Tag: public opinion

Slovakia Dashboard News: At The Last Minute (or Catch a Falling Smer)

This is, I think, the last “dashboard” post for quite some time.  The next time you see me post on this kind of thing, it will be an “election” post, but for the moment we have the kind of unusual situation we usually only see just before elections: three major polls appearing on the same day.  Let me follow my usual pattern and deal with these party-by-party with a few words about coalition v. opposition.

Smer took a huge hit this month, with its lowest results in years: since May 2005 in FOCUS, since June 2006 in Polis, and October 2006 in MVK, an average drop of 4.5 (slightly more in FOCUS and Polis, somewhat less in MVK).  Why the drop should be so large is a bit of a mystery to me.  some journalists attribute it to party financing scandals, but I have a hard time believing that that news was particularly surprising or likely to pry voters away from the party.  I’m more inclined to think that it’s a bit of frustration by soft Smer supporters forced finally to think about making their choice (and it’s notable that SDL has risen significantly in the polls for which we have information, suggesting voters looking for the next best alternative, particularly those with more culturally liberal values).  It’s important to remember that in 2006 the final-week drop in Smer did not play out in the actual election and that the more accurate poll was one taken two weeks before the election, so some of this may be ephemeral.  But nobody in Smer can be happy today.  And for a party which has embodied the slogan “nothing succeeds like succcess,” some must be thinking of how to avoid failing like failure.   I reprint the graphic from the dashboard here only because it is so dramatic:

SNS gets a reprieve this month, probably thanks in part to the assistance of Hungary’s Fidesz, with a 1.5 point gain in FOCUS, a .7 point gain in Polis and no gain at all in MVK (which, however, showed a 1.2 point gain in its previous poll).  National issues may count more than clientelism for some voters and the SNS campaign on this question (which some see as quite effective) may have helped here.  It is hard to say whether the party will lose more from time in opposition (lost clientelist revenues, but time re-purify its image and play the outsider) or another stint in government (posts and money but ever more chances for people to find out how those were obtained).

HZDS also gets a small reprieve losing slightly in FOCUS but recovering to some degree in MVK and Polis, for an overall average of 5.2, far too close for anyone’s comfort.  This recovery may actually help it a bit as those who were on the fence for the party feel comfortable voting for it one last time, but its overall negative momentum and air of decline may be to hard to overcome.  This one is very tough to call

Overall the current coalition dropped three full points in June, to an overall average of 42.0, a remarkable drop for a coaltion that in less than two years ago polled 69.8.  Smer alone had poll averages of 41.0 as recently as January of this year.  The drop is so quick that it is hard to fully accept it and I suspect the overall election final will be a bit more, but we need not wait long now.

Polls of SDKU usually lack a clear monthly pattern and this month is no exception: stable at a high level in Polis, dropping from a middle level in FOCUS, rising from a low level in MVK.  The median stays around 14 where it has been for quite some time.  For SDKU it is especially hard to say whether poll numbers are related to final numbers as for the past 4 elections the party has outperformed its poll, though Martin Slosiarik and others note that the emergence of SaS may diminish that undercounting based on last minute shifts.

KDH has some of the same low-level chaos as SDKU.  No big trends like Smer or HZDS, but lots of movement and poll shifts ultimately adding up to 10% (as it has more or less for almost two decades).  This month the pattern is converging: Polis and MVK dropping from high levels to just over 10%; FOCUS rising from low levels to just under 10%.

SaS finally falls back into the earth’s gravitational pull this month, still rising but by a lower margin (.60) than in all but one month since October 2009.  Both FOCUS and MVK show it stabilizing at around 12 points and while Polis still shows a rise it is to that same 12 point level.  How much of this the party will sustain in the election is an open question: past new parties in Slovakia have lost in the voting booth, but as I’ve mentioned elsewhere, TOPo9 and VV in the recent Czech election managed to mobilize voters.  Could this have something to do with their mastery of social networks and other technological turnout mechanisms?  Hard to say, but if it does, then SaS might manage the same trick.

