Replications in business administration – potential for more scientific reliability

Sören Köcher on his Open Science experiences

Together with Susanne Adler and Marko Sarstedt, you decided to publish the Special Issue: Replications in Marketing and Consumer Research in the “Journal of Business Economics”. What was the reason for this?

SK: With this special issue, we would like to address an important topic that has become increasingly important in many scientific disciplines in recent years: the replicability of research results. In science, replicability is considered a key quality feature – it is crucial for the credibility of and our trust in existing findings. The observation that many scientific results cannot be confirmed in replication studies emphasises the relevance of such studies – also in our field.

However, replication studies have so far played a rather subordinate role in our discipline, although new research is based to a large extent on existing findings. This makes it all the more important that these findings are robust and reliable. With this special issue, we would like to give this topic more visibility and create a platform that contributes to the recognition of replication studies and encourages researchers to carry out and publish corresponding work.

Was there a specific trigger or development that played a role in this?

SK: Replication studies are hardly established in marketing and consumer research – in contrast to disciplines such as medicine or psychology, where replication is an integral part of scientific practice. With this special issue, we would like to provide an impetus and create an outlet that contributes to the establishment of replication studies in our discipline.

What kind of replications do you specifically invite?

SK: We are interested in various forms of replication that can contribute in different ways to the verification and further development of existing findings: Firstly, direct replications with extensions. This involves reproducing the original study design as exactly as possible, but adding additional variables or conditions. Such studies make it possible to identify factors that can change the strength or direction of an effect, or to test alternative explanatory approaches and thus better understand the underlying mechanisms of an effect.

We also invite conceptual replications. The focus here is not on the exact repetition of the original design, but on the investigation of an effect in a different context – for example by using other dependent variables or conducting the study in a different setting. It is particularly interesting when effects that were previously only investigated under laboratory conditions can also be observed in the field . Such studies not only test external validity, but also emphasise the practical relevance of the findings.

In marketing and consumer research, studies cannot be replicated indefinitely – simply because human behaviour changes over time. Consumers today, for example, inform themselves via tools such as ChatGPT or Gemini before making a purchase decision. How old can a study be before contextual changes jeopardise its replicability?

SK: The age of a study is less important to us than its relevance. We are particularly interested in studies that were and are theoretically influential and on which current research is based – regardless of when they were published. Particularly in consumer research, which has a strong experimental orientation, we observe that many older studies worked with small samples – in some cases well below what is considered the minimum methodological standard today. This is particularly problematic if the reported effect sizes are low anyway. In such cases, there is a risk that effects have been overestimated or incorrectly assessed. Replications offer the opportunity to check the robustness of the findings under contemporary methodological standards.

Assuming that you receive numerous interesting and differently designed replication studies for the Special Issue, what other measures could help to anchor the topic of replication more firmly in the economics community in the long term?

SK: In my view, a key aspect is raising awareness: there needs to be a clearer understanding of the functions that replications fulfil and the added value they bring to science. On the one hand, replications strengthen confidence in scientific observations and effects by testing their robustness. On the other hand, they expand our understanding of the conditions under which an effect occurs – or does not occur. In particular, replications with extensions can provide valuable insights, for example by showing which factors moderate an effect or which theoretical limitations exist. They can also contribute to a more differentiated understanding of the underlying mechanisms of action. Testing effects in different contexts – for example in the field rather than in the laboratory – can also provide important information on practical relevance. Another contribution lies in the scientific debate itself: I myself published a replication study with which we responded specifically to a debate within the literature. We replicated a study from the Journal of Consumer Research, which provoked a number of comments after publication. With our replication, we wanted to empirically clarify some of the open questions that were only discussed theoretically there. This is another way in which replications can add theoretical value.

You have a dual role: on the one hand, you are co-editor of the Special Issue and, on the other, you are active in behavioural and consumer research yourself. So you are familiar with incentive mechanisms. From this dual perspective, how could the incentive for researchers in business administration to focus more on replication studies be increased?

SK: From my own experience, I can say that replication studies are a very good way for young researchers in particular to get started in publishing. They provide a clear methodological framework and make it easier to deal with a specific research question in a structured way. This can be particularly helpful for gaining initial publication experience.

What do you think would be a success for this special issue?

SK: In my view, a good success would be if we can put together a balanced and diverse mix of different forms of replication – as I described earlier. If, when looking at this collection, readers recognise that replications do more than just repeat previous studies, namely that they also enable independent theoretical contributions, then an important goal would have been achieved. Ideally, the special issue will motivate other researchers to carry out replication studies themselves.

