Political opinion in Hungary--Medián and Tárki

Two respectable opinion polls came out today with slightly different results but confirming one critical fact--that people who voted for Jobbik in the EU parliamentary elections came largely from the Fidesz camp. Medián specifically asked Jobbik voters about their political preferences in 2006. It turned out that only 14% of them had voted for MSZP. The rest either came from the right or were new voters. Tárki indirectly supports this result. Among those who said they would definitely go to the polls only 58% said that they would vote for Fidesz this Sunday while in May that number was 70%. A twelve percent drop is considerable especially if, again according to Tárki, MSZP gained only six percentage points during the same period. In fact, Tárki specifically states that "today, in comparison to the times prior to the EP elections, the number of Fidesz supporters has shrunk by the same number of percentage points as the Jobbik camp has grown." So although Viktor Orbán tries to minimize the problems stemming from the spectacular growth of Jobbik, his grandiose goals may be in jeopardy.

Another way to look at the change is to compare the survey results from March with those from June. Here I'm reproducing Tárki's bar chart showing shifts in party preferences of those who would definitely vote and who know for whom they would vote. Today MSZP, SZDSZ, and MDF have the same level of support they had in March; Fidesz is down four points, Jobbik is up six points, and "others" are down two points. I should add that in the most recent survey these voters represent at best a slim majority of the total sample; those who at the moment are uncertain for whom they would vote is 40%.  Moreover participation would be very low if elections were held today. According to Medián fewer than 50% of the voters would actually go to the polls. So that is another variable in trying to forecast the outcome of the national elections. Normally participation in national elections is more than 60%. Yet Medián is bold enough to predict that "whenever the next elections are held it seems that Fidesz will get the most votes." One reason for this assumption is that 86% of the voters, including  65% of the MSZP supporters, predict a Fidesz victory.

Medián asked some penetrating questions about Jobbik. Although many people after the EP elections proudly claimed that they knew all along that Jobbik would surpass the 5% minimum, according to Medián "two months prior to the EP elections only every twentieth person predicted such a success." Of course, as usual people are quite forgetful. Now fewer than 30% say that they were surprised at the showing of Jobbik. However, one must keep in mind that only 64% of them could even list the names of the parties that were able to send representatives to Brussels.

The majority were either very happy or more or less happy with the results. Twenty-one percent were very happy and 40 percent more or less happy. I assume that the Jobbik voters (close to 15%) are ecstatic and it is hard to tell what "the more or less" category actually means. I wouldn't draw the conclusion that Medián reaches that "public opinion is more tolerant of Jobbik than earlier." It may simply mean that Fidesz voters might have been more enthusiastic if the party had received 17 seats and not 14. Therefore they are more or less satisfied. Public opinion is very divided as far as Jobbik is concerned. Fifty-four percent consider the party "dangerous"  and 45% think that Jobbik is actually a "fascist" party. At the same time 39% (up from 34%) think that "only Jobbik really cares about the common man." Or they expressed the opinion that "Jobbik at last tells how it is."

Medián probed the topic of the relationship between Fidesz and Jobbik. A huge majority of MSZP voters felt that Fidesz should make clear its refusal to cooperate with an outright Nazi party. Those who voted for Jobbik insist that Fidesz should look upon them as allies. Then Medián continues: "It doesn't matter which way Fidesz moves, the party would lose voters because Fidesz supporters are themselves split on the issue." About one-third of Fidesz voters agree with the majority of MSZP while a little over one-third of them would prefer cooperation with Jobbik.

Medián tried to find an answer for Jobbik's success. It seems that its anti-Gypsy rhetoric was an important factor. Jobbik, a party almost no one had heard of a few months ago, started to gain momentum when crimes committed by Gypsies received widespread publicity. Medián checked the voting results according to the size of the Gypsy population and came to the conclusion that in cities where their population was greater than 5% Jobbik received almost 20% of the votes. Gypsies are highly concentrated in places where MSZP is normally strong, so the common wisdom right after the elections was that most of the Jobbik votes came from the left. However, it was also in these places where participation was low. The Medián assumption is that it was mostly MSZP voters who stayed home. As for where the Jobbik votes came from here is an interesting chart. Medián asked Jobbik voters which party they had voted for in 2006. If one can believe the information they provided, 9% either don't remember or will not tell; 24% are so young that in 2006 they were not eligible to vote; 15% didn't vote but were now inspired by Jobbik; 13% voted for the MIÉP-Jobbik joint ticket; 25% voted for Fidesz-KDNP, and 14% came from the MSZP camp. The very high percentage of first-time voters reaffirms suspicions about the popularity of extreme right ideologies among the youth.

