Jump to content

Talk:2012 Japanese general election

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Polling Issues

[edit]

Three things I wanted to address:

  • 1) With respect to the second Asahi Shimbun poll in party polling, on reading the story that seemed to be the party people most wanted to gain seats, not their indicated voting preference (which was noticeably different).
  • 2) With respect to the first Yomiuri Shimbun poll in party polling, the JRP numbers seem to be listed as the combined total of the support for the Japan Renaissance Party and the Sunrise Party of Japan, even though these two were polled separately.
  • 3) Per looking at the articles in the newspapers, it seems that the polling to date as indicated in the graphs is quite flawed; in particular, the JRP seems to have, per Yomiuri Shimbun, polled much better in several polls.

I'm not sure how to deal with these issues, but they bear addressing.Tyrenon (talk) 10:39, 21 November 2012 (UTC)[reply]

As for the long-term monthly NHK poll graphs, suggestions are very welcome. I've used the NHK poll because 1. the data is easily accessible reaching back about a decade and 2. unlike the major newspapers, the NHK is not as clearly associated with a political viewpoint. And if you compare polls by various news organizations (e.g. on real politics) at least I can't make out a particular bias. The English Yomiuri article shows a different poll, not party support/approval (shijiritsu), but voting intention in the 180-seat proportional race. So, they cannot be compared. --Asakura Akira (talk) 11:11, 21 November 2012 (UTC) P.S.: A timeline graph for the comparable cabinet and party approval questions in the Yomiuri Shimbun poll: http://www.realpolitics.jp/research/yomiuri.html[reply]
Clarification: You can argue about NHK bias. Or write books or blogs about it. What I should have said is: As a public institution, it cannot or should not take sides as openly as others. But as for the polling – and that's the issue here –, I don't think it makes much of a difference. As I said, if someone has a better suggestion for the long-term cabinet/party approval graphs, go ahead. I started those graphs before the last election when there was nothing to give non-Japanese Wikipedia readers a general impression of how (un)popular recent cabinets have been. --Asakura Akira (talk) 12:57, 21 November 2012 (UTC)[reply]
    • First of all, I want to thank you for both the link and the information below. Not being Japanese nor speaking Japanese, what is on here is often all I can find on politics and polls in Japan. Also, I have no problem with using the NHK data...but I also don't know enough to comment on any paper's bias in Japan. With that said, the link is hard for me to interpret because it is in Japanese and...well, the characters don't even mean anything to me. Also, though this is by no means your fault, the 40%+ that indicate no preference complicate the picture immensely. Do you know if there are any pollsters that at least try to flush non-respondents from the pool or figure their voting behavior? Thanks.Tyrenon (talk) 07:36, 28 November 2012 (UTC)[reply]
The shorter answer: I'm not aware of any poll that does what – I guess/extrapolate – you are looking for: Give a simple answer to the question ’How will the next House likely be composed?’. In the upper house election 2010, several newspapers published a prediction based on polling and taking into account the parallel electoral system, district disparities and local nomination strategy. (Most were more or less accurate as they predicted that the coalition would lose its majority, though if my memory serves me right the predicted opposition/government seat margin was narrower than it turned out to be. Even a perfect, 100% accurate generic poll for the district vote could obviously not have predicted the LDP 39/DPJ 28 victory because the popular vote was DPJ 39%/LDP 33%) I expect they'll publish a prediction for the lower house election as well; the complication of multi-member districts in the HC does not apply, but as you have to deal with 300 districts instead of 47, a seemingly even more disillusioned electorate (turnout!) and even more parties, it's probably just as difficult to get it right.
The long answer (a dabbling attempt at summarizing the peculiarities of Japanese elections – for a more extensive and qualified explanation turn to the abundant scientific publications on Japanese politics)

