Tips and Concerns for Conducting Multi-Language Surveys

As globalisation continues to grow it is fast becoming more important for organizations that operate internationally to have the ability to canvas opinion across a number of geographic regions.

Despite the English language establishing itself firmly as the world's preferred business language it is naive for those whose native tongue is English, or who speak English as a native, to assume that communicating globally in English is always the best solution.

There are a number of online survey websites that to some degree offer multi-lingual facilities and some, like www.surveygalaxy.com, that can offer comprehensive multi-lingual facilities at no additional charge, but before embarking on what can be a very rewarding path there are many considerations that should be taken on board to ensure that when published the survey is presented professional in all your target languages.

Language Ability

A person who is quite confident and comfortable with conversational English may not be so confident, or even have the ability, when it comes to reading English and often, words used in conversational English differ to those used when English is written.

If the intention is to seek good intelligence it is important to appreciate that despite the great strides that many people make in learning English as a second language it is one thing to speak English and quite another to really understand the language.

Consider the native language of your respondents and if they are not native, but nevertheless English speaking, consider the level to which they generally speak and understand English.

Although English is specifically mentioned here as an example the same applies for languages such as French, Spanish and Portuguese whose historical reach is similar to that of English.

Machine Translation

Anyone who has had any experience of using a machine language translator will have probably already discovered the limitations of such facilities; language is simply too complex for many of the publically available automated language translators.

A simple test you can conduct to prove the point is to take a sentence and translate it using a machine translator from English to the target language and then take that translation and translate it back to English.

Not only are you likely to experience the problems associated with machine translators but you are also likely to begin to appreciate that it may also be an example of the type of confused meaning that those learning English could equally experience.

Human translators

For serious translation human involvement is essential so that the correct meaning of the question is conveyed but it is paramount to ensure that the person who is tasked with the translation is comfortable in both the translated and translation language.

Take the following example which is has been reproduced verbatim having been sourced from the back of a Chinese bootleg DVD.

Vincent LaMarca job is to arrest kieelers, but this job is different. The suspect he is tracking is his own son. He a cop, LaMarca must tring the acclised to Justice. As a Pather, he must find wining actoce Robert De Niro, Frances Mc Dernad and James Franco. De Nice memorably plays LaMarca, burdened by tragedy and his failings as a father and now putting his life on the line to do by his family and profession. "Do Nino has been sawing us greet stutt for30 years bou. But in this movie he shows us something nes." The San Fancisce Chroniole Mick La's Mick LaSalle wrote. Put another way: a greet actor rederines his greatness in City by the Sea.

In the above translation it is obvious that in some places a phonetic translation has been made and in actual fact the description of the movie as a whole is sufficient to convey the correct meaning, the main problem in this example is more to do with the spelling than the grammar and sentence structure.

Although perhaps an amusing example for someone who speaks English as a native it is in fact a good example of the problems and complexity associated with translations in general. It would be justifiable to surmise that the person responsible for the example translation is probably regarded locally as an expert.

The example is a Chinese poor attempt at an English translation, but their experiences should also be seen as a warning to the possible consequences of producing poor translations from one language into another and also the importance of ensuring that the person who is charged with the translation is properly qualified to translate and that the final translation is, where possible, independently verified.

Qualities of a good translator

Finding a well qualified translator can sometimes be further complicated depending on the subject matter of the survey. If the subject of a survey is related to a specialised field it can be of greater importance that the person responsible for the translation has some basic knowledge of the survey subject or there will be a risk that the translator will not fully understand what it is they are translating and, if the specialised field is technical or scientific, it may even be outside the scope of a well qualified translator.

Ambiguity

Even for single language surveys ambiguity can be a major concern, for multiple language surveys ambiguity can be one of the biggest obstacles to translation. To minimize the problem of ambiguity the source survey needs to be made as clear as possible.

Ambiguity is not always obvious and can sometimes only come to light on a second or third reading, or by two people interpreting the same question differently. Some ambiguity can go unnoticed and can lead to respondents answering a completely different question to that intended by the survey's author in which case the results will be fundamentally flawed and could be misleading and confused.

Regional Differences

There are many regional differences even between English speaking countries. Although the differences between British English and American English are well publicised and can often be the subject of humour, they can also represent a major problem.

Take for example the word 'pants', in British English the word 'pants' is interchangeable with 'underwear' whereas in American English pants refer to what the British would call 'trousers'.

If a survey is to be published across two or more dialects of the same language, all differences need to be addressed and when having the survey professionally translated into a non-English language the person responsible for the translation must know the dialect of the English they are translating.

If it is to be accepted that there are significant and potently problematic differences between established dialects of the same language take time to consider the potential extent of fundamental differences between two distinctly different languages.

The Survey Galaxy website has the facility to cope with surveys where a same language dialect exists and will allow a single survey to be offered to a respondent, for example, as English/British, English/US, English/Canadian.

Dates

In the English speaking countries the convention of how dates are written differs, in simple terms those countries that historically have a British influence might write 5-July-2009 as 5/7/2009 whereas those countries that have an American influence may write the same date as 7/5/2009.

Although a survey can indicate the desired format, i.e. either by requesting the date to be written as 'dd/mm/yyyy' or 'mm/dd/yyyy' you should not underestimate the human desire to always write a date in their own preferred format.

To some the problem of a respondent not writing a date in the required format may be firmly that of the respondent however, if the intention is to collate good intelligence then it is in the survey publisher's interest to identify and then minimise any potential problem.

Because of a globalised workforce it is not even safe to make the assumption that if a survey is to be published in a particular country that the national convention will prevail. If a survey is conducted in a metropolitan city such as London or New York some respondents are likely to be of different nationalities and may consciously or sub-consciously revert to type; in most cases the best solution is to always request dates to be either selected from a pop up calendar or written in full - i.e. 5 July 2009 where there can then be no question as to the actual date entered.

Clean eyes

Proof reading is very much a skill, the ability to force the mind to read what has been written and not what was intended is a difficult skill to master. A simple example of this is when people are asked to count the number of 'F's in the following sentence:

The answer is six but most people on first reading will usually answer between three or four.

It is recommended that when a survey is to be offered in more than one language that where practical, each translation is verified by more than one qualified person. If resources are scarce then the next best alternative would be to have the person responsible for the translation to at least proof read the translation after they have taken a reasonable break.

A good proof reader in any language will be looking to:

• Correct spelling mistakes and typos;
• Correct grammatical errors;
• Ensure that questions are clear and succinct.

Localization

When a survey is to be published across national borders consideration should be given to both the question and answer options. In particular care should be taken where reference is made to monetary values, unit of measures and if manufacturers and/or their products are referred to ensure that they do have a presence in the local country and that they use the same product name.

The Benefits

On considering some of the potential problems discussed here for publishing multi-language surveys it may deter some to offer multiple translations and to instead decree that all surveys will for simplicity be published in English, (or the dominant language).

As to the wisdom of such decisions it would be difficult to generalise but when gathering opinion, if the opinion being gathered is important, then the quality of the gathered intelligence will improve if respondents are asked the question and allowed to answer in their native tongue.

Facilities

Many of the online survey websites support non-English surveys, a large number of those also support the survey's system messages and control buttons to be displayed in the local languages. A few will provide a solution to allow multiple translations of the same survey and one that does offer all these features at no additional charge is www.surveygalaxy.com.

With Survey Galaxy you can nominate the base language and then provide the survey in as many languages as is necessary, the results being automatically rolled up into the base language for reporting purposes.

For more information or to discuss how online surveys can help you please contact surveys@surveygalaxy.com or visit www.surveygalaxy.com the quick, easy and cost effective way to do online surveys.