Dees de Jong

1888 Dutch general election

Hello. Many thanks for adding more detailed results to this article. I assume you compiled the results constituency-by-constituency from the Huygens Instituut websuite. Are you able to paste those onto the talk page so other editors see the calculations? Several of us do this as good practice, so the calculations done by editors can be double-checked (see e.g. here or here).

It would also be helpful if you could break out the 'Other parties' into the individual parties (assuming they are broken out in the source too). Cheers, Number 57 17:33, 24 August 2024 (UTC)

Hi hi, yeah I did compile the results from Huygens. Usually other parties are just independents or parties running in just 1 constituency, that I don’t know the full names of as they are ancient local parties and the Huygens archive only uses abbreviations. I’m on vacation for the weekend so don’t have access to a computer. I could probably put the calculations on the talk page later this week. For the meanwhile I’ve compiled the data in a spreadsheet you can check out. https://docs.google.com/spreadsheets/d/1J6yBDpZpQVt0UefDE9ijJX2D1Qykd4WGwtTOzoCBFhA/view Dees de Jong (talk) 18:42, 24 August 2024 (UTC)

Great, thank you. It looks like you have divided the votes received in some constituencies by the number of seats available? Number 57 21:09, 24 August 2024 (UTC)

Yes, everyone could vote the same amount of times as there were candidates in their district. Every voter in Amsterdam for example had 9 votes. Not dividing the votes by the number of votes every voter had would result in a total turnout of over a 100% and using the data from Huygens would result in a negative number of invalid votes. I think dividing the vote (effectively showing the amount of people supporting a party) is the best way to represent the election. The old source did this as well, but let me know if you feel differently! Dees de Jong (talk) 09:54, 25 August 2024 (UTC)

Counting all the votes doesn't mean a voter turnout of over 100% – many countries allow multiple votes and it is displayed like here or here. Dividing votes by the number of seats is a dangerous bit of WP:OR IMO as not all voters will use all votes.

The remaining question is how you have arrived at the invalid vote totals. I am a bit confused by where the "Votes" figure in column I comes from – it is not a total of valid votes nor a total of valid + blank. It also can't be valid + invalid as if it was, valid + invalid + blank does not equal the total in column L.

For example, for Zudhorn, the total number of votes for candidates is 2,787 and the number of blank is 2. The 'Votes' figure in column L is 2,795 and the total is 2,800. If invalid were 3, then the actual total is 2,792. If the total votes cast is 2,800, invalid votes must be 11 (assuming the votes for candidates and blank is correct), not 3... But where does that leave the 'Votes' figure?

There are also a few other errors – for example, the total number of votes for candidates in Wijk bij Duurstede (3,114) is higher than the number of votes cast (2,938).

I would double check myself, but for some reason the 1888 elections don't show up when I try to access them... Number 57 17:42, 25 August 2024 (UTC)

And re the 1894 elections, the spreadsheet doesn't seem to contain the original vote figures (before dividing by the number of seats) for the multi-seat constituencies. Could you add them?

Separately, I've noticed another major issue with the figures – rather than use the total for 'Other' candidates, you seem to have used whatever number you needed to get the vote total to equal the number of votes cast? Number 57 21:56, 25 August 2024 (UTC)

Ok I had a very busy week, so let's go through everything.

I first want to note I do appreciate your attentivenes and help. I want these pages to be the best they could be, so your close eye helps a lot.

I'll undo the deviding of votes by number of candidates. and I'll add them to the other elections.

The votes colum is taken from the "Aantal stemmen geldig" or in english "Total valid votes"row in the Huygens database. And I've derived at the invalid figure by simply taking Valid and Blanco votes out of the total votes cast. These figures are all listed in the database of Huygens.

This discrapency between valid votes cast listed on Huygens are also the reason the Other's vote total is whatever number needed to match the vote total. I assumed the data did not account for small independent candidates and thus I threw those missing votes onto the Other pile.

In the case of Wijk bij Duurstede it was a typo. I typed 965 instead of 765 when putting everything in a spreadsheet, because sadly Huygens doesn't provide a spreadsheet with the results. I appreciate you catching this and will fix that.

I had already contacted the kiesraad about the missing elections on their site. They contacted me back stating that these indeed were not available, and there is no timeframe in which they will become available again. I've used the Huygens instutuut numbers for the data, if you want to double check my work you can do so here. If you have any other changes you'd suggest I'm happy to defer to your experience and change how I've used the data.

Cheers, Dees de Jong (talk) 15:17, 29 August 2024 (UTC)

Ah, so the problem is that the candidate totals and valid votes are inconsistent in the source? For example, for Zuihorn for 1888, there were four candidates who received a total of 2,787 votes. However, the source claims there were 2,795 valid votes, 2 blank votes and 2,800 votes total.

Also, for constituencies with multiple seats, how were blank votes worked out? Was it an entirely blank ballot, or does it take into account votes not cast (for example, voting for one person less than the number of seats available). The 3,527 blank votes for Rotterdam out of a total of 5,938 ballots cast seems very high, so I suspect it may be uncast votes? Number 57 19:54, 29 August 2024 (UTC)

Yeah, my assumption is that those missing votes were cast on minor candidates not recorded in the data. And yeah blank votes are definitely measured by taking into account votes not cast Dees de Jong (talk) 18:21, 31 August 2024 (UTC)

Hmmm, I think that is a dangerous assumption to make unless we have evidence that there were other candidates. Number 57 21:55, 4 September 2024 (UTC)

Hey I'm trying to get back into updating this. If I ignore the totals given by the source and only use the votes measured, will the vote totals then be acceptable to put on the page? Dees de Jong (talk) 16:29, 8 October 2024 (UTC)

It would be good if we could get to the bottom of the discrepancy between the candidate votes and valid votes. Might there be some newspaper sources from the time that give full results for a constituency or two? I know the Delpher online newspaper library that has been very useful for 19th century Surinamese elections. Number 57 00:25, 9 October 2024 (UTC)

I looked into it and my initial assumption of Huygens not including minor candidates in the table, but including them in total vote count was correct. This paper mentions that there were more candidates than Huygens mentioned in the 1888 Elections in Winschoten, but that these got a negligible amount of the vote. The vote totals do match.

Paper: https://www.delpher.nl/nl/kranten/view?query=verkiezing&facets%5Btype%5D%5B%5D=artikel&facets%5Bspatial%5D%5B%5D=Regionaal%7Clokaal&page=1&cql%5B%5D=%28date+_gte_+%2206-03-1888%22%29&cql%5B%5D=%28date+_lte_+%2210-03-1888%22%29&coll=ddd&redirect=true&identifier=MMRHCG03:163381030:mpeg21:a00001&resultsidentifier=MMRHCG03:163381030:mpeg21:a00001&rowid=1

Huygens: https://resources.huygens.knaw.nl/verkiezingentweedekamer/databank/zoek_district/uitslag_per_verkiezing?uitslag_ID=1492

I think it is safe to say we could add the vote figures back in Dees de Jong (talk) 10:52, 17 October 2024 (UTC)

I do actually think that not dividing the votes of those in multi member constituencies skews with the results and the fact that not all voters will use all votes is reflected in the turnout for the constituency. I believe dividing the figures gives a better reflection of the election and the amount of people voting for each candidate. Dees de Jong (talk) 11:05, 17 October 2024 (UTC)

Actually nevermind you're right I looked into some other pages that have the same thing going and on the Luxembourg pages it works fine Dees de Jong (talk) 11:16, 17 October 2024 (UTC)