The most dramatic aspect of these two opinion polls is that they clearly delineate the problem Fidesz is facing. The problem is  genuine: it doesn't matter which direction Fidesz moves, the party will lose a sizable portion of its voters. Right now the strategy is to ignore the problem. I guess the Fidesz leadership is hoping that Jobbik will lose its appeal. It will make one big mistake or many smaller ones that will make it less attractive. And that is a possibility. It is enough to think of the "conference" held in Szeged of the various extreme right-wing organizations including Gábor Vona (Jobbik) and György Budaházy, who might be the mastermind behind the Arrows of Hungarians, a terrorist organization. Budaházy is already in jail together with five other bomb makers. Then there are rumors that the parliamentary caucus in Brussels that the three Jobbik members wanted to join would take only Krisztina Morvai because she is not a member of a neo-Nazi party. So Orbán's calculations might be well founded. As a fall back position, if Jobbik survives or actually gains ground then Fidesz will make a deal with them in the belief that they can handle them. After all, they made a deal with the Smallholders of József Torgyán and no harm came of it. In fact, Fidesz was so successful that today there are no more Smallholders and József Torgyán pretty well retired from politics. But that is a gamble. Jobbik might be much more resilient and ideologically committed.

Comments

[Malmomonster]

Sorry to be off topic, but I've been told this is one of the the best bi-lingual Hungarian politics sites.

I'm writing a piece on the rise of the European far-right for a Swedish paper.
I just wanted to know please, is this an accurate translation of Kristina Morvai?
http://www.youtube.com/watch?v=4CwilgZBi7Y

Koszonom!

[whoever]

"The very high percentage of first-time voters reaffirms suspicions about the popularity of extreme right ideologies among the youth."

It's interesting, after nearly 8 years of a "socialist-liberal" government (which was neither in many senses), that the younger generation tend to support right-wing radicals. I suspect that it means many families now exist in a "new Hungary" cultural vacuum - where politics and history are rarely discussed in depth, where there is no tendency to read books.

The main failure of the republic to date has been to engage with the people on a meaningful level. It's an inevitable result of the elitist, centralised approach as followed by the SZDSZ and MSZP over the years. If the sky comes crashing on their heads next year they will have to look at themselves. I hope they flush themselves down the drain, to give more credible left parties a chance to emerge.

Having said that, this dumbing-down is occuring across Europe and the USA - seemingly connected with the decline of reading among young people. It appears particularly steep here - there are a lot of houses I have seen with no bookcases at all. Whereas in the UK reading as a leisure and educational activity seems to have been maintained. A good study in Hungary would be to look at reading habits and intellectual activity now compared to 20 years ago.

I suspect that the "new illiteracy" is more pervasive than we would imagine, and that the ignorance this engenders is creating a fertile bedrock for demogogues and cranks.

[Mark]

"Two respectable opinion polls came out today with slightly different results but confirming one critical fact--that people who voted for Jobbik in the EU parliamentary elections came largely from the Fidesz camp."

They don't actually confirm any such thing. The interesting one is Medián because it underlines two important points which have not yet received the attention in the debate they deserve - namely the importance of the act of staying at home as an active determinant in election outcomes, and the importance of those with no prior political affiliation (first time voters). In terms of drawing broader conclusions let me give some health warnings about public opinion polling, especially in the context of party systems which are subject to significant change (as Hungary's is at the moment).

