Talk:2014 Hungarian parliamentary election

Links

>> Support for Jobbik grows but unchanged for Fidesz, left>> Hungary ruling party's poll lead drops, far-right Jobbik jumps>> Orban trumpets harsh Hungarian ‘workfare’ scheme>> Jobbik not to enter any coalition, says Vona + jobbik's manifesto[ >> Hungarians vote in parliamentary elections>> Hungary’s elections: Don’t be fooled by the labels*[User:Lihaas|Lihaas]] (talk) 20:13, 9 March 2014 (UTC)

Context/background

Jobbik have grown exponentially by their presence in local government and hence#'#' popularity. This is perhaps important to show why/how its popular and that it is not mere ideology but also practical experience in govt.(Lihaas (talk) 20:13, 9 March 2014 (UTC)).

Polling section problems

The Nézőpont polls (and maybe others, I didn't look exhaustively) don't add up to 100% due to a whole bunch of undecided voters, and as a result, aren't akin to the other polls. This is not only confusing for the data in the table, but also creates errors when incorporated into the graph. Secondly, I think the format could be improved - the Greek election article has a good example, I think. That page also equalises polling results to just decided voters. Gabrielthursday (talk) 19:11, 27 March 2014 (UTC)

It's also typical to have the most recent survey data at the top of the table, even if it's sortable. Gabrielthursday (talk) 19:37, 27 March 2014 (UTC)

Hi, Gabrielthursday, I'm afraid I consider your edits problematic and will undo them for now, until we can find a better solution here. The Nezopont Institute polls which you recalculated specified, as the footnotes pointed out, two different categories of undecided voters: namely, those who were entirely undecided, and those who were undecided but in favour of a change in government. The latter were thus people who will *not* vote Fidesz-KDNP, even if they didn't yet know which opposition party they will vote for. That's why it's problematic to redistributed the entirety of these undecided votes, as you did, proportionally among all parties, including Fidesz.

To better reflect what these Nezopont polls said, you'd have to redistribute the percentage of entirely undecideds proportionally among all parties, and that of the "undecided but in favour of government change" proportionally among the opposition parties only. I actually did that in the charts, but that already seemed to skirt the Wikipedia's rule on no-original-research, so I was reluctant to do that in the table as well... I could do it still. But the plain recalculation you did is not an accurate reflection of the data in these Nezopont polls.No-itsme (talk) 03:08, 5 April 2014 (UTC)

Hey, first off, I want to thank you for your attention to this article. So please don't take my disagreement as a lack of appreciation. The first thing to say about the Nezopont data is that I did not do any redistribution - I simply excluded undecided voters from the data set (it's mathematically equivalent, but involves making no assumptions about the behaviour of undecided voters). So this is really a question of mathematics, rather than speculation. Polling firms can choose to release their data with a breakdown of undecided voters, or just as a poll of decided voters; both are accepted and ordinary - the underlying data is the same. I'd point out that this exclusion of undecideds is also an accepted and ordinary policy on Wikipedia, if it's appropriate for the article. See Next Greek legislative election for an example of the ordinary use of this method (you may have to hunt through the history and footnotes of the polling section, but I assure you it has been the ordinary and uncontroversial practice). I'd add that this process is not WP:OR, it's simply the presentation of the same information in a manner consistent with the other information in the table. It's the need for consistency that is most important- otherwise, the table would be (at least on the surface) misleading due to the mixing of different types of data.

I would point out that I also left the excluded data noted in the footnote, so readers are free to interpret the breakdown of undecided voters any way they choose. Just because I enjoy this kind of discussion, I could point out that there are good reasons not to distribute the "undecided but in favour of government change" to the opposition parties proportionally. Firstly, there is just one effective option to the right of Fidesz/KDNP - Jobbik. Voters to the right of the government and firmly opposed to the government are less likely to be undecided because there are fewer options. Moreover, the "undecided but in favour of government change" that are to the left of the government probably remain undecided due to unease or dissatisfaction with the leading left-wing option; I would expect this group to break disproportionately for the LMP and other small parties. In addition the "true undecided" by definition excludes those who are definitely opposed to the government, so it's not a neutral pool and shouldn't be redistributed evenly either. One could easily argue that they are more likely to break for the government since they haven't developed a firm opposition this late in the game. My point is that "proportional" redistribution is not always the sensible thing - it's actually more accurate simply to present the decided voters. I'd suggest that's true of the chart as well.