It is interesting that despite the significant drop in the current coalition, the Slovak right did not see corresponding gains this month.  In fact it dropped slightly from 36.7 to 36.1 (suggesting that supporters of the current coalitions are going elsewhere, either from Smer back to HZDS and SNS or to “new” parties like SDL).  This probably is not bad news, as it suggests a certain solidity to the overall vote total of SDKU, KDH and SaS (and indeed the core vote of this population has been quite solid at about this level from one election to the next and its relative success in seats has been affected more by the distribution of the vote between parties over and under 5%.  This year despite lots of expectations to the contrary even a year ago, the right is relatively coherent and, thanks to the small/new party vacuum effect of SaS, should lose little to small parties and so has a good chance of getting seats in proportion to its base).

Both Hungarian parties continue to pass the threshold in all major polls, if only by a hair.  This continues to astound me: if you take two parties whose total support averages 10.6 for the last year and divide the 10.6 at random the chance of getting two parties above the 5% threshold is itself only about 5%, and yet these two parties continue to manage that 1 in 20 shot at not undercutting Hungarian parliamentary representation in exchange for a small chance at maximum gain (though this of course is not what the two party leaders themselves are thinking).  We will see very soon whether their luck will hold out.

Finally, I think it is necessary to say a word about SDL about which I have said nothing for the entire campaign, largely because until this month it averaged less than 2% and never exceeded more than 2.8%.  Suddenly, the party has jumped by a significant margin in every poll and stands at 3.8% and is staring closely at the 5% threshold.  Only two other non-parliamentary parties have exceeded even 3.3% in the past four years and both of them–SaS and Most-Hid–have a good chance of getting into parliament.  It is doubtful that SDL will be able to cross that remaining 1.2% in the final week (SPOZ in the Czech Republic could not manage it, though that’s not much of a guide here) and it is likely that its preferences reflect frustration that will translate into staying home or reluctant Smer voting, but its emergence is a sign of weakness that Smer does not wish to have revealed:

Slovakia Election Update: Pictures worth at least 300 words

I love polling results and I am always surprised and unhappy when they are abused.  Recent articles in SME and Pravda put needless stress on these results by, on the one hand, treating them as unique and absolute indicators of a party’s success or failure (at least until the next one comes out) and, on the other hand, lumping them together into time series regardless of their origin or methodology.  Toward this end, I propose that we fix this by, on the one hand, treating each poll as a unique series which is to be compared to other series, and, on the other hand, when we do lump them together into time series, we smooth them out and look less at individual polls and more at overall trends.  I’ve tried to do both of those things here with some overall coalition/opposition graphs that, I think, speak for themselves in various ways (though that probably won’t stop me from trying to speak for them at some future point.

Coalition, Slovak Right and Hungarian total support according to various polling agencies (and averaged), 2008-2010

Monthly averages of public opinion preferences, smoothed with LOESS, in Slovakia 2002-2010

Monthly averages of public opinion preferences for all major parties in Slovakia, 2002-2010, smoothed with LOESS

Opposition party support, smoothed

Polls, Politics and Parties, Part 9: How Experts (and Bookmakers) Second Guess the Polls

In the eight previous posts in this series (and in this blog in general) I’ve used public opinion as the basic raw-material, but pollsters in Central Europe are quick to note that public opinion only talks about the “current” state of affairs and does not predict what people do once they enter the voting booth.  They walk a fine line on this point, between irrelevance on one hand (if pollsters admit that polls do not predict elections, why should we pay attention to them?) and embarrassment on the other (when the “normal” 24% gap between polls and actual results causes people to notice that polls do not do a very good job of predicting elections).  But the world is full of people who, like myself, who are uncomfortable with waiting for news and need some early indicators, at least, of how things will turn out (my wife has learned from my uncomfortable fidgeting when she says “I have news…” to include a spoiler like “and it’s not bad news, or at least not very bad news.”  This inability to deal with uncertainty in the political world drives me (and perhaps a few other people) to use whatever resources are available to look for the best way to predict the future.

