‹ Tutorials

Tips for analysing international data

Cultural and language differences sometimes make it hard to compare country responses as different cultures tend to answer questions differently.

Sapio Research were a pleasure to work with, Jane was quick to pick up on our specific needs and turn around our complex, multi-national, small business research in a matter of days. We look forward to having the opportunity of working with her and the wider Sapio Research team again in the near future.

Michael Stovell
Commercial Manager
BCSG

In need of insights?

Success story

OpenText

Generating insight that to help position OpenText as thought leaders on Digital Transformation and Industry 4.0 in the Manufacturing sectors in the UK and Nordics.

Read more

We’ve seen in many of our research studies that some countries are more likely to say “yes” than others. This is because of the cultural/language definition of the word “yes” rather than the subject of the questions. Therefore, we have tips for analysing international data.

Tip No. 1 – look at rank order rather than compare percentages

You may be thinking – well, bias makes my research useless doesn’t it? Certainly not! There IS a way to find valuable insights from country by country data. Instead of comparing directly, look at the order that the results come in for each country – i.e. which option was chosen 1st, 2nd and 3rd in each?

Tip No. 2 – make sure the questionnaire is translated by professional native speakers

This may be pretty obvious, however in different countries questions are not asked in the same way. You need to translate the essence of the question. For example, in France a waiter will ask you ‘Que voudriez-vous prendre pour diner’ which literally means ‘What would you like to take for dinner’ whilst in English you’d be having something for dinner (rather than taking it), so be careful people don’t misinterpret the question.

Tip No. 3 – all sample subsets must have at least 50 responses

If you are collecting data from a large sample and breaking it down into subsets (e.g. countries, regions, genders etc) then make sure you collect at least 50 (ideally 100) per subset you’re interested in to allow for statistically robust comparisons. If you need the data to be representative of a certain population but can’t get the required sample size there is an option to weight the results to avoid bias – we’d be happy to speak to you about this if required.

Tip No. 4 – Be aware of cultural tendencies

Some countries have more of a natural disposition to please and agree so are more likely to score agreement questions higher (i.e. strongly agree). We tend to find this more so in the Far and Middle East. Also be aware of different meanings of numbers, for example in Germany ‘one’ tends to be the best score or position for something, whereas in the UK we might say 10 out of 10 is the highest score. Different colour codes (such as on survey buttons) can also lead to different responses as purple means death in some countries, red represents luck and happiness in China etc.

Tip No. 5 – take holidays and events into account

Different countries and cultures have unique public holidays and events throughout the year. These holidays can affect response rates, but may also affect responses. For example if you are asking how often someone has travelled more than 150 miles in the last month, you may find that the response for China would be much higher during their New Year and Spring Festival as they traditionally travel home to parents’ houses, compared to other countries.