CULTURED MINDS
An Investigation of How Culture Affects Decision Making
RESPONSE BIAS
Response bias is a potential confounding factor for comparing feedback from different countries. In their analysis of 6 EU countries (Van Herk, Poortinga & Verhallen, 2003) found that “Rating scale scores did not match differences in actual behaviour between countries. This strongly confirms that ignoring national differences in response styles may lead to invalid inferences in cross-cultural research”
Some methods such as Likert-type scales are more likely to induce response bias than others e.g ranking (Harzing, 2006). Although a full discussion of response bias is beyond the scope of this paper, some key cultural effects on response biases are mentioned below.
2 Main Biases
Acquiescent Bias a.k.a ‘agreement bias’ is the tendency for respondents to agree with questions. Extreme response is the tendency to choose a maximum or minimum point/response.
Cultural Dimensions affect response Bias
Data for these links is patchy. Indeed (Smith & Fischer, 2008) wrote that “some nations do have a stronger culture of acquiescence than others. Unfortunately, we do not yet have available data that would permit an analysis of this type that focuses upon extreme response bias.” But results would appear to be tentatively converging.
For example, a comparison of 6 EU countries indicated that Greeks, whose culture scores extremely high on uncertainty avoidance and medium-high on power distance and masculinity (Geert-hofstede.com, 2015), are more likely to exhibit acquiescent bias and/or than the French, English, Spanish, Italian or Germans (Van Herk, Poortinga & Verhallen, 2003). These results were obtained for both product ratings and self-reports. In support of these findings, (Johnson, 2005) detected links between extreme response style, power distance and masculinity (independently). Further, (Dijk, Datema & Piggen, 2009) detected a positive association between uncertainty avoidance and both acquiescent bias and extreme response.
Meanwhile the English, people from a low-medium uncertainty avoidance and power distance culture (Geert-hofstede.com, 2015), showed the lowest acquiescent bias of all the countries (Van Herk, Poortinga & Verhallen, 2003). Accordingly a negative relationship was observed between acquiescent response style and uncertainty avoidance as well as power distance, individualism (Johnson, 2005; Dijk, Datema & Piggen, 2009) and affluence (Dijk, Datema & Piggen, 2009).
Regulatory Focus affects response Bias
(Crowe & Higgins, 1997) argue that in experiments that involve spotting objects/stimuli (e.g say yes when you see a red dot) respondents with a more promotion focus will display a risky response bias. Promotion-focused individuals will be more likely to answer “yes” when in doubt whereas prevention-focused respondents will tend to respond “no” (displaying risk-averse bias) when in doubt.
Language affects response Bias
In a 26 country study of response biases (Harzing, 2006) noticed that fluency in the questionnaire language (English) resulted in increased extreme response bias, whereas lower competence in the questionnaire language lead to more acquiescent (middle) responses.