If polls aren’t it, then what is?  In a previous post, I’ve looked at alternative quantitative measures for second-guessing the polls–committed voters, committed turnout, past trends–of which only the last seems to have much value (hence my rather cautious use of it).  But the human brain is a marvelous thing, and the answers may not be easily found in the aggregation of statistics, however interesting they are in their own right.  The secret is to find the right brain, or perhaps better said, to know which brain is the right one.  In this we are limited to those who are willing to share the Slovak-election-related contents of their respective brains.  Ideally we would have a significant number of people who have looked carefully at the available information and made an educated guess.  How do we know if they have done so?  Well without close attention to their study habits, the best way is to see whether they have something to gain if they are right (or something to lose if they are wrong).   The discussion boards of SME, Pravda, HN and other news outlets are full of people willing to hazard a guess, but with few consequences for failure (especially since so many are anonymous).  I am left, therefore with two main sources:  those who stand to lose their reputations and those who stand to lose their money.  Among those with reputations at stake, I do not include politicians, whose prognostications are never expected to be true (Pal Csaky predicts 10% for SMK) and so we are left with a small cadre of political scientists and pollsters who are willing to make their guesses public, and I will reference them here.  In particular, I am grateful to Martin Slosiarik (this is only one of the many realms in which I owe him gratitude) for being willing to make his predictions public, and to make them narrow enough to be useful (though he cleverly shies away from one of the hardest questions).

Among those with money at stake, there is the growing field of those who bet on politics (actually has been extremely common in certain past eras) and a whole discourse has arisen about the predictivness of odds markets.  In my own experience, these markets have done quite well in predicting obscure (to me) local races in the US.  They are subject to manipulation, of course, but the more people become involved, the more difficult this becomes.  Slovakia lacks a true odds market for politics.    SME‘s “crystal ball” at http://zajtrajsie.sme.sk/ is a great first step, but the gains, as far as I can tell, are not particularly valuable (the S€ used in the betting cannot yet be used to buy beer) and so participants may make guesses without much genuine forethought, and since this is the first attempt, we won’t be able to tell until the results are in and the markets close (though this is something I will follow closely).  The only other source we have are the people who make their money by encouraging others to bet.  Bookmakers have a strong vested interest in setting the odds right and they can only do that if they make the right kind of predictions.  So in addition to the political scientists and pollsters, it is interesting to look at what the bookies say.

What I have therefore done is to create charts for each party that show the range predicted by Martin Slosiarik, a rough guess based on current poll average projected out two months based on current trends (my own rather superficial method, though one that did not work badly in the recent Czech election) and then the four bookmakers I have been able to find.  In each case a dark bar represents the baseline: in Slosiarik’s case it is the center point of his predicted range, in my case it is the current poll average, and in the case of the bookmakers it is the inflection point of the bet itself (“Will SaS exceed 10% or fall short”).  The small vertical line leading from it shows direction: in Slosiarik’s case the lines show his full range; in the case of the average poll results it shows the what the “two-month-out” method would predict; in the case of the bookmakers, it is a rough assessment of which side the odds favor.  The longer the line, the more the odds favor a result in that direction (it is not a direct measure of what result the party will actually achieve).

What do we learn from these?  For lack of a better method, let me go party by party as I usually do with poll results:

Smer.  I will begin the analysis of Smer numbers in reverse order on the graph below, beginning with the four wagering firms (British firm bwin.com, and firms operating in Slovakia, Doxx, Fortuna and VictoryTip [unlike last election, Nike.sk and Tipos.sk do not appear to be participating].

Betting odds can be hard to compare because the bets of this type are intrinsically binary and must begin from a particular baseline.  In the case of Smer, each of the four begins from a different baseline, though these are relatively close together, spanning a narrow range between 32.0 and 33.5.   These baselines appear in these cases to reflect the odds firm’s baseline guess, and the odds then suggest its subsequent adjustment.  For Bwin and Doxx the odds were even at last check, but both Fortuna and VictoryTip put slightly shorter odds on Smer receiving more than their respective 32.5 and 33 suggesting an outcome slightly above 33.  Average these together and you get an outcome of about 33% predicted by the odds markets as of 8 June.  This is not far from the 34.8 minus 1.4 = 33.4 suggested by the (two-month-out”) poll prediction.  Interestingly, however, it is rather higher than Slosiarik’s 30-33 range or the 30.5/30.6 prediction I derive from two separate but highly consistent SME odds markets (http://zajtrajsie.sme.sk/stavka/pridajstavku/447 and http://zajtrajsie.sme.sk/stavka/pridajstavku/520).  Which of these is more accurate is impossible to say at the moment.  SME readers tend to be somewhat more free market oriented than others and this may represent a bit of wishful thinking.  Slosiarik, on the other hand, knows these numbers inside and out and and seems convinced of the softness of Smer support (he may well be right, but we won’t know for 4 more days).

