Talk:1991 Murcian regional election

Numbers
Hi! We agree that the Murcian regional elections are tough to handle. I have tried to combine the different sources that we have access to and tried to find some logic behind. In the case of the total registered voters, I have taken for good the turnout numbers from CARM, and with the confirmed valid votes, an approximate number of registered voters per constituency can be calculated. The sum of these numbers is very close to the total registered voters given by Sierra. And as the total number of blank and invalid votes has changed (increased), it would be weird if the numbers to compensate them would be taken only from the abstention. Of course, this is only my interpretation and the sources are confusing. Togiad (talk) 23:40, 6 December 2019 (UTC)
 * I have tried to combine the different sources that we have access to and tried to find some logic behind. While I agree that this may be needed in very dire situations where numbers are very confusing, I'd strongly call against doing this unless absolutely necessary, because it would constitute synthesis, rather seeking alternative and (in the end) simpler approaches. In this case, sources do sometimes appear as more confusing that they actually are: there can be a wide variety of historical references for results, but most of these typically limit themselves to replicating a "core" source (the primary one). In this case, the "Sierra" source seems to be using results from the BORM for the 1991 election (as explicitly stated in page 321, and which you can easily confirm by the results matching). So, while it's true that in page 123 it provides a different figure for registered voters, it comes from a different source, is presented in an entire different and separate way to actual results and seems (according to the author's own word) to partly constitute an "own work". Thus, it may be correct or it may be not, and at this point it comes at conflict with other figures. This, being consistent would imply taking the BORM figures as a whole, which also includes the 778,256 figure, disregarding both the 777,934 one (which can very well be coming from an early estimation; this is typical of Spanish elections) and the 779,296 one (which seems to have been calculated by the author itself combining INE and CREM data). Impru 20  talk 00:16, 7 December 2019 (UTC)
 * For the 1987 election, results for smaller parties at the constituency level seem to have been entirely fucked up in sources, so this will be harder to handle. Impru 20  talk 00:17, 7 December 2019 (UTC)
 * I see your point and I agree with it, although I'm not fully convinced because of the wide number of contradictions between sources. In fact, the turnout figure from CREM coincides with the value obtained from 779,296. However, I would point out that the "own work" for the table seems to be referring to the ratio calculations of the last columns, while the figures from the first columns come from the cited sources.
 * I will check the other elections and hope to get something clear, because it is indeed very messed up. Togiad (talk) 00:40, 7 December 2019 (UTC)