Accounting for cultural bias in b2b research

However, delivering on this when carrying out international research can prove difficult – how can one ensure the reliability of standardised data that has been captured across countries and continents?

Despite being a recognised challenge in consumer research, cultural bias has not been discussed at length in a b2b context, and a knowledge gap exists. It is just as important for b2b marketers and researchers to understand the variety of ways in which respondents of different nationalities answer survey questions. As businesses increasingly look to reach a global and more diverse target audience, it is imperative that researchers are aware of the cultural predispositions and biases at play, in order to provide reliable, impactful studies in b2b markets.

Identifying cultural biases in international research

Let’s firstly identify the cultural biases that typically affect responses to scaled questions – namely 10-point numerical scales.

Different nationalities will use rating scales in various ways. Where US respondents tend to use all parts of the scale, those from Western Europe – namely France and the UK – tend to use the middle ratings ( 5-8 ) that result in lower aggregate scores. This is known as the midpoint response style. In contrast, Latin American markets – primarily Brazil and Mexico – use the extreme ends of the scale ( 1-4 and 9-10 ), which is known as the extreme response style. Generally, this leads to higher aggregate scores, which is why organisations often assume they are performing better in these markets.

Another response style influenced by nationality and culture is acquiescence bias. This is the inclination to agree with the interviewer or survey sponsor, i.e. to provide the desired response. Inclination to agree is higher in developing markets, such as Mexico, Brazil, China, and lower in the Pacific region, namely Singapore and Australia.

The methodology of research can also affect the strength of, or propensity for, cultural bias. For example, native language surveys often prompt more extreme responses (for example a 1-4 or 9-10 on a 1-10 scale), while non-native language survey administration (e.g. interviewing a Norwegian in English) can encourage a midpoint response style. Answering in a second language is likely lead to lower aggregate scores than if surveys were administered in the mother tongue – respondents generally feel more comfortable providing a definitive opinion if they are able to articulate themselves in their native language.

How can we reduce and/or account for cultural bias?

Although it is an inevitable part of international studies, market researchers must be able to identify and account for cultural bias to avoid misinterpretation and unreliable conclusions. Here are three simple ways to approach global research in order to limit the impact of cultural bias and develop actionable insights:

1. Wherever possible, surveys should be administered in the respondents’ native language. While this doesn’t eradicate bias, it lessens the negative skew caused by a strong mid-point response
2. ‘Worded’ scales can be more useful than numerical ones to avoid the flattening effect of the mid-point response style. Using ‘extremely/fairly/slightly and dissatisfied/satisfied’ is a good option, although there can be issues – such as the interpretation of the words ‘fairly’ and ‘slightly’. At the very least, numerical scales should be anchored, for example where one equals not at all satisfied and 10 is extremely satisfied
3. Researchers can ensure sampling consistency in tracker studies by setting quotas or establishing weighting criteria and then applying the same conditions to future surveys. Quotas or weighting criteria can be based on either the revenue split by country, number of customers by country, or by a determination of each country’s long-term potential to the business. This way, researchers can benchmark and track global metrics because the cultural bias will be consistent from wave to wave

In order to provide clients with reliable, meaningful results and insights that are actionable across borders, cultural influences in b2b market research must be understood and identified – just as they are in consumer research – to prevent recommendations built on skewed data.

Conor Wilcock is research director at B2B International