CLTs in Japan

Just about every research methodology has its own local twist to it when conducted in Japan. Central Location Tests (CLTs) are no exception, and while there are some similarities to other countries, it's best to be aware of the differences when planning this kind of research in Japan. Here are just a few to bear in mind:

1)	Locations First off, where do you conduct CLTs in Japan if you wish to capture a response that is nationally representative of the Japanese population? Well the good news is that unlike in most other developed markets, relatively few locations are needed in Japan. The population is so homogeneous in geographical terms that typically surveying metropolitan Tokyo alone is sufficient. The usual alternative to this is to also conduct the CLT in Osaka. The people of west Japan, which Osaka represents, sometimes show differences in tastes compared with those of east Japan (Tokyo). Tokyo and Osaka are often sampled in a 60:40 ratio. But its said that 90% of the comedians in Japan come from Osaka... so beware!
2)	Recruitment One of the decisions to be made for any CLT is whether respondents will be recruited using "Street Catch" or "Pre-recruitment". "Mall Catch" is generally not available in Japan, as this requires permission from the management of the mall, which is often not forthcoming. In most cases, "Street Catch" works well, and locations within Tokyo can be selected according to the particular specifications of the sample. For instance, the area around JMI 's own office in Shibuya is a Mecca for young, hip and trendy shoppers. If older or more suburban respondents are sought, an area such as Kichijoji is ideal. "Pre-recruitment" is usually required for very low incidence samples, long interviews and studies where a follow-up callback is included. The famous conscientiousness of Japanese is a great benefit when you do pre-recruit - expect hardly any "no shows" come rain, shine or earthquakes!
3)	Getting the right answers Japanese respondents are very serious about completing questionnaires. This means the quality of data collected tends to be very high, accurately reflecting their true feelings and opinions. Frivolous and falsified responses are generally not as much a problem in Japan as in many other countries. However, Japanese respondents tend NOT to answer in extremes. In other words, it is unlikely that many "extremely good (bad)", "strongly agree (disagree)" or " I am the first one among friends" responses will be seen. Because of this it's often best to use 7-pt rather than 5-pt scales wherever possible - but don't expect to see very many 1s and 7s!
4)	Lost in translation Of course high quality translation of questionnaires is vital. Translation is a skill that leads to success when done correctly, but to tragedy when done poorly. For examples of these kind of translations and the comic results that ensued, please see the following site: http://www.engrish.com/recent.php Use professional translators who are experienced in market research terminology -- both on front and back-end translations. This helps to ensure that culturally specific nuances are both captured and preserved in the research.
5)	How computers can help Lately many JMI clients are requesting web-based CAPI. Such facilities are available, in addition to the more traditional PC-based approach. However, whether you use traditional CAPI or web-based CAPI, please keep in mind that the Japanese language, like other Asian languages, contains double byte characters. This makes software programming more difficult. It is therefore recommended to factor in ample time to program and test questionnaires using computers in Japan. Often, we have found that Japanese questionnaires, self-programmed by clients on their own web site, are illegible on computers in Japan.
6)	And even cell phones! With over 70% of the Japanese adult population online 24/7 via their mobile phones, a short m-survey is great way to do a quick follow-up to a CLT. Upon a client's request, JMI is now providing this service on its CLTs conducted in Japan. How does it work? A cell-phone jabber is used at the end of the hall-interview, with the respondent's consent. They place their phone on the device for 10 seconds so that the necessary information is automatically gathered to allow us to send a short follow-up survey (5-10 questions) to respondents' phone. This can be done after the main CLT results have been analyzed. This way questions arising from the research can be answered easily and quickly, ensuring closure on the research for the end client. So if you do plan to conduct CLTs in Japan, please bear in mind some of these issues. In doing so the success of your project is assured. And if you plan on coming to town to see the fieldwork, we'd love to show you the delights of Tokyo!