(1) We know from the success of sites like fivethirtyeight.com in the United States that public opinion polls are not reliable or precise predictors of election outcomes. Normally their raw data has to be subject to some statistical manipulation before they can be considered accurate.
(2) This is for a number of reasons - a researcher asks someone a question in fieldwork for a poll (they come to the voter). In an election the voter has to remember their identity card, walk down to the polling place, and vote themselves. Furthermore, while in an election a voter can trust in the fact that their ballot is secret, they cannot do so to the same degree with an opinion pollster. The distorting effect of this effect on polling figures has been measured in a number of elections - the UK general election of 1992, the German Federal election in 2005, and, of course, Hungary's parliamentary elections in 2002.
(3) Because of this if a party is deemed in the public sphere (media etc.) to be socially unacceptable, yet is widely supported, opinions polls will tend to understate its support, because voters are unwilling to tell a researcher they support it. This is a widely documented phenomenon with far-right parties, and I would guess, is what is behind opinion polls' under-reporting of the Jobbik score.
(4) Polls are notably bad at predicting scores in low turnout mid-term elections. When disrepancies are revealed polling after the election tends to correct to the election result - this happened after the 1989 European elections in the UK. This is why the FIDESZ vote fell by 12 points in the Tárki survey. It reflects no change in actual opinions - it is just reflecting the fact that an election has taken place where people actually voted.
(5) The polls are overreporting likely MSZP scores and continuing to underreport the Jobbik score. I'm guessing the Tárki survey was conducted late last week and last weekend, i.e. within two weeks of the European election. There is no reason to believe that any events have occurred that would have changed significantly any votes since. The anticipated turnouts for national election are very close to those for the EP. In this context, if one compares these survey results to the actual vote, then only the FIDESZ score is within the poll's margin or error. The Jobbik score in this poll should be no lower than 12%, but the actual score is 10%. Likewise the MSZP's score should be no higher than 20%. It looks to me like there are a small number of voters who are telling pollsters they will vote for the MSZP, when actually in the polling place they will cast their votes for Jobbik!
(6) There are two big problems from drawing too many conclusions from Medián's survey. The first one is that the sample is clearly a sub-set of their broader sample, and given that they are surveying a small number of people that the statistical margin of error is extremely high (perhaps as high as +, or -20%), and so high as to render the precise figures close to meaningless. But, it is interesting that neither the HVG nor Medián are telling us anything about their methodology which would enable proper and independent checking of whether their conclusions are supported by their research!
(7) The second problem is that we know that asking people about their prior voting behaviour in polls is problematic. It has been shown in previous elections in a number of countries (including Hungary)that what people say about their previous voting history is influenced by their current opinions. So, we would expect, for example, given the current unpopularity of the MSZP, that a significant section of 2006 MSZP voters would tell an opinion pollster now they had voted for someone else!

There is no way of settling the substantive question through opinion polling. Someone needs to do some detailed statistical analysis of the raw results - something like the Wählerstromanalysen that are produced by opinion pollsters for newspaper and broadcasters after elections in Austria and Germany.

[Mark]

"Medián checked the voting results according to the size of the Gypsy population and came to the conclusion that in cities where their population was greater than 5% Jobbik received almost 20% of the votes."

Why just concentrate on cities? Surely given the concentrations of Roma populations in rural areas this isn't very meaningful. And more importantly - what they say is true for northern Hungary, and the northern Great Plain region, it is not true west of the Danube. They need to explain the regional divergence.

[Eva S. Balogh]

Mark: "They don't actually confirm any such thing"

Well, we disagree. A drop of 12% from one month to the next is fairly convincing to me. Especially when the same pollsters show a large jump in the popularity of Jobbik during the same period. Medián's questioning Jobbik voters' past voting habits is also convincing to me. I don't understand why you stick with such gusto to the theory that former MSZP voters make up the bulk of the extreme right's constituency. This is simply cannot be maintained when you look at what MSZP voters think of Jobbik. Especially if you compare that with the detailed analysis of current Fidesz voters' attitude toward Jobbik.

[Eva S. Balogh]

Mark: "Why just concentrate on cities?"

I don't think that they did. Most likely the label is wrong but I can ask.

[Mark]

Éva: "I don't understand why you stick with such gusto to the theory that former MSZP voters make up the bulk of the extreme right's constituency."

This isn't quite what I'm saying - in fact I'm absolutely sure that some of the more extreme interpretations (that five-sixths of the Jobbik vote came from the MSZP)are wrong. It doesn't look to me though that any more than a third of Jobbik voters are ex-FIDESZ voters (this actually could be argued to be consistent with Medián). Jobbik also has disproportionate appeal in areas where the local milieu let's say has produced high MSZP votes in the past (this isn't the same as saying that all of these people are former MSZP voters - though I suspect at least a quarter of them are). Clearly there is movement from first time voters and non-voters into Jobbik's camp. I think the truth is a bit more worrying than either of these interpretations - that Jobbik has created a third camp which cannot be easily pigeon-holed into the established right-left spectrum.