Finally, I'm going to revert back my changes to the Nezopont data. I made my comments here a number of days before taking action, and was happy to listen to discussion over the proposed changes. While I'm not suggesting any impropriety on your part, I did not have the chance to address your concerns before you reverted the article. I also believe I've adequately answered your concerns, so I too will be bold and make the move back to the status quo ante. Gabrielthursday (talk) 10:14, 5 April 2014 (UTC)

Well, I don't agree. Normally, a pollster just shows different sets of data for (a) the general population and (b) decided voters. That's the common practice, and in such cases your solution makes perfect sense. Nezopont, however, as the links document, has taken an unconventional approach, for the entire period from February 2013 through to last month. It provided a set of data for the general population, with lots of undecided voters; and then this separate selection of "active voters", which still had a share of undecided voters, but a much smaller one, and divided up those undecided voters into "overall undecided" and "undecided but in favour of government change".

This is unusual, of course, and posits a dilemma for us. And I believe that in this case, the solution you have applied, which would otherwise make sense - and would make sense even to some extent if applied to the "general population" sample - substantively distorts the results. This distortion is also immediately visible on a gut check, when comparing the numbers from their latest poll. They have, annoyingly, again changed practices, and this time omitted the "active voters" sample, providing only a general population sample and a "potential voters" sample, which includes no undecided at all. We'll have to use the latter of course, which turns out to lead to a jarring contrast with the numbers for their last two polls as you corrected them - making it seem like Fidesz somehow lost 12% in one week, even as both Unity and Jobbik gained 5% in one week - and all this at a time when the other polls are showing Fidesz staying level and Unity dropping. Now if it was the *opinion pollster* itself which presented such a jarring shift in voter preferences in contrast with other polls, we should of course present it accordingly. But they didn't: after all, their previous polls already explicitly indicated that among the undecided but "active voters", there was a substantial share that was committed to voting, and voting against Fidesz. So the jarring contrast in this case it is purely a manifestation of our own choice of excluding part of the data they presented.

Again, I absolutely understand how the simple exclusion of all undecided voters makes perfect mathematical sense with any conventionally presented data that includes a generic "undecided" section, when no alternative data that adds up to 100% is presented by the pollster. But Nezopont explicitly chose to distinguish the overall undecided in the general population sample from this more specific data, where you have "active voters" but even those include a smaller number of undecided, and those undecided are in turn disntinguished in two categories, one of which implies a form of voting preference (against the government). So in this case, excluding them all is not just the mathematically straightforward exclusion of undecided one would normally be doing, and I think the "gut check" of comparing the most recent Nezopont data bears that out.

That is also not the only problem I see. The edit you made leads to a jarring contrast in both directions, timewise, because it was applied selectively only to the last two Nezopont polls before the most recent one, when all the ones before, going back to February 2013, still use the non-exclusion method. This in turn makes it seem like Fidesz suddenly gained 13% in one month in February. That all seems very confusing. I would argue that consistency in data presentation trumps the desire to see any one individual poll's data add up to 100%. So unless we'd feel like doing the same modification retroactively throughout the previous year, it ends up being more misleading, and having more of a distorting effect, to have the data suddenly jerk into another pattern for two weeks before settling down again in the most recent one, than it was to just present the Nezopont numbers as given by the pollster and not add up to 100%. The cure, here, it seems, is worse than the flaw it was laudably intended to solve.

I have not reverted your reversion, of course, because I don't want to get into an edit war, but I do think they should be. No-itsme (talk) 12:31, 5 April 2014 (UTC)

Edit: I've asked Norden1990 to see if he'd like to look at this and weigh in, he wrote most of the rest of the page - might be helpful.No-itsme (talk) 12:51, 5 April 2014 (UTC)

We should indicate the different method of Nézőpont polls, and, I think, current format is also comprehensible and suitable, like version in Greek election page. Nevertheless, I'm not an expert on public opinion research methods, tools and its visualization. --Norden1990 (talk) 15:23, 5 April 2014 (UTC)

Thank you for your thoughtful response. I welcome Norden's contribution as well. Firstly, you are quite right that I ought to have presented all the Nezopont poll data as decided voters rather than just the February & March ones. I am happy to remedy that, though perhaps you would like us to reach some agreement here first.