Problems for pollsters seem to be the parallel electoral system (that some Western observers seem to be totally unaware of or want to keep their readers/listeners/viewers ignorant of, judging from what is occasionally published in Western media), the (connected) party system and the highly personalized and localized campaign organizations, nomination strategies and voting behaviour. First, I'd avoid the term "party polling" without specification – you have to specify what was asked for: A. party approval (where as had been the case once in the 1990s, today roughly half of the population disapproves of any political party in Nagatachō), B. voting intention in the (180-seat) proportional race (where it looks as if many of the discontent seem to go for a "third force" this year, i.e. JRA, YP or if it succeeds Mirai no Tō/"Future Party") or C. the 300 single-member districts. The latter are what you would want to know about because they are almost two thirds of the House and swing heavily unlike the proportional vote which is by definition, well, proportional (though "block" constituencies with as few as six seats mean that larger parties still get a significant advantage). But, a generic SMD poll is not very useful (unlike, for example, generic House polling in U.S. congressional elections; but there you have only two major parties that contest more or less every district in the country. And while there are some "ticket splitting" voters, while local candidate quality obviously plays a big role and a (currently decreasing) segment of candidates is not aligned with their party on many issues, party affiliation is still a major factor Edit: An ill-chosen example in several ways, most strikingly: In the 2012 House election, apparently a popular vote of D49%/R48% (preliminary results) led to a seat distribution of R234/D200; but the essential point I wanted to make is that you could estimate based on a pre-election generic congressional poll that the GOP would retain control of the House and approximately by how many seats – if you had an idea from past results how local voting patterns, gerrymandered districts etc. would affect seat distribution; in the Japanese election, the additional factors seem to be more complex. A better comparison would probably have been recent UK general elections where you have three parties in a FPTP-system and the local Scottish/Welsh/Northern Irish parties complicating the picture; estimating the composition is still possible from a generic poll; but you have to have additional information. But where do you get that from if the likely third and fourth party (JRP & JFP) have only been founded a few weeks ago? JRP is second in many generic polls; JFP has a high number of incumbents. Any seat distribution estimate is a qualified guess – in Japan the relation between raw party vote and seat distribution is more complicated than in other countries, I think. --Asakura Akira (talk) 16:33, 2 December 2012 (UTC)):[reply]
1. Nomination is local. Example: If you asked a voter in Tochigi 3rd district in 2009 in a generic poll, his or her answer to want to vote for a Democrat or a Liberal Democrat was useless – there was none in the race. This year, LDP-Kōmeitō have candidates in almost all 300 districts (in the 290s I think), the DPJ IIRC had announced earlier this year to send "assassins" after every defector; but the last time I checked (shortly after dissolution) they were in the low 200s – good candidates don't grow on trees. The JRP, despite Hashimoto's long preparation for entering national politics, would apparently love such a tree as well. The Future Party which has ~60 incumbents has announced to nominate only ~100 candidates overall (SMDs+PR). YP has even fewer candidates. The only other party with a more or less countrywide presence in the (~290 I think) are the Communists; but they do not seem to be competitive in SMDs anymore even in their former strongholds. (ja: Jiji Press estimate of candidate numbers by party as of 11/17 (= obsolete, but as a first impression)) The other parties have only a few dozen candidates at best; but locally, they may have a shot. Which brings me to
2. Voting behaviour is candidate-based. If Japanese voters, particularly the more predictable rural voters, have written Koizumi or Hatoyama or Kishi or... on ballots for generations, they seem to be unlikely to change that even if the current candidate belongs to a different party. And more generally beyond "hereditary" politics, as one poll below shows, many voters do not vote for an SMD candidate based on party. At least they say so, but the actual results often show that it is true. Illustration: Some LDP families had a Socialist "black sheep" (Former Yamaguchi governor Tarō Ozawa's son Katsusuke Ozawa who served four terms in his father's district in Yamaguchi is the first who comes to my mind).
3. Campaign organization is to a high degree local and candidate-based. Example: Yūichirō Hata has decided to stick to the party line on hereditary candidate's and not run for his father's seat in Nagano's 3rd district. But if he hadn't, Tsutomu Hata's Kōenkai had already announced to support him even without DPJ nomination. So, he would have taken the local campaign organization with him.
4. Some other factors can not be covered by a straightforward "Who will you vote for?" poll. For example, Kōmeitō has a highly organized base that will almost always turn up. So if overall turnout is low, their relative share of the vote increases. And polls generally seem to underestimate the Kōmeitō vote while the party seems to know pretty well how many voters it has. While their total number of votes is high in the proportional race, Kōmeitō has often (with the exception of the landslide 2009 HR election) a very high ratio of district winners/candidates and therefore a high ratio of district seats/votes. In other words, if Kōmeitō does nominate a district candidate it's usually a winner; an example would be the Tokyo assembly elections just a few weeks before the HR general election defeat of 2009: Its 23 candidates won 23 seats. On the opposite end in this measure are the Communists: In the 2012 HR general, they will probably get 0 district seats out of some 290 district candidates, meaning several million "wasted" district votes.
Some things have changed since the 1990s, e.g. party HQs now have more money thanks to state subsidies, and there has been a trend of increasing importance of the national party leader in voting behaviour, and many argue that while not fully successful, the political reforms of the 1990s have indeed strengthened the role of issues in campaigns. But even if you are an optimist looking at the changes since the 1990s, local factors play a huge role and a generic SMD poll misses them. And you can't do local polling until you know for certain who runs. Which is on December 4 when the official campaign starts. A campaign of only twelve days (what Ohio voters dream of...) also means: Unless there is a huge change in how Japanese voters vote, the incumbent (or the more experienced former incumbent, as is the case this year with many LDP candidates) has a huge advantage regardless of party affiliation. Because a challenger has only twelve days to make roughly 100,000+ voters – depending on district, in some very malapportioned districts or in a genuine three+-way race significantly less –, for a start, make them remember his name – because of the way Japanese votes are cast – and convince voters that he or she can actually achieve more for them than the known face who has represented them in Nagatachō for years, in many cases, decades.
Considering these factors, polling is probably much more difficult than elsewhere, and I'd not read too much into those that are out there.