Éva: "This is simply cannot be maintained when you look at what MSZP voters think of Jobbik."

When one looks at the far right Europe-wide this apparent paradox is really not hard to explain. You can say the same about Labour voters in the UK in relation to the neo-Nazi British National Party (which wins most of its votes from working-class voters in medium-sized ex-industrial cities in northern England). You can certainly say the same about Austrian Social Democrats and the Freedom Party, or French Socialists and the Front National. The mistake you make is in assuming that social democratic voters are a homogeneous block. Typically (and the MSZP is no exception)the social democratic coalition is made up of a culturally liberal, public-sector middle class (who are extremely anti-racist, and culturally progressive) and a larger working-class constituency which votes for the left on bread-and-butter issues, but when asked about cultural issues (law-and-order, the death penality, immigration, race)tends to have attitudes that place them on the hard right. The real story of social democrats across Europe is that as they have retreated from offering this voter group what they want in terms of jobs, welfare etc., they instead gravitate to parties on the far right which are more in line with their cultural attitudes. I'd maintain that what is happening in Hungary is very little different to what has happened in western Europe. The difference is the speed of the MSZP's collapse, and that because of the degree of economic insecurity (objective and subjective) the potential space for the far right is greater.

Éva: "A drop of 12% from one month to the next is fairly convincing to me."

My approach to the polling is determined by the fact that if you look across Europe since 1945, no free-and-fair election has occurred in any country where the winning party has taken a vote share approaching 70% (there are some free, but not fair elections like, say, Romania in 1990 where the vote share approached two-thirds). Hungary is not so far out of line with the rest of Europe in terms of its political behaviour, which leads one to another conclusion: the polls have been exaggerating the FIDESZ leads among certain voters because of methodological flaws. And, let's face it, they have a record for this. They did it in 2002, and before the EP elections in 2004. And the polls conducted that were published on the evening of the 7th June itself suffered from the same flaw - which suggests that this isn't a sudden change in voting preferences. Let's face it - they got the size of the FIDESZ vote wrong, and they are now just a bit more correct than they were. And, if this is the case, we need to be very careful in drawing anything other than fairly simple conclusions from them.

[Eva S. Balogh]

Mark: "This isn't quite what I'm saying - in fact I'm absolutely sure that some of the more extreme interpretations (that five-sixths of the Jobbik vote came from the MSZP)are wrong."

As you know very well there is no certainty when it comes to the secret ballot. (See the chief justice's election when both sides claimed that the other side had voted against him.) One can only approach the question by indirect means. There are just too many signs that the approximately 14% of Jobbik voters came from MSZP is more or less accurate. No deep analysis will ever reveal the exact numbers.

[Mark]

Éva: "One can only approach the question by indirect means."

It can be done relatively reliably through using the tools provided by regression analysis to compare the detailed results of two sets of elections in order to provide an account of the movements of votes between parties and non-voters. Because the electoral authorities in Hungary publish results by polling place it should be possible to do this with a very high degree of reliability. It requires a nice big computer and people to input the data. This method is used routinely to compare the results of two sets of elections in countries other than Hungary, and companies like Infratest Dimap, for example, have been doing this for years for public service television in Germany. This kind of analysis will tell us infinitely more than opinion polls with tiny samples and flawed methodology ever will about where Jobbik's voters came from.

[whoever]

Mark talks of "a larger working-class constituency which votes for the left on bread-and-butter issues, but when asked about cultural issues (law-and-order, the death penality, immigration, race)tends to have attitudes that place them on the hard right."

This working class does exist in Hungary and it is disenfranchised in the way that you describe. It isn't clear, though, that it maintains its own institutions - trade unions - here in the same way that it would in France, for example. This would make the appeal of the far-right far harder to counter.

To make things more complicated, younger generations are aspirational in a sense, and the vast majority of students appear to go into further study after leaving school. (I haven't got the numbers for this). It's not clear that these younger people would consider themselves working class at all after a lengthy college education, often geared towards business needs. The crucial question is: what future will exist for this group - what jobs? And who will they blame?

[Gábor]

"I suspect that the "new illiteracy" is more pervasive than we would imagine, and that the ignorance this engenders is creating a fertile bedrock for demogogues and cranks."