I can offer little explanation of why Nezopont's poll results vary so dramatically from March to April. I suspect it is a combination of a changed methodology (it's a little unclear, but it seems probable that Nezopont dropped their "active voter" screen) and statistical noise. Maybe there was some real voter movement there as well. Unexplained variance between polls happens often. All we have to go on is the data provided, and we should simply present it in the clearest and most consistent manner. In that respect, it is not really a question of needing the numbers to add up to 100, it is more that we should be presenting the data in a consistent manner. It would be entirely proper to have a poll table with an undecideds column (which could even be divided up for the Nezopont data); but the table as it stands now does not have such a column, and is largely populated with decided voter data.

With respect to the integrity of the Nezopont data, I'm not entirely sure how the fact that Nezopont broke the undecideds into two groups could affect the data about decided voters. I would assume Nezopont simply asked respondents who indicated they were undecided an additional question, thereby giving them the two groups. How this could affect the decided voter data is a mystery to me. As a result, presenting the Nezopont decided voter data shouldn't be problematic. I do agree that the undecided numbers give additional information; but it is additional information that was never in the table in the first place, and has not been removed. Perhaps the Nezopont data on the undecideds could be presented in a more visible and helpful manner, but then again, perhaps that would be taking this article a little too far into the weeds of polling data.

Finally, you mentioned that for the chart you allocated the undecideds from Nezopont in a manner that benefited the opposition parties. I am not sure that this is an appropriate way of doing things. If we just had a certain number of undecideds, would we not just allocate them proportionately (or simply use the decided voter data)? When the undecideds are broken into anti-government undecideds and "true" undecideds, the "true" undecideds change tendencies as well, simply by the exclusion of those more likely to vote for opposition parties. Wouldn't just using the decided voter data be the simplest solution for the chart data as well as for the table? Gabrielthursday (talk) 16:44, 5 April 2014 (UTC)

Hi, thanks for your quick and thoughtful response. I'm going to align left again because otherwise our columns become very narrow.

You write that you "can offer little explanation of why Nezopont's poll results vary so dramatically from March to April", but I think that's missing the point. Their results did not actually vary all that dramatically. It's, to be blunt, primarily (though not wholly) your change in how we present their data that makes it *look* like they varied dramatically. It's at least in part an artifact of the way we changed the presentation of their results, not their results themselves. That's why I objected to your edits, and why I still think they should be reverted.

Nezopont presented, ever since February 2013, two samples. One of the general population, with a large proportion of undecideds (a huge proportion even, something which is quite common to Hungary). If we only had that sample, it would make sense to convert them the way you did, by simply excluding the undecideds and recalculating the remaining numbers to reflect percentages for decided voters only.

Secondly, however, they presented a sample of "active voters". This seems to have been a kind of hybrid between a global population sample and a decided voters sample. An unconventional format for sure, and to a large extent our discussion here can be blamed on this. From as far as I can tell, this sample consists of those who expressed a determination to vote, but still included some who did not yet know whom for. Moreover, that latter number was subdivided in those who did not know at all, and those who *did* know they would vote against Fidesz. And it is this (highly unconventional) element which, in my opinion, causes the integrity of the polling data to be harmed when simply excluding all the undecided voters and recalculating the rest to 100% even though a part of those undecided did express a voting preference, if only a negative one.

I.e., in its March poll, in its "Active voters" sample, Nezopont found 48% preferring Fidesz, 19% Unity, 11% Jobbik, 6% intending only to vote *against* Fidesz, and 13% entirely undecided. In its April poll, it finds 47% preferring Fidesz, 28% Unity, 19% Jobbik, and no more undecideds included. The shift it finds is therefore that the undecideds transferred to Unity and Jobbik - which, considering that a third of them had already been expressing a determination to vote against Fidesz, is not such a dramatic shift.

Now exclude all undecideds from March's number, as we've done here; not just the generically undecided, as is often done in these kind of tables like you say, but also this unconventional category of respondents being recorded as having a definite negative preference, if no positive one yet. This results in Fidesz share of the vote in February-March shooting up to 59%; only to abruptly drop 12 points to 47% again in April when Nezopont's own data adds up to 100%. An abrupt break with Nezopont's polling trend over time, and contrast with every other poll. Of course this doesn't *prove* that we made a wrong choice in how we present their Feb/March data; but it certainly seems to suggest that we might have. If the point of the edits was, as you initially said, to make our presentation of the data less confusing, it seems like it might have had the opposite effect.

There is no good option here, I think (thanks, Nezopont!). Only a choice of least bad solutions:

(1) would be to use Nezopont's entire population sample, and recalculate that data by excluding the undecideds, who are not subdivided or anything in that sample. But why use an entire population sample if the pollster provided a more narrow one too?

