultural Surveys recommend “a team translation approach for survey instrument production” noting that “[o]ther approaches, such as back translation, although recommended in the past, do not comply with the latest translation research” [21]. ‘Team translation’ (also known as ‘committee translation’) is considered preferable to other approaches on the basis that it enables people with complementary knowledge and expertise (e.g. language and culture, survey methodology, knowledge relevant to the particular survey topic) to work together to arrive at the best translation to ensure that survey items convey what they were intended to convey to the target audience.[22]

While team translation can assume a variety of forms, the approach known as TRAPD (Translation, Review, Adjudication, Pretesting, and Documentation) is the version endorsed in the Guidelines for Best Practice in Cross-Cultural Surveys.[21] Under the TRAPD model, In team translation two independent translations are produced and are then compared (item-by-item) by bilingual reviewers who possess study design and subject-matter knowledge, who and work with an adjudicator to identify the ‘best’ translation for pretesting; each step is documented for transparency.[21, 23]

2. I would suggest the authors to review the aim of this article – Is it a methodological discussion or research paper? If it focuses on the methodology, a more critical discussion is warranted to compare the TRAPD model with other translation approaches in terms of its reliability and practicality, with sharing of practical experience and insightful recommendations. If it is more research-focused, findings need to be presented in an analytical manner in respective to a well-defined research aim. [Reviewer]

Thank you for this suggestion. We have revised the Introduction to more clearly communicate the intended aim and scope of our manuscript:

In this article, we apply the TRAPD model to translate an English-language sexual health and blood-borne virus survey into four languages for migrants living in Australia. The aim of the study is to:

(1) document how TRAPD can be applied in practice, including any challenges in its application;

(2) provide examples of issues identified through the processes of team-based ‘review and adjudication’ and pretesting, which are key features of the TRAPD model;

(3) offer guidance to future researchers who seek to use the TRAPD method; and

(4) provide recommendations for further research priorities on the subject of survey translation.

The aim is not to empirically test the effectiveness of TRAPD relative to other methods of translation.

The study makes an important contribution to the literature since there are However, there are relatively few published examples describing how TRAPD team translation has been implemented in practice the context of survey research, despite it being the model endorsed in the Guidelines for Best Practice in Cross-Cultural Surveys., and m Much of the available TRAPD literature relates to translations carried out as part of large and relatively well-resourced surveys, such as the European Social Survey.[24-28] However, as Sha and Lai have argued that “[i]t is important to identify a viable translation process that can be adapted and tailored to the varying level of expertise and resources available”.[29] The purpose of this paper is to document the process, results and lessons from a project to translate an English-language survey into four languages for migrants living in Australia. The project formed one part of study to ascertain the feasibility of conducting a periodic national survey in Australia to measure migrants’ knowledge, attitudes and practices in relation to sexual health and blood-borne viruses.

Amendments have also been made to the Results and Discussion sections of the paper to more explicitly reflect the aims described above.

Methods

3. “The authors have further explained the reason for recruiting a relatively small sample for pretesting, which is understandable. While budget is limited, I notice that the number of respondents is disproportionate among different communities. Only 3 Chinese (3.8% of the estimated population size) but 20 Vietnamese (66.7%) were recruited. This difference cannot be solely explained by budgetary constraint. Were there any practical reasons?” [Reviewer; editor also requested justification for the disproportionate sample size]

We have provided the following additional explanation:

The larger sample size for Vietnamese pretesting was opportunistic in the sense that a group of 20 participants were gathered for another purpose and expressed willingness to provide feedback on the translated instrument; although this resulted in the Vietnamese pretest sample being larger than those representing other language groups, the opportunity to obtain more feedback with minimal additional resources was recognised as an efficient means of obtaining more data to check instrument validity. There was no intention to engage in any statistical comparison of differences in pretest responses between the communities.

4. Please explain why simplified Chinese does not need to be reviewed separately? In fact, you mentioned that differences existed in Chinese used in Mainland China, Taiwan, and Hong Kong (in fact, there are other major communities using Chinese).