I wouldn't call them illiterates in a strict sense, just look at those important elements of the extreme rightist subculture, Schythia Bookstores - it's actually a chain - Gede testvérek, or just consider the success story of the Wass Albert texts. It's a flourishing business, or at leest seems so and the youth belonging to this culture reads quite much.

The problem is far more complex, I suspect that it begins with the public instruction system, where the state (more precisely local authorities) run system is slowly degraded, while a small island of private schools emerged seen as the last resort of quality teaching. But for example as the larger part of those schools are church run ones, you can imagine how the quite nationalistic clergy and church authorities in Hungary will indoctrinate their pupils.

As the youth enters a university they won't really find anything (at least at History Departments), but a lot of disillusioned and cynical teachers, who are still living and teaching in the '80s, who manage a "scientific" system based on patronage, favoritism and corruption, whose authority is non-existent and therefore what they teach lacks any credibility. Not to speak of the conflict between personal experiences and theories.(And as the radical and extreme right is quite active in student self-governments they have first hand experience that this old guard of teachers are rady to make the most doubious deals with them, crossing the floor from left to the extreme right if it comes to money and influence...) As I've met some good colleauge as a student who started as dumb, one-sided, biased people and due to their willingness to learn and the influence of some good teachers they completely changed during their years as students, I would conclude that it is fairly important. And emphatetically I can understand those youngsters that they are fascinated with a fantastic story protraying their nation as the bravest, most honest, kindest etc. as they've read it in their books...

[Sophist]

Got to agree with Gábor: in my experience - as a highschool teacher - the schoolkids advocating the "szebb jövőt" are more likely to be articulate and literate, what they lack is the logical skills to discriminate and the role models advocating sg positive about the status quo.

There is also a significant generational thing going on, in that new voters are now born after the system change. The charms of social and economic liberalism don't appeal to them, because they - unlike their parents' generation - haven't experienced their absence.

I actually find the anti-consumerism the civic activism refreshing - reminds me of CND and Red Wedge in 80's Britain. But here and now, it's unfortunately allied to "racism" and irredentism.

[Gábor]

I actually tend to agree with Mark regarding the Median survey on the composition of Jobbik's voter base, the problems with opinion polling, the importance of the rate of participation in determining the elections' outcomes etc. I have my doubts concerning Eva's disbelief on Jobbik's capability to appeal to former MSZP voters. But I don't think that those polls are necessarily problematic, questionabble etc. First of all they didn't really registered any movement except in the case of Jobbik, but it can easily be what you mentioned, Jobbik supporters becoming less afraid of publicly declaring their allegiance. The figures for the total population - the reliable ones as the other derivatives are quite shaky because of the usually false prediction of voter turnout - didn't change significantly.

On the other hand it is not necessarily justified to take the EU elections as an gigantic opinion poll and compare polls taken on voting intentions for a parliamentary election to it, because of the low and uneven turnout. It doesn't necessarily means that the MSZP has more chances at the latter, but it is possible that other factors are driving the crucial decision on whether to participate or not in the case of a general election, therefore somewhat changing the distribution of votes as well. And as the group of the so called "biztos szavazó pártválasztók" is a realtively small group - but usually larger than the turnout figure was at the EU elections -, it is not inconceivable that those realtively small differences to your expectations are real ones.

Otherwise I would list the MSZP as one of the parties the support of wich is not really popular and therefore giving reason for voters to hide their opinion from pollsters. I'm aware of the fact that it probably does not mean more then 2-3% at the best and the MSZP has no large hidden reserves to mobilize at election day, but this ammount of voters can be responsible for such unexpected movements as well. Not that it would bring a bright prospect for MSZP either their votes turned to Jobbik or are only determined to remain at home the yhave serious problems...

[Eva S. Balogh]

Gábor: "I have my doubts concerning Eva's disbelief on Jobbik's capability to appeal to former MSZP voters."

I don't remember saying anything like that. According to Medián (and I was just listening to Gergely Karácsony, Medián research director on Kontra, Klub Rádió) Jobbik drew from everywhere, including MSZP BUT most of their voters came from the right or from youngsters who three years ago were not old enough to vote. Karácsony was bitterly complaining about so-called political scientists and certain pollsters who serve political purposes and disregard "facts." He was enraged by irresponsible commentators who without any proof kept saying that the majority of Jobbik voters came from the left.

