Common or tailored approaches to the international questionnaire surveys

When faced with the prospect of conducting a study across a number of countries the first issue is whether to write a separate questionnaire for each country or a single questionnaire that varies only on items such as brand lists.

This can only be answered by examining the objectives of the study and the known or likely differences between the markets. Downham (Worcester and Downham, 1978) lists the following differences that can have an effect upon the questionnaire:

• Language. There may be different languages not only between coun­tries but also within countries. Is it necessary to include all minority languages in all countries? Apparently common languages may have different usages, eg English in the UK and the USA.
• Ethnic differences. Different ethnic groups may speak different languages. Where they don’t, they may have different consumer habits and attitudes.
• Religion. This may be associated with ethnic differences, but may have implications for attitudes, lifestyle, and consumption of products such as alcohol and meat, for which different questions will be required both to make sense and not to offend.
• Culture and tradition. It would be wrong to ignore cultural differences, and questions must allow for the machismo culture in some Latin countries, the issue of ‘face’ in the Far East, and the different levels of importance given to gifting in different cultures.
• Literacy. Literacy levels vary between countries, and even official statistics can overstate it. Low literacy levels among the sample mean that aids such as verbal prompt material cannot be used, let alone self­completion questionnaires.
• Geography and climate. Differences in climate can mean that product usage patterns are different, particularly with regard to food products that are suited to either a warmer or cooler climate, such as butter and olive oil. Issues such as water hardness can also create different usage patterns for the same product.
• Institutional factors. Different market backgrounds often require differ­ent questions to be asked. Baths are more common than showers in some countries but rarely taken in others; approaches to clothes washing, savings and credit cards all vary between countries for reason of history and market development.
• Distribution. Supermarkets, hypermarkets and shopping malls domi­nate distribution of many goods in some countries but are unknown in others, where different questions may be needed.
• Media and advertising. The media that carry advertising vary between countries, and, even more so, the access to the media may vary.

To this list can be added:

• Infrastructure. Different infrastructures may have an impact on usage and attitudes. The greater use of communal heating systems in some countries than others, different transport systems, different stages of development in telecommunications, and different approaches to health care may all affect the way in which the questionnaire is written for different countries.

It may be relatively easy to have a common format for a brand awareness and image study in the pasta sauce market across a number of European countries, for example. The same spontaneous and prompted brand awareness questions can be used, and the same format used to determine brand images. The brand list will almost certainly vary between countries in most markets and the image dimensions measured may need sensitive adaptation, but the structure of the questionnaire can remain the same. There are a number of reasons, though, why the questionnaire approach may need to be different.

1. Different usage of product

In some product fields and markets a study may require completely different approaches for different countries. Some products are used in completely different ways in different parts of the world. For example, milk-based products that are used as night-time drinks in Europe are frequently used as aphrodisiacs or body-building products in parts of Africa and the West Indies, and razor blades are used to shear sheep in some parts of the world. It is unlikely that a single questionnaire could be used that would adequately describe the usage patterns of these products in all regions.

2. Different market segments

Market segments that exist in one country may not exist in another. Low and mid-priced Scotch whisky segments, which can account for the majority of the market in Western countries, may not exist in some Asian countries where only luxury brands are available. The usage questions and image dimensions that are appropriate for a market segment with a strong mid-priced segment of many brands may not be of any use in countries where the competitive set is not just Scotch but other high- priced luxury drinks.

3. Brands in different segments

Brands may be in different segments in different countries. This can happen in any market and is quite likely to happen in countries where distributors are used who are independent of the manufacturer and who have historically been given the authority to position the brand as they wish. Brands that in one country would be considered mid-priced may elsewhere be luxury brands. Good market data and local knowledge should identify this type of problem.

For most clients and researchers, the more the same questions can be asked in all countries under study, the easier it is to manage, analyse and report the study and the more likely it is that the client can adopt a common marketing approach. There can therefore sometimes be consid­erable pressure on the survey designer and questionnaire writer to adopt a common approach and set of questions. The client may want to adopt a common marketing strategy, but the researcher would not be doing his or her job if the client was led to believe that the markets possessed only a number of common characteristics and was left unaware of the differ­ences because they were not asked about.

The biggest danger is the assumption that because a questionnaire has been used successfully in one country it can be used in any country.

4. Comparability

Where a common research approach is adopted across countries, then, there are many reasons to try to make the questionnaires, and hence the data output, as comparable as possible. Downham (Worcester and Downham, 1978) again suggests that:

• Time and money are saved by using a standardized approach.
• Life is simplified for the researcher.
• End -users often have greater confidence in a standardized approach, rather than one that has many variations.
• Absolute uniformity is essential in some cases, particularly in the data required for the technical development of products.