(2) would be to use Nezopont's active voters sample, which still has undecideds but many fewer, who in turn are subdivided by negative or no preference; and simply present that data as-is in our table. Downside: this leaves a share of undecideds in the Nezopont polls when there aren't any in those by the other pollsters. Include a separate column for those if necessary. Imperfect as this is, I think this would probably be the most conscientious option, considering the unconventional specificity of Nezopont's subcategories of committed-yet-(partly) undecided voters.

(3) would be to recalculate the data from Nezopont's active voters by excluding the undecideds - both subcategories of them - as you did. This would be the standard method; but it discards the unconventional, yet specific information in Nezopont's subcategories of undecideds from the data in our table (even if it's still mentioned in the footnote). And the comparison with Nezopont's April data at least seems to confirm that this leads to a significant distortion.

(4) would be to recalculate the data from Nezopont's active voters by excluding the fully undecided, and assigning the anti-Fidesz voters among the undecideds to the opposition parties in a proportional manner. This might get the closest to what was actually being measured. It's also the least-bad option I'm using in the charts, so that speaks for using it in the table too. But it might come uncomfortably close to Wikipedia's ban on original research. (It also still leaves that jarring contrast between Nezopont's March and April data, though it's smaller.)

I fully agree that we need to present the data in a clear and consistent manner. Since we have a rather unconventional case at hand here, my solution would be to stick closely to the data as reported by the polling agency, and avoid the usual exclusion of undecideds since the data about them in this case is more specified. That admittedly leads to some distortion in comparison with the other pollsters' data, but - on the face of it - less so than option (3).

Re: your final note, I am not "allocating the undecideds from Nezopont in a manner that benefited the opposition parties" in the charts. I'm allocating that share of the undecideds who explicitly state that they would vote for the opposition, to the opposition parties.

(Did some editing for clarity, done now).No-itsme (talk) 22:24, 5 April 2014 (UTC)

Firstly, forgive me if I am about to write about things you already know well; but it seems to me that our disagreement goes to some issues of how polls generally work. So, I gather your primary concern is how to deal with the "undecided-but-against-the-government" voters identified by Nezopont. Let me preface some of my thoughts by noting that while I am far from an expert on polling, I have answered polls, administered polls, read a fair number of polling scripts, discussed polling with a professional pollster and read a little on how polling works. So while I certainly don't claim to have exhaustive knowledge of the field, neither am I just guessing. One would always ask about which party one intends to vote for before asking questions designed to obtain more information about those who state they are undecided. So the data about decided voters in the Nezopont polls is directly parallel to the data about decided voters in the other polls. The information about undecided anti-government respondents would not affect the decided voter numbers in any way, since the question that identified the undecided anti-government group would only be asked after the general voting preference question.

I think the issue about the "active voter" sample is a bit of a red herring. Nezopont apparently uses (or used) what is typically called in North America a likely voter screen. It's a very common methodological tool, so I would disagree with your assessment of it being unconventional. I gather most if not all of the other pollsters cited in the table use a similar kind of screen, so this kind of method would appear to be fairly typical in Hungary as well.

Excluding the undecided voters in the Nezopont data is exactly parallel to what every other pollster has done; they found undecideds, they stripped them out, and published the data of decided voters (often alongside the broader data). Nezopont chose not to present their data in this way, but the decided voter data is right there - and, for the sake of consistency and accurate comprehension, we can and should do to Nezopont's data exactly what all the other polling firms did to their own data. As for their high numbers for Fidesz, I can only speculate - maybe it's a different methodology, maybe some institutional bias, maybe publication bias; they might even have faked the data. But I am not artificially increasing their numbers in any way - that's just what their decided voter data is. Yes, it looks slightly more consistent if you're comparing apples to oranges, but that's a false and misleading "consistency".

Regarding option 4 that you used for the chart, it's not so much the proportional allocation of the anti-government undecideds to the opposition parties that's the difficulty; it's the proportional allocation of the remaining undecideds among all parties. Simply put, the remnant "undecided" group is not going to be a neutral group - by definition, the poll took all the undecideds and removed those most opposed to the government. As a result, it would necessarily be more inclined to vote for the government than all the undecideds together. So the procedure under option 4 would only be useful if we believed that undecideds as a group were going to favour opposition parties - for which supposition we have absolutely no good evidence. Even then, the resulting numbers would favour the opposition parties by an arbitrary amount.