[whoever]

Obliged to clarify - I absolutely agree that the younger people in the radical right in Hungary can be often articulate and literate, in my experience.

What I am saying is that mainstream modern Hungarian culture is anti-intellectual. I'm not talking about the paid-up activists - but the people who make up the majority of young people. So in the absence of a popular but intelligent mass culture, the space for the far-right to operate seems enormous.

The public libraries are run-down and I guess so are the school libraries. The culture of enquiry doesn't exist - perhaps it never did, but in order to gain ground in the Party, at least people were forced to read a bit of something theoretical.

[Sophist]

whoever,

My school library is a sad joke - we possibly have more books at home, but this represents 3 generations of Sophists - the local library isn't, nor is the new children's library I visited recently.

But the book industry is deeply surprising: three major chains, Alexandra, Libri, Lira and Lant have large shops here, a county town. Knowing that my students aren't reading anything other than Twilight, I have to wonder who is keeping all these guys going .... the mafia laundering money perhaps?

Having just read "Darkness at Noon", I should point out that the Party did a good job at murdering "cultures of enquiry." Social and Economic liberalism has done nothing to resurrect them.

[Mark]

Éva: "Karácsony was bitterly complaining about so-called political scientists and certain pollsters who serve political purposes and disregard "facts." "

I absolutely agree with him - but he might just be in a stronger position if Medián itself was more transparent about its methods. Medián hasn't put its research report up yet on its own website, and - in contravention of every commonly accepted international standard in the reporting of opinion poll data - the HVG article contains no information whatsoever about the survey its figures are based on. So, we know nothing about overall sample size, dates, methodology, or margin of error. It is at least common practice in the United States and the UK for polling companies to publish the raw data on their website to enable independent checking. Without this transparency, how does anyone know whether this survey is based on any more on "fact" than the opinions of "so-called political scientists".

This is quite important because it really makes all the difference as to whether this was a focussed survey of Jobbik voters with a decent sample size (say 400, for this purpose), or whether it was the subset of a standard survey of public opinion (probably 6% of a 1,200 person poll) which would have a margin of error of around + or -20%. So, that figure of the proportion of ex-FIDESZ voters could be anything from between 15% to 55% of the total, and the MSZP anything between 0 and 34%! It really does make the world of difference.

[Gábor]

" the HVG article contains no information whatsoever about the survey its figures are based on"

Actually it is on the illustration showing the popularity of politicians. As the results regarding Jobbik voters' composition are in a complementary article it is the most probable that they surveyed the subset of the sample, 8% of the 1200 people. Not that reliable, although common practice in Hungary, Századvég used it for ages and now Forsense and Nézőpont (obviously among the pollsters serving political purposes, it is funny to have a pollong firm the results of which are always in full compliance with Fidesz's political messages) are always ready to make serious-looking analyses of the opinions of the voters of different parties reagrding important policy issues.

[Gábor]

Éva, you are eager to dig out every bit of information, interpretation, data, assumption undermining the idea that Jobbik's breakthrough is a result of its appeal to former Mszp voters. I do not condemn it, quite the contrary, having an advocatus diaboli is really advantageous for the discussion and you have quite legitimate points. But on the whole you are more inclined to accept that Jobbik's result is due to a flow of radical Fidesz voters and former non-committed rather than leftists. I was only summarizing the lengthy debate in one sentence.

Otherwise I don't think that even if Karácsony and Medián would have right, the Mszp would have any chance to survive with supporting Bajnai's government wothout objection. Not that I think its policies are completely flawed, based on a wishful thinking and a very limited knowledge even of economic thinking and theories, it is almost irrelevant in this case. Those policies simply won't deliver any substantial result till next year and nobody will be really thankful, while the Mszp will simply seem to be a party of endless spending cuts and austerity measures. It can be appealing to some maniac economic liberals, but even the business circles favored by the almost complete implementation of the Reform Alliance program will remain only contemptous towards the socialists. (Just loook at Parragh László's announcements yestreday (?).)

[Eva S. Balogh]

Gábor: "Otherwise I don't think that even if Karácsony and Medián would have right, the Mszp would have any chance to survive with supporting Bajnai's government wothout objection."

I'm afraid they will have to support the Bajnai government. They have no choice. Otherwise the country will go to the dogs. And the party with it.

What Parragh says? I don't accept anything else from him.

