This article is about bringing qualitive and quantitave research together with integrating numbers and representativity into qualitative research. Quantitative research is primarely counting and statistics.
The topic of qualitative market research is understanding. What does the world of my target group look like? What emotional and financial settings do they have? Quantitative research, on the other hand, determines how many people in the target group have which attitudes, how many they are and how their budget looks. A random sample is drawn so that not all members have to be questioned.
Some entrepreneurs therefore believe that unstructured observation can replace qualitative market research. Sometimes that works too. Quantitative market research uses representative, large samples according to statistical rules and brings different results due to the large number of subjects.
Large institutes can save costs by comparing similar studies and using similarities to minimize the required sample size. Others sell the same report to multiple customers – that is Boutique Market Research.
The claim of qualitative market research
Qualitative market researchers understand the customers. Its like some newspapers articles, where journalists describe some individuals where they think they are good for the numbers. Quantitave shows the numbers – for example 60% believe that the government does a good job in education. Discussions in focus groups, in-depth interviews, images, social media and much more can be used as data sources for qualitative research.
Also compare here (New Market Research Blog). The qualitative market researcher is an active part of the research process. He or she uses their subjectivity to better understand the phenomenon being studied. You want to understand why people believe in what and see the world a certain way. A qualitative researcher wants to understand the process of selecting a product, for example. A quantitative researcher wants numbers, averages, and more. Manufacturers and dealers can then base their decisions on this.
How qualitative researchers can do representative studies with small numbers
A qualitative researcher makes small samples. Understanding the process and the mindset of the subjects, this researcher assumes some validity. To ensure this this procedure can help:
The result of the small sample are suitable when no changes are to be expected from a larger sample. This can be checked by first evaluating 30 questionnaires. Hopefully there is no significant change with a further 10 questionnaires. If there is, another 10 questionnaires are added. If this does not maintain the hypothesis that the differences in responses are statistically insignificant, the study can be considered representative provided the responding participants represent the target group.
Some subjects suitable for qualitative studies:
- Lists of ideas – if the ideas are repeated, the study can be ended. Studies that do not record the subjects’ ideas via structured questions but via free text information fall into this category.
- All – or – none results. If every participant in a small study says the same thing—like, “I see a train station in this ad,” or “I prefer the new packaging”—the conclusions are likely to be valid, even with small samples.
- Strong hypotheses for the study to support If we have a hunch and it is supported by the small sample, we can be sure that the hunch is correct. All we have to do is verify that the underlying hypothesis is not just speculation. This type of market research is popular with journalists who can get by with just a few interviews.
- Understanding instead of measuring – for example customer journey. This analysis helps testing new products.
The problem of representativeness in qualitative research remains.
With small samples, the problem of representativeness remains. There is no statistical way to ensure the representativeness of a non-probable sample. The market researcher’s judgment must be added here: can the result be as the study suggests or are there other, contradictory results? Trusting only statistically proven results without knowing the underlying connections can be misleading.