Making sense? .The Japanese language and global research

Among advanced economies Japan is unique in terms of the complexity of its language AND the low levels of English generally spoken. This combination presents a challenge to market researchers. Indeed the integrity of the findings from global research projects conducted in Japan can be questioned when the issue has not been addressed sufficiently.

In terms of scores achieved in the Test of English as a Foreign Language (TOEFL), Japan has one of the lowest rankings in the world. One would expect more from the second largest economy globally. Yet astonishingly, given the amount of money poured into English education in Japan, the Japanese still ranked 18th out of 21 Asian nations in 1999 TOEFL tests and in the bottom 10% of the world! Against this background of poor English, it goes without saying that the Japanese language is extremely complex. Its written form draws on three distinctly different character sets. This structural complexity is compounded by intricacies concerning politeness, grammar, and of course culture.
		What does it all mean?

Foreign language can present an obstacle in multiple phases of the market research process. The Sapir-Whorf hypothesis is relevant here: "Distinctions encoded in one language are unique to that language alone."

The peculiarities of the Japanese language affect market research at several levels. Here we touch on three areas researchers should be aware of when conducting global projects in Japan:

1.	Translation of documents used to set-up and execute research e.g. questionnaires, prompts, discussion guides and test materials
Translations may seem a routine, even mundane, part of the set-up process. However simply handing a document to a professional translator may lead to more confusion than ever imagined:
•	Culture complicates! The word 'moisturize,' which can be translated 3 ways in Japanese - all with slightly different nuances for a Japanese woman shopping for hand soap. The Japanese word 'uruoi' could mean 'not dry,' while 'shittori' would have a damper connotation, and 'hoshitsu' is closer to 'maintaining moisture.' How can you determine which translation to use without a Japan-based client? If it is a key word in a branding campaign, it's definitely worth a teleconference with your Japan research partner to clarify the intended meaning.
•	Beware the quality control practice of back-translation! Like a wayward 'whisper down the lane,' translated Japanese being translated back into English can result in entirely different meanings! The word "unique" in English would be translated into Japanese with words that mean "unique," but when back translated may return as "different to other products." Additionally some words have different connotations that can then be interpreted and misunderstood by an independent translator. So it is preferable to have the translation checked against the original English by a Japanese person, such as the branch office of your client or by a Japanese person rather than back translating and comparing English scripts. It should also save you time and money!
•	No direct translations exist for some words! One extreme example of this was the big hoopla raised in the early 1990s as the Japanese and US government clashed over the trade deficit. US negotiators complained that the Japanese would say 'yes' when they meant 'no.' It is true that in certain circumstances, there is no exact translation for the English word 'no.' Accusations of deliberate deceptions followed, and the US got payback when then US President George Bush Sr. vomited beef sashimi in the lap of the Japanese prime minister. Don't let haphazard translation/interpretation throw you off. (.and lay off the raw beef if you have a tender stomach)
2.	Translation and interpretation of documents generated by research e.g. OAs, group transcripts ....
For certain projects, particularly qualitative, a Japanese supplier's deliverable to you might be verbatim answers or transcripts. Beyond the need for translation into smooth reading, the way in which Japanese speak may end up confusing your research analysts:
•	Implied meanings: Japanese is a language where the real important meaning is often conveyed by what is not said.       Often, phrases will just end in 'but.' with the key phrase left unsaid. It is important for your Japan research partner to either confirm what exactly the person's thoughts are here, OR include in parentheses, the implied meaning as understood by the interviewer.       "Yes, I'd love to go out on Friday,but it's a little." (never in a million years buddy!)
•	Six or more ways to say 'I?' Depending on the social situation and the way a person wants to be perceived, there are at least 4-6 different ways to refer to oneself (excluding regional dialects). This illustrates the importance of taking into account the environment when analyzing a transcript. It can be helpful if the research agency makes clear what kind of social situation that respondents are in and the implications with respect to their attitudes. This is especially pertinent in in-depth interviews or focus groups involving businesspersons!
•	In the backroom; As a foreign market researcher, it is nearly impossible to understand all of what is happening in a focus group or in-depth interview through a simultaneous translator. The linguistic situation between interviewer and respondent can have implications on the research data. And body language cannot be fully interpreted by a simultaneous translator.Rely on a bilingual moderator's debrief for interpretation of subtexts, connotations, and hidden meanings.

3. The use of programming -e.g. CATI, CAPI .

Japanese in common with other East Asian languages contains double byte characters, which makes the software programming for CATI and CAPI much more difficult.

The main technical problems in dealing with double byte text are in distinguishing single byte characters from double byte characters to ensure that double byte characters are never split up during display or printing. Further difficulties arise in table formatting with Asian characters in columns being formatted top to bottom rather than left to right. This can be complex when text can contain a mixture of Roman, Hiragana, Katakana or Kanji characters.

While CATI and CAPI are available now in Japan, access to more demanding functionalities such as automated OA grouping, can be difficult. Similarly, expect a delay (2-3 years) for new methodologies such as DS-CAPI to establish themselves in this market.

The importance of language in market research

In conclusion, even with the best questionnaire flows and advanced methods available, the quality of global research conducted in Japan will be defined by how linguistic differences can be handled by your research partner. It is so often the little things in the research process that can make or break a study in the end. But be assured the time invested in ensuring that language differences do not distort your findings will pay off handsomely in the end.