SNS. For SNS the baseline measures of three of the four betting firms is the same–7.5 while the fourth is lower at 6.5.  All but one suggest a slight downward trend with an implied average of around 7.0.  This is also the median of Slosiarik’s 6-8 range.  It is interesting that for this party–the betting and expert models are most out of sync with the average polling results, which are under 6 or the trend, which would point the party even lower.  This may be because these experts believe in the (not insignificant) effect of recent Hungarian policy decisions and/or believe that some Slovaks may be hiding their vote for a party that says publicly what some people think privately but might not be willing to admit.   It is difficult for me to disagree with the experts in order to engage in some wishful thinking of my own.

HZDS. HZDS is even more consistent than Smer or SNS.  Only one betting firm even put HZDS’s baseline above the magical 5% threshold and that one, like all the others, put shorter odds on the party falling below the baseline, which in three out of the four cases means below the threshold for entry into parliament.  Likewise Slosiarik puts HZDS in a 4-6 range that sets its median at 5%, though he carefully does not make any predictions about whether it will be slightly above or slightly below.  The opinion polls here weigh in on the negative, both in terms of absolute numbers and two-months-out trend which puts it closer to 4 than to 5.  (The SME odds markets also give HZDS a less than even chance of returning to parliament: 42.8:57.2.  (http://zajtrajsie.sme.sk/stavka/pridajstavku/87)

SDKU. SDKU has a relatively narrow range of baselines but rather broader range of odds spreads (even though one of the firms, Bwin, does not include the party among its range of bets.  The range hovers between 14.5 and 16.  Public opinion polls put the party at 15 with no discernible trend while Slosiarik puts it higher, on the grounds that it may pick up some voters who at the last moment decide that they are not comfortable voting for SaS (This may be the grounds for Fortuna’s quite short odds on the party receiving more than the baseline).  Still, even this does not change the range particularly.

KDH. KDH produces very consistent results: 9.5 or slightly above.  This is true whether you look at the four betting firms, all of which chose the same baseline, or at Slosiarik’s 8-11 range or at the current polls which put the party at 10%.  This is, not coincidentally also the party’s basic result for nearly every parliamentary election in which it has run independently since 1990, though that itself is rather odd since the party has changed, its voters have changed, its leaders have changed, the country has changed and yet KDH manages the same old 10% every time.

SaS. SaS shows a relatively narrow range of baselines–from 8.5 to 10.0 with odds pointing in opposite directions (the lowest baseline, Doxx giving short odds for a higher score and the highest baseline, Fortuna, giving short odds for a lower score.  They meet somewhere around the middle–around the 9.5 that is in the middle of Slosiarik’s range.  As with SNS, SaS odds and experts differ from the polling average for the party and its trend, perhaps reflecting insider knowledge (or the belief therein) that new parties poll high and then end up lower (as I’ve discussed above, this was true for almost every new party to enter Slovakia’s system: ZRS, SOP, ANO, Smer, HZD, SF) but, interestingly, it was not true for the Czech VV which in many ways is quite similar to SaS or for the rather different but equally new TOP09.  I tend to agree with the experts in pegging SaS down a few points but the Czech case gives me just a bit of pause.

MKP-SMK. MKP has an extremely evenset of baselines–between 5.5 and 6.0 with only the barest hint of a trend in one or two. This is identical to the public opinion average.  Both are lower than Slosiarik’s range of 6 to 8.  The betting firms may just be playing it safe here since there is really no way of telling here.  Slosiarik clearly expects voters to fall back to MKP from Most (just as he seems to expect them to fall back to SDKU from SaS.  As above, there is good reason to agree but past precedent may not always dictate current action.  This is the question I will be most curious to see answered on election day.