As stated in the Introduction, the four languages for translation using the TRAPD method were Traditional Chinese, Karen, Khmer and Vietnamese. Translation into Simplified Chinese using the TRAPD method was not a stated aim of this research and, consequently, the results from the Simplified Chinese adaptation process are not presented in the Results section (or in Appendix S6). We have decided to remove the reference to Simplified Chinese in the Methods section to avoid confusion. The process of adapting the Traditional Chinese survey into Simplified Chinese was distinct from the TRAPD method and, as such, does not fall within the scope of the current paper; it is better suited to a separate manuscript on the subject of the process of adapting translated survey tools for use with different linguistic communities/dialects.

Findings

5. In response to Question 4 (“Have the authors made all data underlying the findings in their manuscript fully available?”) the Reviewer has answered “No”. [Reviewer]

We note that we have provided raw data in the document marked Appendix 3 – TRAPD results. The results for each language are included in a separate workbook within the Excel file (screen shot in attached).

6. Were findings of all languages presented in the Results section? I notice that much of the results focuses on the Traditional Chinese version. [Reviewer]

We note that the results for each language are included in a separate workbook within the Excel file marked Appendix 3 – TRAPD results (see above). We have also significantly amended the structure of the Results section, including the introduction of a new Table 5 (see in attached version) to highlight examples from all four languages.

Discussion

7. “I think the numbering of paragraphs is not necessary, especially when some points are actually related (e.g., #6 and #7).” [Reviewer]

Thank you for this suggestion. We have now removed the numbering from the paragraphs. The Discussion section has been significantly redrafted in light of the feedback (point 2 above) that “a more critical discussion is warranted to compare the TRAPD model with other translation approaches in terms of its reliability and practicality, with sharing of practical experience and insightful recommendations.”

8. “[A]fter revisiting the discussion, I have questions about the advantages of TRAPD model – (i) In this study, the ‘leaner’ TRAPD model could save time and resources for translation. However, is there evidence that the outcome of TRAPD model (e.g., culture adaptability and language) is comparable to the original TRAPD model and the conventional approach?” [Reviewer]

We have added the following paragraphs in our significantly reworked Discussion:

Whereas much of the existing TRAPD literature reports on the use of the method in the context of large and relatively well-resourced, international surveys, this study documents the experience of applying the TRAPD method to a modestly-resourced survey in one country. The purpose of this study is not to provide evidence that TRAPD is the final word in effective translation methods. As has been noted elsewhere, empirical data on “the contribution of the TRAPD process … to the overall quality of the translated survey instrument”[18] remains limited, and more studies comparing the results of different translation methods are needed. Instead, this paper shows that the TRAPD method has value in identifying errors and exposing nuances in translation, but that it can be difficult to implement in practice, particularly where resources are low and translations into many languages are required. As Knight and colleagues acknowledge, the reality of studying minority populations is that “research context often makes full compliance with … best practice recommendations impossible and … the best one can do is to make incremental approximations of these recommendations.”[34]

As was reported in the Methods, deviations from TRAPD ‘best practice’ occurred. Significantly, it was not possible to employ or compensate translation reviewers; instead, the task of reviewing was undertaken in-kind which made it difficult both to recruit multiple reviewers for each language, and to rigidly proscribe the manner in which the translation reviews were undertaken. For instance, while reviewers were instructed to code translation errors using the European Social Survey (ESS) Verifier Intervention Categories, few reviewers used the ESS categories, and some did not document any reasons for preferring one translation over another which made the process of adjudication difficult. Problems around documentation in the TRAPD review process have been noted in other studies, including the European Social Survey where it was noted that “maintaining documentation can be burdensome … The documentation provided by the country teams on the development of ESS translations was at times meagre.”[21]