Having a common questionnaire is also likely to lead to fewer errors in survey administration than if there are a number of different ones.

Given these reasons, most organizations would agree that a standard­ized questionnaire is always preferable and should be used unless there are good reasons that can be demonstrated why it would not be suitable for a particular country or group of countries.

One approach to writing questionnaires for a multi-country study is to start by writing the questionnaire with one country in mind. Once that has been refined, it should be tested for its appropriateness in every country in which it is to be used, even those sharing a common language. Amendments should then be made in order to accommodate differences between markets. This may require changes only in the brand lists, but it may also require changes in image dimensions, advertising media and prompts used, methods of distribution in the market, absolute prices, relative prices, the competitive product set, frequency of use bands, or completely different behavioural questions. The researcher reaches a point where the changes are so significant that it becomes a different questionnaire.

5. Coordinating common elements

Even if a study is able to use a standard questionnaire across a number of different countries, there will nearly always be minor variations to be accommodated.

5.1. Brand lists

Almost invariably the brand list will change in most consumer markets. There may be local brands that are available only in that country or region, and the multinational companies may sell different brands in different countries. Some brands of Scotch whisky, for example, are sold only in the Asia Pacific region. Others only have a significant level of distribution in a small number of European countries. The brand list in many product sectors is unlikely to be the same in any two countries.

The questionnaire writer needs to be aware of these differences, which will affect the brand lists used both as pre-codes and as prompts for ques­tions such as brand awareness, purchase and usage.

5.2. Brand image

Brand image questions are frequently asked of a small number of brands deemed to be important either in the market or in the direct competitive set to the client’s brand. Even if the long list of brands available is similar in two countries, the short list of brands that are the most relevant to be asked about in image and brand-positioning questions may vary between countries.

Frequently the client will be able to advise on the appropriate brands for each country both for the long and the short lists. This may come from the company’s marketing plans for each country and from the company’s office or representatives or distributors. It is always worthwhile to check the list with local representatives, who may be aware of new local brands that have not yet made it into the company’s global marketing strategy. It is also worthwhile for the research agency to ask its own representatives in each country for their views on the brand lists, for the same reason.

5.3. Image dimensions

Frequently the objective is to produce a single, global, brand image map on which variations between countries can be plotted. If insufficient care is taken in choosing the image dimensions relevant to each country, this can result in a misleading picture being produced for some countries because the brand position has been measured using a set of image dimensions developed for a different country and a different competitive brand set.

To achieve the ideal set of image dimensions the researcher should determine all the relevant image dimensions for each country, bearing in mind that the positioning and the competitors could be different. A preliminary stage of qualitative research to explore the way in which consumers in each country perceive the market and the brands in it can be used to give the most appropriate image dimensions for each country. For studies across many countries, however, this is frequently too costly and time-consuming to carry out. Findings from qualitative research that has already been conducted in a country for other purposes can often be used to provide a consumer-led picture of the market structure and brand perceptions. If that does not exist, reliance will sometimes be placed on qualitative research carried out in a few countries that are thought to be representative of a group of countries. Where this occurs, it is particularly important to pilot the questionnaire in the countries in which no qualita­tive research was carried out.

However it is arrived at, a distillation of all relevant image attributes across the countries in the study can be compiled to form a ‘master set’ of image dimensions.

If the intention is to use a technique such as correspondence analysis to produce a global map, then all image dimensions may have to be used in all countries regardless of their relevance. There is a danger that the list, in trying to accommodate the key points for each country without becom­ing overlong, will contain too many compromises. While it will provide a global overview, it will not be sufficiently detailed to provide an accurate positioning in any one country. Supplementary questions specific to each country may be required for that to be achieved.

5.4. Attitudinal questions

Attitudinal questions can sometimes be difficult in maintaining compara­bility between countries. Not only may consumers have different atti­tudes to a market or product area in different countries, but what is important to them in arriving at those attitudes may also be completely different.

Frequently, the attitude dimensions to be measured should be the same in each country, although with the expectation that response patterns will be very different between countries. If a battery of attitudinal rating scales is to be used, the wording of each dimension must be appropriate for each country, and care must be taken to avoid offence, in relation to both cultural and religious attitudes.

Source: Brace Ian (2018), Questionnaire Design: How to Plan, Structure and Write Survey Material for Effective Market Research, Kogan Page; 4th edition.