Most-Hid. Surprisingly to me, Most-Hid has the most consistent set of baselines–all at 5.5–and odds (nothing shorter than 1.75 or longer than 1.85).  The reason that this is surprising is that Most-Hid has perhaps the most unpredictable electorate.  We really have little basis for judging whether they will switch back to the tried and true MKP at the last minute, as Slosiarik clearly thinks many will.  He says Most-Hid will be “very close to 5%” but if you look at his prediction for MKP-SMK and assume that the total electorate for Hungarian parties is less than 12%, his actual prediction seems to be slightly lower than 50-50 that Most-Hid will make it into parliament, something that he probably avoided saying directly lest it produce the headline “FOCUS expert rules out Most-Hid”

So what happens when we put all of these together?  The table below gives a rough estimate of the median point of the odds-makers. (There is no simple way to calculate these because there is no pure way to factor in what the odds mean for preferences.  Very short odds on a party that has a baseline of 5.0 might mean that the pollster is absolutely certain that the party will finish with 4.9 or that it will finish with 2.0.  Of course all things being equal the shorter or longer odds are likely to be somehow proportional to the distance from the baseline since that distance is the most probable basis for certainty.  If there’s anybody in the odds business who can set me straight, I’d very much appreciate it.)

Party Odds Markets Slosiarik
Percentage Seats Percentage Seats
Model Minus
Most-Hid		 Model Minus
Most-Hid		 Minus
HZDS		 Minus
Most-Hid
and HZDS
Smer 33.0 58 62 31.5 52 55 55 59
SNS 7.0 12 13 7.0 12 12 12 13
HZDS 4.8 0 0 5.0 8 9 0 0
SDKU 15.5 27 29 16.5 27 29 29 31
KDH 9.8 16 17 9.5 11 12 12 13
SaS 9.0 10 11 9.5 16 16 17 17
MK 5.5 17 18 7.0 8 0 9 0
Most 5.5 10 5.0 16 17 16 17
Current Coalition 44.8 70 75 43.5 72 76 67 72
Current Opposition 45.3 80 75 47.5 78 74 83 78

The results are not particularly auspicious for the current coalition as it depends on two narrow chances: HZDS in parliament and Most-Hid out.  Without both of these conditions–the odds makers narrowly predict HZDS’s exclusion–the best Smer could hope for (according to the oddsakers, at least) would be for the failure of Most-Hid which in this case would produce a 75-75 split.  Otherwise, Smer will need to reach across the aisle or face the possibility of an opposition coalition.  Slosiarik’s slightly different numbers produce the same conclusion.  Only the presence of HZDS and absence of Most-Hid produces a coalition majority in his model, and then only a very narrow one.

Indeed the striking piece of information here, beyond the importance of thresholds to which I’ve alluded on numerous occasions, is the narrowness of governmental margins.  Slovakia may be entering a period that looks a bit like recent Czech political history with fragile or even minority governments (especially if MKP-SMK were to become untouchable”).

Of course all this depends on whether the odds makers and experts know what they are talking about.

P.S. Want to bet on this?  You can figure out from the above where the best odds are.  Here are the oddsmakers:

https://www.bwin.com/politicsB

Bureau of Meaningless Statistics: The (Non)Effect of Undecided Voters

There is an old saying that “figures don’t lie, but liars do figure” (which I’m sure has some equivalent in almost every language) and there is a wonderful book written in 1954 provocatively entitled “How to Lie with Statistics.”  In Slovakia’s election coverage in 2010 the challenge is not so much statistical lies as lazyness.  Figures appear from various polling firms and they are duly published by newspapers that what people to pay attention whether they have solid basis in fact (nobody’s lying, per se, but they also have no way of knowing whether they are telling the truth) or whether they have any impact.  As a case in point take today’s article in SME, “Smer and SaS can score among undecideds.” (HN does the same)  It is, perhaps, interesting that this is the case, but the article makes little effort to deal with the two real underlying questions:

