I would like to analyse the data of one company. Namely the number of distribution channels used (for one product) in one country and the influence of these channels on performance (e.g. profit). And compare this relation across the different countries in which the firm operates. This comparison would also be possible over time, e.g. 10 years. I don't know which technique is suitable for such an analysis.
Hi Nicola,
I think you don't have to see the "company" as your statistical unit. You don't have a "single case", but different countries with different products, and you want to see if the number of channels is associated to profit.
One profit is associated to i) one channel? or ii) one product? in each country. This will defined your statistical unit.
First, plot what you want to see, and then you will be able to ask the question you want to answer(with a statistical method if needed, namely if you want to predict what will happen with more distribution channels in a determined country)
If you have the complete data then just show it. If you have only some data from some countries, for instance, then you could use statistics to estimate the numbers of channels for those.
All comparisons and their interpretation is not a matter of any statistical analysis.
Hi Nicola,
I think you don't have to see the "company" as your statistical unit. You don't have a "single case", but different countries with different products, and you want to see if the number of channels is associated to profit.
One profit is associated to i) one channel? or ii) one product? in each country. This will defined your statistical unit.
First, plot what you want to see, and then you will be able to ask the question you want to answer(with a statistical method if needed, namely if you want to predict what will happen with more distribution channels in a determined country)
Try to make some interviews in this company, too, if you can: then you can also answer 'why' and 'how' questions as it's not always enough just to look at numbers. Search for the term 'mixed methods'
Thank you very much for your answers! Yes, I try to do a study 1 with interviews to ask why they use this specific distribution channel system and in study 2 I try to test a model to see if the relation between the factors that lead to the channel choice influence performance. Thank you very much Audrey, that's a good point with the statistical unit!
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Structural Equation Modeling