In addition to being time-consuming, the practice of coding translation errors may not be familiar to bilingual reviewers who do not necessarily possess a research background. Many of the distinctions between the categories – for instance, register/wording issues, grammar/syntax issues and minor linguistic defects (which can include grammar) – are subtle or require a high level of linguistic awareness. This suggests that the use of alternative coding protocols should be considered and tested, and efforts should be made to simplify the process of documenting reasons for decision at each stage of the translation process. In an effort to understand the difficulties they experienced in obtaining adequate documentation from translation partners, Harkness, Villar and Edwards surmised “[t]his may be because those involved were not familiar with how and what to document, but it is also likely that the effort involved in manual documentation played a role.”[21] One solution may be for a member of the research team to assume responsibility for documentation by sitting with each panel member (either physically or virtually) while they independently review the translations. The researcher could record the preferred translation for each item and the reviewer’s reasons, using probing questions where necessary; this was the approach taken for the Karen translation in our study. An alternative would be to investigate the development and use of specialised software to manage the documentation process. It is noted that an online, interactive Translation Management Tool was developed for the Survey of Health, Ageing and Retirement in Europe (SHARE) and tested in round 8 of the European Social Survey in 2016, although a number of areas for improvement were identified.[21] Currently, the tool is not publicly available but can be tailored to support specific survey requirements at a cost.

It is not possible to determine whether closer adherence to TRAPD ‘best practice’ in this study would have improved the quality of the resulting translations. A literature review of studies using multi-step, team-based translations of health quality of life questionnaires found that “[a]lthough there is some evidence that different methods (i.e., ‘light/simpler’ vs. ‘heavy/complex’) yield similar results, this has been tested empirically only on a very limited scale” [12]. However, our study does demonstrate that even ‘light/simpler’ adaptations of the TRAPD method can successfully identify issues that may not have been apparent had non-team-based or single-round translation approaches been adopted. For instance:

• relying on only one translator to conduct a forwards-only translation without further review would have resulted in important errors being overlooked (e.g. the Khmer translation of ‘casual sexual partner’ as ‘unprotected sexual partner’, and the Karen translation of ‘committed relationship’ as ‘willing in mind to live together’);

• relying on forward-backward translation only may not have revealed importance nuances (e.g. the translation of hepatitis B as 乙型肝炎 would likely have been translated back into English as ‘hepatitis B’ suggesting no problem with the translation, when in fact the review process highlighted that it would not be inclusive of/familiar to Taiwanese-Chinese speakers);

• not including researchers as partners in the translation process may have resulted in translations which did not reflect the intended meaning of the survey items (e.g. a bilingual reviewer without a sexual health and blood-borne virus background favoured the narrow, heteronormative translation of Karen translation of ‘sexual activities’ which meant ‘sex among men and women’);

• pretesting assisted in identifying typographical/formatting errors made when incorporating the reviewed translations into the survey tools (e.g. the omission for the term ‘no/not’ in one item in the Karen survey, and parts of the Karen script not being displayed correctly);

• the benefits of assembling a team comprising people with complementary expertise was particularly evident when attempting to translate technical, medical terminology in a way that, on the one hand, is meaningful to the target audience and, on the other hand, guards against perpetuating terms which are stigmatising or give rise to inaccurate assumptions (e.g. the difference between ‘clinical’ and ‘lay’ translations of HIV into Traditional Chinese).

It is, of course, possible that some errors in translation were not detected as a result of the approach we adopted. For this reason, we have taken the rare step of making all of our translation data publicly available (Appendix S3)…

9. “Regarding the rigour, I notice that the TRAPD model was not adhered thoroughly. For example, some reviewers did not provide their comments for adjudication and documentation was simplified. Would the failure to adhere suggest the difficulty maintaining the rigour of the model in practice? (As you argued that the TRAPD model is the ‘real practice’)” [Reviewer]

The Reviewer is correct to identify our conclusion that it may not be possible to rigidly apply TRAPD in all contexts. As noted in our response to Comment 8 (above), we argue that ‘light/simpler’ adaptations may be necessary and justifiable, and that more research needs to be undertaken on this point.

