I don’t think I’d be making the declaration of the century by saying that our cultural origins influence our habits and values. That would be too obvious.

What if I were to say that our cultural origins influence our way of taking surveys? Okay, probably not the declaration of the century either, but it might spark your curiosity a little.

In a recent qualitative study conducted for one of our clients, we asked participants to rate their satisfaction towards a service on a scale of 1 to 10 (revolutionary, isn’t it)? Most answers were between 7 and 9. We repeated the exercise, with the same question and the same topic, but this time exclusively with participants of a specific cultural group. Answers were almost exclusively 5’s and 6’s.

Do respondents from this specific cultural origin receive a lower quality service? Do they suffer from discrimination or racism when receiving this service? No, of course not. In fact, when looking at the reasoning behind their rating and the various positive and negative aspects of the service, the two different groups appear quite similar.

These aspects are also discussed in the same terms and participants place the same importance on them. Why in that case do we see different results?

The title might have spoiled the answer: cultural differences could be the cause. Fortunately, the topic has been studied at length.

One study[1] (among many) was conducted in 26 countries and showed several major differences in the way people answer surveys.  It confirmed hypotheses and previous research results:

  • For a similar satisfaction, it was observed that ratings from 1 to 10 given to illustrate the satisfaction level may be higher or lower based on the respondent’s country of origin. For example, a 7 in the U.S. may not mean the same as a 7 in Sweden;
  • For 4-point scale questions (e.g. very much, somewhat, not very much, not at all), answers tend to sway towards the middle when respondents answer in their second language (typically English). Generally, more answers are found at the extremes when they answer in their first language. Although respondents may be proficient in the survey’s language, their answers may not fully represent their feeling. This may lead to inaccuracies in particularly cosmopolitan or bilingual areas.

I could go on, but I will stop with these two examples.

These results beg the question: how can we accurately compare results from different countries? How should we interpret survey answers, particularly in cosmopolitan areas? In aera of unprecedented human mobility, research firms will have to find an answer in order to produce results mirroring today’s societies.


[1] Harzing, Anne-Wil (2006). Response styles in cross-national survey research: a 26-country study, International Journal of Crosscultural Management, vol. 6, no. 2, 37p.