 In qualitative research we typically want to also make causal claims. However, the logic how we make causal claims in qualitative research is quite different from quantitative research. To understand why that's the case and how we make causal claims using qualitative data, it's useful to start from how we make those claims using quantitative data. So when you make a causal claim using quantitative data, you have to demonstrate generally three conditions. First you have to demonstrate that the cause X and the effect Y are associated. Then you have to demonstrate that the cause comes before the effect and not the other way around. And then you have to eliminate any rival explanations. For example it could be that X and Y have a common cause and therefore they are correlated. This is fairly straightforward to do except for elimination of the rival explanations because there are some general strategies that can be applied. So to demonstrate the direction of influence we typically measure the cause before the effect and that in most cases takes care of the problem. To eliminate rival explanations we have two strategies. We have the strategy of randomized experiment and the strategy of statistical controls. There are also others but these are the two most common ones. The idea of randomized experiment was that because you randomized people into treatment and control, there cannot be any differences before the treatment and only the differences after the treatment must therefore be because of the treatment. The idea of statistical controls was that you step into the shoes of the skeptic who doesn't believe that your XY correlation is evidence of causality and then you consider what kind of counterarguments that person would make and then you model those counterarguments with control variables. So for example if we claim that CO-gender causes profitability such that women-led companies are more profitable than men-led companies and we attribute that to the causal effect of women being better. Leaders, then we need to think about alternate explanations such as certain industries being more likely to hire women and those industries also being more profitable for some reasons. In qualitative research you don't see much of this. In some articles you see that association so you can see that some cases typically you would have maybe 4-12-20 cases. You could see a table that demonstrates that the cases that are high on X are also high on Y but that doesn't really qualify as an association because the sample size is so small that we can't rule out chances and explanation for that association. The direction of influence also is not explicitly mentioned and I've never seen a qualitative data analysis or article using qualitative data discuss control variables or do randomized experiments. So how exactly do qualitative studies make causal claim? To understand that we have to understand that causality is not three conditions. Causality is a process, it's a process through which X influences Y and if we now take a look at these three conditions here the question is where exactly is the causal process in these three conditions? So basically it's not there. We are only looking at X and Y and ruling out alternate explanations but we are not looking at the causal process itself when we make causal claims using quantitative techniques. In contrast, in qualitative techniques this causal process is something that must be explicitly present. So the idea of in quantitative analysis is that the causal process is a black box, we don't touch it we only observe the inputs and the outputs and then we infer that there must be a process between but we don't really see the process. In qualitative data analysis, how we do it is completely the opposite. We don't focus on X and Y and their association. Instead we focus on explaining and describing the causal process. If you can convincingly argue that there is a process through which X causes Y and you can demonstrate that that process plays out in your data or has played out in the data in the past if you do a retrospective case study then you have a pretty solid way of making a causal claim. So if you can argue that I saw that the causal process played out in this particular way there cannot be any other explanation for the data. So if you see that something happens and then you document it well that's a very solid causal claim. You don't need to care about the three conditions for causality which are only for quantitative research because quantitative researchers are limited to looking at causality or causal process as a black box because we only have like these numbers that quantify static snapshots over time. So what will happen if you ignore the causal process? If you do a qualitative study and you merely say that X is high in some cases Y is high in those other cases, X is low in some cases, Y is low in other cases. I have some first-hand experience in doing that and the outcome is generally is not positive. So if you ignore the causal process and explaining how and why something happens and you do a qualitative study then the reviewers and the editors of your study are forced to basically evaluate your study which in our case had six cases against the three conditions. And if you look at can you make a convincing claim that you have a robust association between two variables using six cases, the answer is no. Also you can't really rule out alternate explanations because you don't have any control variables. And this is basically what Edyker told us when we tried to submit a paper that did not explain the causal process. So it's not sufficient that you say that X and Y are associated. You really need to dig into the black box and explain the causal process like what steps does it take from getting from X to Y and then that gives you your new theory and your new contribution.