This seems to be a central aspect. In scientific practice, replication research is often perceived as subordinate – among other things with the argument that the actual intellectual achievement lies with the first authors of the original study, while a replication merely repeats its structure. Against this background, the question arises: To what extent can replication research, especially when supplemented by extensions, still be understood as an independent scientific contribution?

SK: That is indeed a key point. Replication research is often underestimated. It is usually overlooked that replications – especially with extensions – can make an independent contribution: They help us to better understand an established effect. Of course, in many cases the theoretical contribution is different from that of a study that identifies a completely new effect. This is probably one of the reasons why replication studies in our discipline have so far been less frequently included in scientific journals. Nevertheless, it is important to understand established effects on which current and future research is based as well as possible. Even smaller contributions can be valuable here – for example, if they show under which conditions an effect does not occur or is particularly pronounced. Such findings can in turn be helpful if later studies build on these results.

Ultimately, the testing of effects also plays a role in practice. Precisely because marketing and consumer research is highly relevant for companies, start-ups, NGOs and other user-orientated stakeholders, it is crucial that research results are replicable and robust.

SK: Exactly. Our discipline is strongly application-orientated and delivers not only scientifically relevant but also practical findings. This makes it all the more important that our recommendations for business practice are based on robust and verified results. Conversely, companies benefit when they know under which conditions a certain effect reliably occurs – or not. This strengthens both the scientific validity and the practical usability of our research.

What role do replication projects play in your teaching and to what extent do you think they are suitable for teaching scientific skills to students?

SK: We use replications specifically in teaching to introduce students to scientific work – for example in dealing with data, applying empirical methods or critically analysing existing studies. Replication projects are very suitable in this context, even independently of a publication. I am also currently experimenting with a slightly different approach with my students: we are trying to replicate known effects using synthetic data – i.e. with data sets that we generate via ChatGPT. The aim is to find out to what extent such artificially generated data can be used to simulate realistic response behaviour and whether established effects can be replicated.

How did you become sensitised to the topic of Open Science? My impression is that it is neither firmly anchored in university education nor systematically anchored in the doctoral phase. When and how did you first come into contact with it?

SK: My first concrete contact with Open Science was about five or six years ago. At the time, I was working with an American co-author on a joint project. In the course of our collaboration, he suggested preregistering our studies and uploading the data to the Open Science Framework to give reviewers access.

Open science didn’t play a role in my own doctoral training – it wasn’t an established part of the academic qualification back then. However, I have the impression that this is increasingly changing. I am convinced that open science will be more strongly integrated into structured doctoral training in the future.

If someone is new to the topic of Open Science – an area that is quite complex and extensive – what specific recommendations would you make? From your experience, what would be sensible first steps?

SK: I think a central obstacle often lies in a certain inertia towards the implementation of open science practices. However, many measures can be realised with little effort. A good and pragmatic first step, for example, is the pre-registration of studies. It is worth thinking about how you want to analyse the data later before you start collecting it. I have to make these considerations in the course of the analysis anyway – so why not record them in a structured way right from the start? Another simple step is to share materials and data via platforms such as the Open Science Framework. This is also technically straightforward and creates long-term transparency and traceability – not only for others, but also for your own work. If I structure and document my data clearly from the outset and archive it properly, I also benefit from it later – for example when I want to access the data or further develop a study.

One last question to conclude: Where do you see the topic of Open Science in marketing and consumer research in the coming years?

SK: I am convinced that open science practices will become much more important in our discipline. The first developments can already be seen: Some of the leading journals have already led the way here and now require the disclosure of datasets and materials in order to ensure the traceability, transparency and replicability of research. I firmly believe that this practice will become more widespread and will be a central component of good scientific standards in the future.

Thank you very much!

*The interview was conducted on 4 June 2025 by Dr Doreen Siegfried.
This text was translated on 18 June 2025 using DeeplPro.

About Dr Sören Köcher:

Dr Sören Köcher is currently a deputy professor of marketing at the Technical University of Dortmund. His research focuses on consumer behaviour in digital contexts, decision anomalies, sustainable consumption and replication studies. International research visits have taken him to York University, Texas State University and Florida Atlantic University, among others. He has received several awards for his scientific achievements, including from the American Marketing Association, the Academy of Marketing Science and the NRW Consumer Advice Centre.

Contact: https://marketing.wiwi.tu-dortmund.de/professur/team/soeren-koecher/

LinkedIn: https://www.linkedin.com/in/soeren-koecher-26777249/

ResearchGate: https://www.researchgate.net/profile/Soeren-Koecher

OSF: https://osf.io/rdmgw/

About the Special Issue: Replications in Marketing and Consumer Research in the “Journal of Business Economics”

to Open Science Magazine