Market analysis project: from problem decomposition via analysis of raw data to higher-order recommendations (NN2)

Learning consultants’ reflection on NN connections of this activity 

NN2 highlights that the way we analyse data can impact the outcomes. This capability can be developed by relating the analytical findings to a higher-order problem and reflecting on how our conclusions were affected by the analytical methods we used. 

In this example, students integrate the knowledge from different courses to solve a broadly formulated problem, which requires both strong analytical skills and the ability to move to higher levels of abstraction. Students get to select appropriate methods of market research analysis in order to make sense of the available data and then formulate a recommendation on how to approach the given higher-order problem in a language that is easy to understand. The multitude of possible recommendations highlights how choices we make as analysts can significantly impact the resulting solution. 

Teaching philosophy 

 ‘Analytics is not just about decomposition. It is about decomposing in such a manner that you can build up again to a higher level of abstraction. You cannot just say, ‘We want to be sustainable’ or ‘We want to be climate friendly’; we need to ask what that is. If I have a company that wants to be more sustainable, I need to identify concrete analytical issues that I should then manage. But when I get insights into these, I need to figure out how they fit together and how I can move back up the ladder and show my customers that I am a sustainable company. So, the analytics and the higher level of abstraction go hand in hand, in my opinion.   

What we do with the students is that we integrate actual problem settings, where they have first to decompose the problem and then increase the level of abstraction afterwards. We tell them that the dataset should be seen as an information basket, and it’s them, the students, who should transform this information into knowledge based on the competences they have acquired during the course.’ 

Torben Hansen, Professor at the Department of Marketing 

Key objective(s) aligned with this activity 

• Identify and develop relevant marketing research problems, marketing research questions and hypotheses based on a marketing decision problem 
• Develop a relevant analysis design that will be suitable for a given marketing research problem 
• Evaluate and select appropriate methods for data collection and sampling 
• Describe, evaluate, select and apply market research techniques 
• Assess and communicate the solution of specific marketing research questions and marketing research problems independently and on a reflective scientific basis 

Description of the activity:

The course is organised in two parts: qualitative and quantitative. The quantitative part takes up about 80% of the course, based on the experience that students find it much more challenging to learn about the quantitative part than the qualitative part. 

Most of the teaching is organised by first having a 3-hour lecture and then exercises, where the class is divided into two groups. Both lectures and exercise classes are complemented by videos that offer students asynchronous access to all the course content. The exercise classes then become the space for dialogue with the teachers, which students appreciate and find necessary for their progress, as proven by attendance rates. 

The exam comprises two parts. In the Fall semester, students are asked to build a theoretical model that integrates the elements from different courses. For example, this year (2022/2023), the focus was on consumer behaviour and the factors determining consumers’ choice of supermarkets. In the second part, students get a dataset in SPSS and are asked to modify their model to reflect the answers from the data file. The students need to detect the relevant constructs that can be identified in the data file using the methods they had been taught. They are asked to use at least 3 of the methods and report their findings as recommendations to the problem owner. This means that the students must demonstrate their skills working with the various phases of the market analysis process to solve a higher-order problem requiring high levels of abstraction.