10. “The authors suggested additional research on the application of TRAPD model in translating survey instruments. Why kind of research or data do they recommend?” [Reviewer]

11. “discuss the impact and significance of this study on future research.” [Editor]

We have responded to Comments 10 and 11 through the addition of the following paragraphs:

Given the dearth of clear empirical evidence about the most accurate and feasible method of undertaking translations, we encourage future researchers to follow our example. Greater transparency around survey translation research processes and results will: (1) allow the accuracy of different approaches to be studied in greater detail; (2) provide practical insight into effective ways of adapting established processes in response to different needs, settings and levels of resourcing; and (3) enable the validity of the survey results to be assessed with reference to the quality of the underlying translations.

Unless a culture of transparent trial-and-error in translation is fostered, negative consequences will follow. First, survey researchers may feel daunted by the complexity (and uncertainty) of translation and may therefore avoid attempts to engage with linguistically diverse populations; this would either result in: (a) underrepresented sections of the population continuing to be overlooked in studies geared towards collecting data to improve health outcomes, or (b) respondents having no option but to complete surveys in a language other than their preferred language, thus increasing the potential for survey items to be misunderstood and data quality to be compromised. Second, survey researchers who have attempted to undertake translations but whose methods have deviated from ‘best practice’ (e.g. due to circumstantial challenges like budget limitations) may hesitate to publish their translation processes and results for fear of being criticised; the consequence of this would be a continued dearth of information about the most effective and efficient means of undertaking survey translations. As a recent editorial in Nature Human Behaviour makes plain, “[s]cience is messy, and the results of research rarely conform fully to plan or expectation. ‘Clean’ narratives are an artefact of inappropriate pressures and the culture they have generated.”[36] It is only through more detailed accounts of the processes and results of survey translation that we can come closer to understanding which methods (and variations of methods) are best suited to produce the most accurate results in specific contexts. Given the increase in the scale of international (pre-pandemic) migration,[37] and the growing appetite for multinational, multicultural and multiregional surveys,[38] the importance of continuing to build the evidence-base for survey translation is both apparent and urgent.

We trust that these amendments adequately address your comments and we are grateful for the role you have played in strengthening our manuscript.

Yours sincerely,

The authors

Submitted filename: Response to reviewers.docx

Click here for additional data file.

19 Nov 2021

PONE-D-21-17384R2Translating best practice into real practice: methods, results and lessons from a project to translate an English sexual health survey into four Asian languagesPLOS ONE

Dear Dr. Vujcich,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

==============================Thank you for submitting the revised manuscript. Basically, the authors have adequately addressed all the comments. However, there are still some minor comments and suggestions for improving the paper which are listed as follows:

1. In the abstract, in Line 44 and 45, the sentence "...TRAPD method can identify issues that may not have been apparent had non-team-based or single-round translation approaches have been adopted" is difficult to understand. Please rephrase and revise.

2. In Line 116, the sentence "the aim is not to empirically test the effectiveness of TRAPD relative to other methods of translation." can be deleted.

3. I would like to clarify that supplementary files including S1, S2 and S3 will not be included in the published manuscript.

4. In the section or pre-testing, in Line 205-219, "The minimum target sample size ...... who would likely to speak English not well or not at all is unknown" should be deleted as the calculated sample size was not achieved and not related to the pre-testing survey.

5. In Table 4, it is shown that 9 questions resulted in identical independent translations, but in Line 249, it is stated that 12 were identical. Please clarify and revise accordingly. Please ensure that your decision is justified on PLOS ONE’s publication criteria and not, for example, on novelty or perceived impact.

For Lab, Study and Registered Report Protocols: These article types are not expected to include results but may include pilot data.