Incidentally, that Ipsos forecast was very odd. I have to imagine they are seeing some late movement in their data for them to project Fidesz noticeably lower than both their decided voters or their "certain and decided voters". Gabrielthursday (talk) 01:14, 6 April 2014 (UTC)

Thanks, and yes, I've been following and studying polls and the profession of polling myself for quite a while too. I haven't administered polls myself, but otherwise to the same extent as you list about yourself.

Regarding what you say about how the data about decided voters in the Nezopont polls is directly parallel to the data about decided voters in the other polls *even if* you strip out the decided anti-Fidesz voters, I don't agree. This goes back to the issue of how different pollsters push their undecided respondents to different degrees, sometimes using a follow up question, and how the data they publish reflects the overall results of that choice. (E.g. I've seen polls that asked a follow up question along the lines of "if you're not certain which candidate you'll vote for, which candidate are you most leaning to vote for?", and integrate the answers in their topline results.) Nezopont's unconventional data about determined-yet-undecided anti-government respondents reflected such a decision on how far to push undecideds. In their case, far enough to yield a finding that these respondents are determined to come out to vote, and to vote against the government; but not to express a specific preliminary party preference.

Simply stripping this group out changes the political balance of the sample. The remaining determined & decided voter sample with a specific party preference would lean more to Fidesz than the original determined voter sample which includes those who are determined to come out, and decided to vote against Fidesz, even if they didn't express a specific party preference. The gut check comparison with Nezopont's very last poll does not provide any proof, of course, but at least seems to confirm this.

What is unusual about Nezopont's data here is not, of course, simply that they use a likely voter screen, which I'm well aware is commonly used in the US as well. What's unusual is that they subdivided the undecided respondents that remained even in their likely ("active") voter screen, identifying a specific subset as having at least a negative voting intention (against Fidesz), if no positive voting intention. This is a category of information I have not encountered in other polling data (neither in Hungary nor in Germany, Holland, the US, Britain or France). And the question of how to process that piece of data is at the heart of our disagreement, I think. Or rather, the question of whether to process that information in the numbers we present at all or discard it and leave it to the footnotes.

(In addition, I thought it unusual that even their likely voter screen still had a sizable number of undecideds at all, but that may just be me, and I might be wrong about that.)

Regarding your comments about option 4, I don't think I can follow two of the assumptions you seem to be making. You say that the problem is "not so much the proportional allocation of the anti-government undecideds to the opposition parties", so we agree on that point. Those who indicated that they would vote, and vote against the government, would vote for the opposition. Simply reallocating their vote to the opposition parties proportionally is an imperfect solution, for sure; but at least it ensures that we don't discard the information about their voting intention altogether, the way we are doing now in the table.

Instead, you argue that the remaining undecideds - those who expressed being "fully undecided" - should not be simply excluded, proportionally increasing the percentages for all parties, because they will not "be a neutral group". You base this argument on the fact that, as a group, they will be more pro-government than the sum total of them *plus* the determined anti-government voters. But how is that relevant? There is only one government party/alliance in Hungary (Fidesz/KDNP). So there can be no group of pro-government, yet undecided voters, in a way that would mirror the subcategory of anti-government, yet undecided voters. So there can also not be such a group still mixed up in this remaining 'wholly undecided' sample. Those who expressed being wholly undecided will therefore really have been wholly undecided, also on the choice between government party and opposition parties. Excluding them (resulting in the percentages for all parties, government and opposition alike, going up proportionally) reflects that.

The conclusion you draw here is that, by doing the above, we would be implying that "undecideds as a group were going to favour opposition parties," even though "we have absolutely no good evidence [for this] supposition". Let's look at this. Nezopont found that the undecideds in their active voter sample, as little or as much as they were pushed, were not in fact all wholly undecided. They identified that a sizable proportion of the undecideds were in fact determined to vote against Fidesz. They identified no similar group of pro-government undecideds (unsurprisingly, since that would be something of a logical impossibility, considering there is only one government party). And finally, they identified a group of wholly undecideds (who didn't express a lean either to or against government). You argue that excluding or reallocating the last group's data distorts the data in the opposition's favour, because this last group is not in fact "neutral", but relatively more government-friendly. As I explained above, I don't think this holds water. In fact, a smell-test check with the last Nezopont poll seems to suggest that even these undecideds swung largely to the opposition, though it's impossible to tease that apart from the effects of what was probably a less restrictive likely voter sample ("potential voters" vs "active voters").

