untenable assumptions/types of counterfactualy

[Stefanie Thurm]

Hi to all,

I have a question about the types of counterfactuals according to Dușa 2019/ Dușa 2019 and apologize in dvance for potentially asking a dumb question but I'd be extremely helpful if someone could point me in the right direction.

So, I'm currently reading into the different types of counterfactuals and the implications of SA/(T)ESA. To do so, I've been reading "QCA with R", the 2019 book version as well as the 2024 online resource. When comparing the two, the graphs "The truth table composition" exhibit one major difference, namely, in the 2019 version, there is a small red area "untenable assumptions" located within the easy counterfactuals area which is not the case in the 2024 version. However, the are discussed in the text in both versions.
Now, I've been wondering if the the overlap of easy counterfactuals/untenable assumptions in the newer version is intentionally left out and presents a newer development or if this was unintentional.

I'd be very happy about any opinions. Thanks!

Best

Stefanie 

[Adrian Dușa]

Hello Stephanie,

To be honest, I don't remember the exact reason why I modified the Figure 8.8 in section 8.6.

It probably had to do with my understanding of the untenable assumptions, which I previously thought were possible to be listed in the easy counterfactuals.

Later, when I delved more into how the intermediate solution is calculated, especially for conjunctural directional expectations, I probably realised they can not be found there at all (I stand corrected though).

I find the updated figure in the online version of the book easier to read, the purpose being to understand that not all remainders are alike, and there are significant differences between them that affect the quality of the parsimonious solution.

Either in the published or in the online version, this idea emerges clear: one should be aware the parsimonious solution involves all sorts of impossible, untenable and difficult counterfactuals the contribute to the minimization purposes when in fact they shouldn't. As a result, the parsimonious solution is over-minimized and necessary conditions might be erroneously eliminated.

I hope this explains it,

Adrian

--
You received this message because you are subscribed to the Google Groups "QCA with R" group.
To unsubscribe from this group and stop receiving emails from it, send an email to qcawithr+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/qcawithr/6bdacde1-7351-45a5-8230-bf580721eb97n%40googlegroups.com.

[Stefanie Thurm]

Dear Adrian,

thanks a lot for getting back to me on that question. Your explanation is very, very helpful! :-)

All the best and thanks again

Stefanie