--Asakura Akira (talk) 17:14, 29 November 2012 (UTC)[reply]

Other polls

[edit]

A very incomplete collection of post-dissolution polling, maybe it saves some time for someone who wants to create a more comprehensive overview. --Asakura Akira (talk) 13:46, 26 November 2012 (UTC)[reply]

abbreviations: PR proportional representation segment (180 seats), SMD single-member district segment (300 seats), PM prime minister

Source Date Noda Cabinet approval Party approval PR vote Other questions
Approve Disapprove DPJ LDP Others None/
don't know
DPJ LDP JRP-SPJ Others None/
won't vote/
don't know
Kyōdō (ja) 2012/11/17–18 10.8% 23.0% 7.8% ...
Asahi TV Hōdō Station (ja) 2012/11/17–18 21.0% 56.3% 18.7% 29.5% JRP-SPJ 10.6%, Kōmeitō 3.6%, Others 7.2% 30.7% 12% 23% 10% (∑) 10% 45% PM (open): Abe 19%, Noda 12%, Ishihara Shintarō 6%, Ishiba 4%, ...
Fuji News Network (ja) 2012/11/17–18 21.3% 65.5% 13.5% 18.5% JRP-SPJ 9.9%+2.3%, Kōmeitō 3.2%, ... 41.1%+3.6% 14.8% 22.9% 22.4% (∑) 19.7% 18.2% PM (from a list of 9): Hashimoto 15.6%, Ishiba 13.0%, Abe 11.9%, Noda 10.9%, Ishihara Shintarō 10.5%, ..., NOTA 20.7%
Nippon TV (ja) 2012/11/16–18 23.8% 56.0% 15.9% 24.1% JRP-SPJ 4.7%+2.0%, Kōmeitō 3.2%, ... 32.8%+9.4% 15.5% 25.6% 5.9%+2.6%
(includes Genzei Nippon)
(∑) 12.1% 38.5% SMD vote: based on candidate 51.6%, based on party 36.7%; PM (open): Abe 19.0%, Noda 13.5%, Ishihara 7.6%, ...
NHK – [1] & [2] (ja) 2012/11/23–25 22% 64% 16.6% 24.3% JRP 4.2%, Kōmeitō 4.3%, ... 35.3% Ruling coalition: DPJ-led 9%, LDP-led 22%, DPJ-LDP 28%, without DPJ or LDP 25%
Kyōdō – Japan Times article (en) 2012/11/–25 23.9% 8.4% 18.7% 10.3% SMD vote: DPJ 8.8%, LDP 19.8%, JRP 9.5%; PM (method?): Abe 33.9%, Noda 30.0%, ...
Asahi (ja) 2012/11/24–25 18% 63% 12% 16% JRP 6%, ... 38% 13% 23% 9% Kōmeitō 4%, YP 3%, ... 41% SMD vote based on: candidate 34%, party leader 15%, manifesto 45%
Yomiuri (ja) 2012/11/23–25 21% 69% 10% 25% 14% ... SMD vote: DPJ 9%, LDP 27%, JRP 14%; Ruling coalition: LDP-Kōmei-JRP 21%, DPJ-LDP-Kōmei 13%, LDP-Kōmei 13%, DPJ-led 8%
Asahi (ja) 12/1–2 20% 59% 12% 15% JRP 5%, NFP 2%, ... No party 36%, don't know 23% 15% 20% 9% Kōmeitō 4%, JFP 3%, YP 3%, ... 41%
Yomiuri (ja) 11/30–12/2 13% 19% 13% JFP 5%, ... 41% SMD vote: LDP 22%, DPJ 13%, JRP 12%, ...

