Talk:Opinion polling for the next Czech parliamentary election

Phoenix Polls
Hi all I am looking at some of the polling data and I've got concerns that phoenix don't meet usual standards of transparency amongst pollsters. They don't seem to publish their results anywhere other than on a news site, and the information released by that news site is very limited. As an example, they bundled 'other' and 'don't know' votes together, which throws off the data as the don't know figure is necessary to recalculating the numbers. They also don't release a sample size or specific fieldwork dates some of the time, and the most recent poll the news site suggested the poll had 'over 1000' respondents. Hence, I would suggest that for the time being or until we can find a new source, Phoenix polls should be removed from the graph; I am going to remove them from the graph because the data cannot be properly recalculated or dated. cc @Braganza @Bedivere.cs @Nadvšenec Quinby ( talk ) 10:20, 29 April 2023 (UTC)


 * I am neutral and support anything you agree on Braganza (talk) 10:30, 29 April 2023 (UTC)
 * Hi
 * I would agree with this. Phoenix doesn't release any documentation for the polling - in fact, for the April poll, they haven't released anything. Also, they are not member of SIMAR, unlike the other big pollsters.
 * Its ownership is also shady (https://zpravy.aktualne.cz/domaci/phoenix-research-predpovida-uspech-trikolore-patri-prodejci/r~7b72a3c8952e11eaa7deac1f6b220ee8/), and they're polling widely differently (and wrongly) from the other pollsters. I would remove them from the overview, and maybe consider doing the same with SANEP, which has similiar issues. Kroulacek (talk) 10:53, 4 June 2023 (UTC)
 * Hi, It is true that Phoenix (and also SANEP) is a problematic pollster but I wouldn't remove them from graph completely but rather suggest to move them to a newly made section which would include explanation. I'm not certain what ould be the name of the new section. Still if these polls are removed I don't have problem with it. Bedivere.cs (talk) 05:11, 12 July 2023 (UTC)
 * My standpoint is similar, it is not unusual for opinion polling articles to include polls with unknown fieldwork dates, approximate numbers of respondents or unclear differentiation between others und don't know (I had to deal with such cases a lot here for the graph). Nevertheless, if all problems occur together, while all other pollsters don't have such issues, I would at least agree with omitting them from the graph. Gbuvn (talk) 08:34, 12 July 2023 (UTC)