==============================

Please submit your revised manuscript by 3 December 2021. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ka Ming Chow

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: N/A

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Thank you very much for thoughtfully addressing our comments. The revised manuscript provides a critical reflection on the adoption of the TRAPD method to tranlate the sexual health and blood-borne virus survey. The revised Introduction is well-written and demonstrates the significance of using TRAPD method. In the Discussion, the authors successfully examined the processes, shared the lessons learnt, and compared the current practice with the original TRAPD method.

One suggestion is to add a brief conclusion at the end of the manuscript.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Marques Ng

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

22 Nov 2021

Dear Editors,

Translating best practice into real practice: methods, results and lessons from a project to translate an English sexual health survey into four Asian languages

Thank you for providing additional feedback on our manuscript. We have reproduced and responded to each item of feedback below.

FEEDBACK FROM EDITOR

1. In the abstract, in Line 44 and 45, the sentence "...TRAPD method can identify issues that may not have been apparent had non-team-based or single-round translation approaches have been adopted" is difficult to understand. Please rephrase and revise.

Both the unmarked and tracked versions of the manuscript had the correct wording, as follows:

However, our study does demonstrate that even adaptations of the TRAPD method can identify issues that may not have been apparent had non-team-based or single-round translation approaches been adopted.

2. In Line 116, the sentence "the aim is not to empirically test the effectiveness of TRAPD relative to other methods of translation." can be deleted.

We have deleted the sentence.

3. I would like to clarify that supplementary files including S1, S2 and S3 will not be included in the published manuscript.

We understand from PLOS ONE guidelines that:

Supporting information is auxiliary to the main content of the article. In the online version of the published article, readers access the files via hyperlinks in the Supporting Information section of the article. PLOS hosts these files on its servers and also deposits them on Figshare to increase compliance with the FAIR principles of data sharing. Supporting information files are published exactly as provided, and are not copyedited … The PLOS publishing platform supports any file type for supporting information files.

We are happy for the supplementary files to be accessible via hyperlinks to the documents hosted on the PLOS servers/Figshare. However, if you require us to provide the files in a different format, or to host them on our project website, please let us know so that we can accommodate your request.

4. In the section or pre-testing, in Line 205-219, "The minimum target sample size ...... who would likely to speak English not well or not at all is unknown" should be deleted as the calculated sample size was not achieved and not related to the pre-testing survey.

We have deleted these lines as indicated in the attached response to reviewers, and amended reference numbers accordingly.

5. In Table 4, it is shown that 9 questions resulted in identical independent translations, but in Line 249, it is stated that 12 were identical. Please clarify and revise accordingly.

Thank you for pointing this out; it should read ‘nine’; we have amended accordingly.

FEEDBACK FROM REVIEWER 1

6. One suggestion is to add a brief conclusion at the end of the manuscript.

We have added the following sub-section:

Conclusions

Survey translation is time-consuming and costly and, as a result, deviations from TRAPD ‘best practice’ occurred. It is not possible to determine whether closer adherence to TRAPD ‘best practice’ would have improved the quality of the resulting translations. However, our study does demonstrate that even adaptations of the TRAPD method can identify issues that may not have been apparent had non-team-based or single-round translation approaches been adopted. Given the dearth of clear empirical evidence about the most accurate and feasible method of undertaking translations, we encourage future researchers to follow our example of making translation data publicly available to enhance transparency and enable critical appraisal.

We trust that these amendments adequately address your comments. Please advise whether you require anything further from us.

Yours sincerely,

The authors

Submitted filename: Response to reviewers.docx

Click here for additional data file.

24 Nov 2021

Translating best practice into real practice: methods, results and lessons from a project to translate an English sexual health survey into four Asian languages

PONE-D-21-17384R3

Dear Dr. Vujcich,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Ka Ming Chow

Academic Editor

PLOS ONE

29 Nov 2021

PONE-D-21-17384R3

Translating best practice into real practice: methods, results and lessons from a project to translate an English sexual health survey into four Asian languages

Dear Dr. Vujcich:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Ka Ming Chow

Academic Editor

PLOS ONE