Seat distribution estimates:

  • Asahi Shimbun 12/4–5: LDP majority, DPJ below 100, JRP about 50, "Tomorrow Party of Japan" (as the JFP now chose to translate its name to English) about 8+ (eight PR seats and trouble in the SMDs – the "effectively being wiped off the map" kind of trouble)
  • Nikkei LDP gains majority, DPJ loses more than half of its current 230 seats
  • Sankei LDP-Kōmeitō majority (inluding ~200 SMDs with a clear LDP lead and ~5 with a comfortable lead for Kōmeitō), DPJ less than 100, JRP sluggish (~15 in Hashimoto's Osaka, but weak in Ishihara's Tokyo), JFP significantly down (struggling in SMDs to even be the runner-up get into double-digits), YP (Vote splitting in the 28 districts where YP and JRP run against each other), fight for survival: SDP, NPD, PNP (ed.: have only 2 SMD+1 PR=3 candidates anyway), NPN (ed.: only one Tanaka)
  • Nippon TV/Yomiuri 12/4–5 LDP comfortable majority (ed.: even if this holds, without significant DPJ defections or at least a three-party coalition after the HR general a nejire kokkai ("twisted Diet"/~divided government) still looms, LDP-Kōmeitō have no majority in the HC), DPJ southbound, JRP third party (has a chance for PR seats nationwide, first party in the Kinki bloc), JFP losing fight for SMDs, JCP little chance for SMD seats (ed.: despite their presence in 299 out of 300 districts) and struggling to keep current strength (ed.: Shii aimed to double their seat total), Kōmeitō and YP gain seats overall, struggling: SDP, NPD, PNP, NPN, NRP
  • Kyōdō 12/4–5 (Mainichi article, Japan Times article in English) LDP-Kōmeitō ~300, DPJ ~70, JFP ~15, JRP <50, JCP ≤9, SDP and NPD 1–2

Applies to all estimates: The number of undecideds is still very high. → volatility, turnout

Japanese Communist Party

[edit]

Why it's reported that "Japanese Communist Party continued to decline in strength and relevance"? In three years it's grown up from 4,22 to 7,88 in the local constituency vote and decreased only by 0,86% in the PR Block vote. It's lost 1 seat but has increased its percentage so it has lost in relevance but not in strenght. --I-pod keroro (talk) 22:19, 6 January 2013 (UTC)[reply]

[edit]

Hello fellow Wikipedians,

I have just modified one external link on Japanese general election, 2012. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

  • If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
  • If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 10:40, 22 November 2017 (UTC)[reply]