conservative and parsimonious solution

[Sarah Wang]

Hi everyone, 

Could I please ask a question about conservative solution and parsimonious solution? 

This is the result that I produced- see below and also attached (the first sol_ny is the conservative solution for the negated outcome; the 2nd sol_nyp is the parsimonious solution for the negated outcome )

My question is: 1. Does it look normal to you? The parsimonious solutions look more complicated to me (3 solutions), compared to the conservative solution (1 solution). But the combinations of conditions are simpler.  

2. will you tend to use the conservative solution or parsimonious solution? why (if they all make sense theoretically due to the lack enough evidence in theoretically). 

Thank you

Kind regards

Sarah 

> sol_ny <- minimize(TT_ny,
+                    details = TRUE,
+                    use.tilde = TRUE)
> sol_ny

M1: ~comp*~selec*~track*~inequa*~edmat + ~selec*~track*~inequa*edmat*~privat + comp*selec*~track*inequa*~edmat*~privat -> ~segr

                                            inclS   PRI   covS   covU   cases
----------------------------------------------------------------------------------------
1       ~comp*~selec*~track*~inequa*~edmat  1.000  1.000  0.334  0.224  FIN,NOR,SWE; DNK
2      ~selec*~track*~inequa*edmat*~privat  1.000  1.000  0.200  0.200  ISL; IRL
3  comp*selec*~track*inequa*~edmat*~privat  1.000  1.000  0.177  0.067  KOR
----------------------------------------------------------------------------------------
                                        M1  1.000  1.000  0.601


> sol_nyp <- minimize(TT_ny,
+                     details = TRUE,
+                     include = "?",
+                     use.tilde = TRUE,
+                     row.dom = TRUE)
> sol_nyp

M1: ~selec*~inequa + (comp*selec*inequa*~edmat) -> ~segr
M2: ~selec*~inequa + (comp*~track*~edmat*~privat) -> ~segr
M3: ~selec*~inequa + (comp*inequa*~edmat*~privat) -> ~segr

                                                    --------------------------
                               inclS   PRI   covS   covU   (M1)   (M2)   (M3)   cases
----------------------------------------------------------------------------------------------------------
1              ~selec*~inequa  1.000  1.000  0.777  0.490  0.534  0.600  0.534  FIN,NOR,SWE; DNK; ISL; IRL
----------------------------------------------------------------------------------------------------------
2    comp*selec*inequa*~edmat  0.929  0.507  0.288  0.000  0.045                KOR
3  comp*~track*~edmat*~privat  1.000  1.000  0.222  0.000         0.045         KOR
4  comp*inequa*~edmat*~privat  1.000  1.000  0.288  0.000                0.045  KOR
----------------------------------------------------------------------------------------------------------
                           M1  0.974  0.918  0.822
                           M2  1.000  1.000  0.822
                           M3  1.000  1.000  0.822 

[Adrian Dușa]

Dear Sarah,

The parsimonious solution is not more complicated, it has more ambiguity, which is another thing entirely.

For the second question, I wonder what is your reason to want those solutions for the negated outcome. If you do it in the context of (T)ESA, to identify contradictory simplifying assumptions, or to identify and exclude remainders containing negations of necessary conditions, those are examples of when and why to employ the analysis of the negated outcome.

Otherwise, it all depends on what you want with it but if you're simply interested in the negation of the outcom per se, this is not different from analysing the presence of the outcome.

Preferring the parsimonious solution over the conservative really depends on what your research purposes are. The parsimonious solution however always employs a good number of difficult counterfactuals, so you might want to be aware of what QCA theory indicates in relation to this.

I hope this helps,

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/800695f4-8879-436d-ab43-1b76f1d66c74n%40googlegroups.com.