Lastly, about that Ipsos forecast - I don't think that it's that they were "seeing some late movement in their data" that led to their projection for Fidesz being noticeably lower than the polling data about certain and decided voters that they published at the same time. It's more likely that they realized that their voter screen was too narrow, in comparison to what actual voter turnout is likely to be. And that the voters who fell beyond their committed-certain voter screens, but will nevertheless turn out to vote, will disproportionally lean to the opposition. That's what the article says, actually: "Ipsos said it expected a higher participation rate on actual election day, so there are likely to be divergences from figures in the committed-certain camp." This, I suppose, might be exactly the problem Nezopont was trying to address by including its undecided-but-for-government-change category in its active voters sample.

Median has done the same thing, by the way. They also published a projection of results (see the last chart) that is far more friendly to the opposition than the polling results they published at the same time for either of their voter screens. We'll soon find out whether they were right. :-)

Anyway, election day is almost over. I don't agree with your edits, but it's not like I'm going to revert them - edit wars are silly. And there will be no further polls to add for us to worry about. So I suppose this whole discussion has few practical implications.No-itsme (talk) 16:40, 6 April 2014 (UTC)

Results

I think changes of seats compared to 2010 is not an useful data as new electoral law has cut the size of the parliament. Maybe the comparison of the proportion of seats would be more interesting. --Norden1990 (talk) 16:00, 7 April 2014 (UTC)

If I recall correctly, other election pages have scaled the seats of the outgoing parliament to the new size, and compared the "adjusted" number with the new number. Gabrielthursday (talk) 17:36, 7 April 2014 (UTC)

Makes more sense, this is deveptive. We can add a note there too. Or remove it all togherLihaas (talk) 15:41, 9 April 2014 (UTC)

Infobox

Hi there,

I don't edit this article but noticed that the infobox data is from the previous election in 2010. Could someone please update? Szaboci (talk) 11:02, 8 April 2014 (UTC)

Soon. There is no final result for now. We are looking forward to the votes from the foreign representations. Fidesz's two-third victory isn't sure yet as there are several pending constituencies in Budapest (District XVIII notably), Miskolc e. g. --Norden1990 (talk) 18:05, 8 April 2014 (UTC)

Copyright problem removed

Prior content in this article duplicated one or more previously published sources. The material was copied from: multiple sources (see Wikipedia:Contributor copyright investigations/Norden1990). Copied or closely paraphrased material has been rewritten or removed and must not be restored, unless it is duly released under a compatible license. (For more information, please see "using copyrighted works from others" if you are not the copyright holder of this material, or "donating copyrighted materials" if you are.) For legal reasons, we cannot accept copyrighted text or images borrowed from other web sites or published material; such additions will be deleted. Contributors may use copyrighted publications as a source of information, but not as a source of sentences or phrases. Accordingly, the material may be rewritten, but only if it does not infringe on the copyright of the original or plagiarize from that source. Please see our guideline on non-free text for how to properly implement limited quotations of copyrighted text. Wikipedia takes copyright violations very seriously, and persistent violators will be blocked from editing. While we appreciate contributions, we must require all contributors to understand and comply with these policies. Thank you. Moonriddengirl ^(talk) 10:11, 9 May 2014 (UTC)

Bring back the rest of the article!

This was once a well-written, expansive article on the topic. What happened? Now it's just a bunch of statistics. KFan II (talk) 22:55, 11 May 2014 (UTC)

Copyright problems. See above. Gabrielthursday (talk) 07:25, 12 May 2014 (UTC)

External links modified

Hello fellow Wikipedians,

I have just modified 2 external links on Hungarian parliamentary election, 2014. 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:

• Added archive https://web.archive.org/web/20140116131232/http://www.hazaeshaladas.hu/ftp/speech_of_mr_gordon_bajnai_-_23_october_2012.pdf to http://www.hazaeshaladas.hu/ftp/speech_of_mr_gordon_bajnai_-_23_october_2012.pdf
• Added archive https://web.archive.org/web/20141021193827/http://www.hirado.hu/Hirek/2013/03/08/11/Partta_alakult_az_Egyutt_2014__kozismert_arcok.aspx to http://www.hirado.hu/Hirek/2013/03/08/11/Partta_alakult_az_Egyutt_2014__kozismert_arcok.aspx

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) 20:59, 8 November 2017 (UTC)