From myth to added value

Too small, too unscientific, too cumbersome. Although widely used, the prejudices against qualitative research are numerous. Artificial intelligence will not change this either. Ines Imdahl and Jens Lönneker from Rheingold Salon, both graduate psychologists and passionate qualitative researchers, thoroughly dispel the myths surrounding qualitative research.

Anyone who conducts qualitative research is confronted with a number of prejudices in their professional life: The research is probably not really scientific. Because of the small number of cases. At best, it is suitable for generating hypotheses. These theses then require quantitative or representative verification, no matter how exciting or convincing they may be.

It is not uncommon for the qualitative part of the research to be smiled at somewhat or at least not taken quite so seriously. Motto: "Yes, yes, we'll do that too - just for a moment - with." No statistics, no numbers, just openly ask for opinions and put together what people have said - and the research is done. Incidentally, behind such an approach there is often the assumption that qualitative market research can easily be acquired on your own. As with marketing and psychology, most people believe they are experts per se. Some large research institutes even train their qualitative researchers in just two to three days. For comparison: At Rheingold Salon, this takes up to four years.

OUR MOST IMPORTANT MYTH BUSTERS

1. Scientific rigor depends not on the number of cases, but on the theory and systematic evaluation.

2. Quantitative research cannot verify qualitative research, as both approaches solve different tasks and answer different questions. The approaches are usefully complementary.

3. Qualitative research is not an afterthought, but must be learned and practiced with great effort - it is not an easy path and requires time, training and a lot of basic know-how

4. AI is helpful if the theory-guided systematization approach is mapped by the evaluation program - otherwise, AI evaluation is no better than "gut feeling", just less emotional.

Yes, and then there is now also the AI, which can take over the lengthy evaluation of open questions, delivers insights at the touch of a button and brings a previously unknown calculability into qualitative research. Let's take a closer look.

1. Good qualitative research follows a scientifically based theory

---

The myths surrounding qualitative research are based above all on the constant attempts to differentiate it from quantitative research. Numbers seem somehow "more reliable" than the word or the description. However, the cognitive value and scientific nature of a procedure depend in no way on the type of data, i.e. qualitative or quantitative, but on whether a theory-guided and systematic, learnable implementation and evaluation is possible.

By the way, the debate surrounding both types of research was already conducted in Germany in the 1950s and 60s: Subjectivity, non-reproducibility and the alleged unscientific nature as reservations about qualitative research were countered by the accusation of meaninglessness, lack of connection and irrelevance of data mountains in quantitative research. The sometimes somewhat headless hoarding of data has reached a new peak with big data today.
The increasingly used AI-supported tools do not replace the classification into meaningfulness and context. Understanding the meaning for the human being is at the heart of the design and evaluation of qualitative research - an AI does not simply bring this with it, it has to learn it, just like the researcher.

The key is: There is a theory underlying both the research design, the development of a question guide and the evaluation. This also applies in every way to the digital space. The theory can be named, its methods and evaluation criteria can be learned. If only open questions are used and the answers are put together loosely or according to gut feeling, then this is not yet qualitative research. Depending on the respective theory, research requires years of practice and learning in order to be able to describe, analyze and systematize the wealth of everyday data as comprehensively as possible.

2. Qualitative research has a variety of scientific systematization models as its origin

---

What is summarized under qualitative market and media research comes from completely different disciplines - models are drawn from scientific disciplines, such as psychology, ethnography, linguistics or sociology. The heterogeneity of the approaches is further reinforced by the fact that even within a science, different models are used in qualitative research. For example, in psychology: Here, assessments are based on depth psychological, cognitive psychological, neuropsychological models. Even within depth psychology, there are variants that are more strongly based on Jung or Adler, are systemically oriented or operate with a modern, independent approach such as morphological psychology.

In addition, there are approaches such as semiotics, which is important in the academic environment across disciplines from philosophy to linguistics and social sciences. A good overview of the spectrum is provided by the handbook "Qualitative Market Research in Theory and Practice" by Gabriele Naderer and Eva Balzer.

THE ADDED VALUE OF QUALITY RESEARCH

1. Everyday experiences of people become comprehensible, meaningful and tangible for companies, advertisers, marketers and decision-makers.
2. Complexity and diversity are retained and provide relevant details for targeted customer approaches.
3. Explanations create context and meaning - this applies internally to the company as well as to the external customer approach.
4. Qualitative research often uncovers structures overlooked in everyday business, which help to expand perspectives in a meaningful way.

To be clear: The mere use of digital or AI-based evaluation tools does not constitute an independent systematization approach. This is because the models are based on different systematizations, which must be incorporated as a complex system into the AI or deep learning models used. If the AI simply evaluates language or other general syntax - this is neither objective nor scientific. This is because data is merely collected here and then put together "arbitrarily".\nThis is then a pseudo-systematization based on the gut feeling of the programmers, who are often not even market researchers.

3. Meaning & context - explaining and understanding are characteristics and added value of qualitative approaches

---

But how is it that such heterogeneous scientific disciplines can compete in one and the same field? What is the common and connecting element? And does it even exist? The economist and social scientist Prof. Dr. Gerhard Kleining characterizes the special feature of qualitative approaches as follows:

• Relevance to everyday life in terms of topic and methods
• Openness of the research staff when dealing with the research subject
• Extensive field research
• Inclusion of the researching subjects in the cognitive process
• Low abstraction of the data - preservation of its meaning
• Variation of methods, including quantification (one would like to add digital, AI-based models)
• Adaptation of the procedures to the research subject
• Analysis by comparing the data, searching and finding commonalities
• Presentation of concrete results, not just a discussion of methods
• Theoretical ability of the results

The attractiveness of qualitative research therefore lies in the fact that it preserves phenomenal connections in their complexity and, despite a theory-guided approach, enables great openness and flexibility in the research process. In research aimed at determining quantitative data, on the other hand, a higher degree of standardization and abstraction is usually sought, which hardly allows immediate access to these original complex connections.

Progressive digital transformation easily suggests the thesis that qualitative research is also becoming more predictable and objective as a result. However, this is debatable in two respects: Firstly, qualitative research is no less objective than quantitative research (provided it is theory-guided). And secondly, digitalization - when used optimally - is an acceleration, but it does not provide an explanation, but evaluates according to the programmed specifications. It does not create the context, but initially only more data.

By systematically seeking to capture connections and meaning, qualitative research has a methodological goal, is more than a pure data form and is gaining a new, even more relevant meaning, especially in times of meaning and purpose development of companies.

The important thing here is: Quantitative research does not verify qualitative findings. It cannot do that. Prof. Gerhard Kleining writes: "Qualitative investigations without quantification are possible; of course, one can remain on the more complex qualitative level. In contrast, purely quantitative investigations are risky if the qualities underlying the quantities or frequencies are not sufficiently known." And further: "The two levels of abstraction cannot be used for mutual verification, especially not the qualitative by the quantitative."

by Ines Imdahl & Jens Lönneker, published in planung&analyse on 23.06.2023

available at Qualitative Forschung: Vom Mythos zum Mehrwert – planung&analyse (horizont.net)