[Gábor]

Oh, I'm always delighted to read so sophisticated argumentations. I'm completely disarmed and overwhelmed by the thorough and detailed reasoning.

[Mark]

Gábor: " As the results regarding Jobbik voters' composition are in a complementary article it is the most probable that they surveyed the subset of the sample, 8% of the 1200 people. Not that reliable, although common practice in Hungary."

Well if that is the case all we can say with confidence is that they took significant numbers of votes from MSZP, FIDESZ, themselves, non-voters and first-time voters, but the margin of error is so large as to render the percentages meaningless.

As for it being common practice, it is clear from the way the results of such surveys are written up that there is a marked lack of understanding of methodology. Such articles end up being little more than unsupported opinion liberally laced with a few percentages deplyed by the author for rhetorical effect.

[Eva S. Balogh]

Mark: "Well if that is the case all we can say with confidence is that they took significant numbers of votes from MSZP, FIDESZ, themselves, non-voters and first-time voters, but the margin of error is so large as to render the percentages meaningless."

The two of you, Gábor and you, are trying too hard to pin the success of Jobbik on MSZP voters. Everything, but everything we know up to now contradicts it. Common sense also does. It is enough to see the huge drop in the Fidesz numbers and it is clear that the Jobbik voters came from there.

[Mark]

Éva: "Everything, but everything we know up to now contradicts it."

We have to be very clear that it doesn't. What we do know suggests two things:

1) That thinking about this in the terms of the bi-polar left-right split and trying to classify Jobbik as "left" or "right" is flawed. In fact, as a way of thinking it says more about political prejudices of those advancing the argument (either way) than it does about Jobbik support. What we are seeing is the emergence of a new form of political division with a large right-wing party (FIDESZ) in the centre of the Hungarian spectrum, flanked by two smaller camps - the remains of the left, and a right radical one. The real question is how we describe that new camp.

2) Most committed FIDESZ voters are actually much more enthusiastic and committed to voting for their party than in the 2002-2006 cycle. Therefore the rate of attrition of votes from FIDESZ to Jobbik accounts for no more than a third of Jobbik's total support. Whereas you suggest Jobbik's emergence poses a difficulty for FIDESZ in the 2010 election, this conclusion would lead one to believe that the real threat Jobbik poses to FIDESZ will be after 2010 once it is in government.

Let me just say something about the evidence:

1) Given the inaccuracy of Hungarian opinion polls no reputable political analyst outside the country would base any conclusions about where a party's support came from on the basis of month-by-month fluctuations in the headline figures. I doubt actually that they would even if they had been proved reliable. It is just not an appropriate or safe use of the evidence they provide (certainly in connection with US presidential elections there is quite a lot available on what can and can't be done safely with opinion polls).
2) The only safe way of explaining where a party's vote came from is to subject the votes themselves to proper statistical analysis. My poor computer is just not large enough to do this properly, but I have done this in a very rudimentary fashion. This reveals three things relevant to this discussion - (1) FIDESZ voters are far more likely to turn out and support their party than before 2006 - they are very committed to their party - so a high rate of attrition to the far right is unlikely, (2) the geographical distribution of turnout suggests that where Jobbik is strong, turnout is higher than one would expect given the regional distribution of voters in previous elections - in other words low voter participation is not an acceptable explanation of high Jobbik percentage shares, and (3) if you look at the regional distribution (by county) of the MSZP + Jobbik share and compare it to the county distribution of the 2006 MSZP share there is an almost perfect correlation. In other words where the MSZP vote fell most, Jobbik gains were the greatest. The movement of votes from the MSZP is almost certainly greater than that from FIDESZ to Jobbik.

If someone does proper statistical analysis of the votes and proves me wrong that's fine, but no-one is even attemtping it.

[Eva S. Balogh]

Mark: "Éva: "Everything, but everything we know up to now contradicts it."

We have to be very clear that it doesn't."

I'll give up. As for the accuracy of Hungarian polls. Yes, there are some which are terrible. Nézőpont, for example. But the other three polls that came out lately are pretty close to each other and that means something. I'm not saying that the voters come from the left or from the right but that it seems that more votes came from Fidesz than from MSZP. I know that Mr. Stumpf would like to see it otherwise and came out boldly with the notion that Jobbik had done a great favor to Fidesz because it took away votes from MSZP. Wishful thinking.