  • First, is this a useful way of adjusting polling numbers? I don’t know, but neither does SME.  I don’t have any evidence imediately at hand, though I will look to see if I have any precedents from 2006.
  • Second, if this were useful in adjusting numbers, would it have any effect on the overall outcome.  Here the calculation is the work of about 5 minutes at a spreadsheet (use focus data to figure out the overall share of undecideds, multiply this by the percentages printed in the article, add this to the original percentage gained by the party, recalculate to equal 100).  The results are in the table below.  And the answer is “not much”
Party May Poll Share Share among undecideds Contribution of undecideds Revised preferences
(sums to more than 100)		 Revised total share
Smer 35.3 16.9 2.6 37.9 34.6
SDKU 14.0 6.6 1.0 15.0 13.7
SaS 13.3 12.8 1.9 15.2 13.9
KDH 8.3 5.8 0.9 9.2 8.4
SNS 6.1 3.5 0.5 6.6 6.1
MK 5.9 2.8 0.4 6.3 5.8
Most-Hid 5.6 4.6 0.7 6.3 5.8
HZDS 5.1 3.5 0.5 5.6 5.1
KSS 1.6 1.6 1.5
Unie 0.8 0.8 0.7
SDL 0.8 0.8 0.7
Paliho Kapurkova 0.7 4.0 0.9 1.6 1.4
ND 0.7 0.7 0.6
EDS 0.7 0.7 0.6
ZRS 0.6 0.6 0.5
Nase Slovensko 0.4 0.4 0.4
SRK 0.1 0.1 0.1
Azen 0.0 0.0 0.0
100.0 109.4 100.0
Won’t vote 16.1
Undecided 15.1

No party shifts its share by more than 0.7 percentage points, no party drops below the threshold, and the only shift in relative ranking is that SaS slightly overtakes SDKU (and Most-Hid ties MKP-SMK).  And what effect would this have on overall parliamentary outcomes?  Well almost nothing. As the graph below shows, run these percentages through the seat calculator and you get the following results: Current coalition minus 1, current opposition plus 1.

May Parliamentary Seats Revised Parliamentary Seats Seat Change
Smer 57 56 -1
SDKU 23 22 -1
SaS 21 22 +1
KDH 13 14 +1
SNS 10 10 0
MK 9 9 0
Most-Hid 9 9 0
HZDS 8 8 0
150 150

Buried in these results is actually a strong incentive for papers to do the deeper (which is to say not very deep at all, but at least not utterly superficial) calculation.  The coalition v. opposition numbers for the original FOCUS poll (without undecideds) is 75:75.  Add in the undecideds and we get a new parliamentary balance: Coalition 74, Opposition 76.  Had SME only run the numbers, could have run the equally meaningless but far sexier headline, “Undecideds give opposition majority in parliament.”  Maybe it’s a good thing that the busy reporters at Slovakia’s papers don’t have time to do the extra work.

Slovakia Dashboard News: AZEN and the sound of one poll clapping

Almost a month ago, I reported doubts about the poll produced by the previously unknown pollster AVVM.  Last week, AVVM issued a new poll, but though it only increased my doubts, I decided not to report on it because even too-credulous (or circulation focused) Pravda described the results as “strange” (cudny) and a search of other major news outlets produce other stories worth responding to and trying to correct.  And yesterday an article in SME by Miroslav Kern (“Pre-election juggling with preferences“) dealt nicely with the issue in broader terms.  I try not to do what the Slovak press is already doing well (and despite my occasional criticism, there is a lot that they do well) and so I figured there was nothing to write about here.

And then, in a search for results of another idiosyncratic poll (Presov University, on which more later) I found myself on the website of Novy cas.

I generally do not read Novy cas, as it does not have a particularly serious reputation, but I realize now that I should read it more often, for a variety of reasons.  My first reaction to the Novy cas election website (after getting over my annoyance that it required me to download Microsoft Silverlight to even work) was “Wow.”  Not only does the website look great (waving Slovak flag and all) but it had a “Your Ideal Quiz” (because of Silverlight I can’t actually link to it but the page it is on is here: http://volby.cas.sk/) that actually asked pretty good questions, showed the result of each answer on the suitability of particular parties and, when I tried to make answers point to a certain party, actually increased the score for that party.  So far so good.

Novy cas also has lots of party information all in one place: tagged news stories, party list, party program (with an interesting mini-wordle that shows the most common word in the party’s program), and development of party preferences over time.  This last, however, raised some questions: it listed only one poll per week, but polls in Slovakia are not done on a weekly basis so the polls must come from different sources and so cannot be easily compared. It did not list the sources, so I had to reconstruct it by comparing the data to my own poll database.  The first of the four turned out to be FOCUS, the second Polis and the third and fourth…AVVM (a detail that Kern didn’t know or was too polite to mention).  Let me restate that in stronger terms:

The self-proclaimed “most read daily newspaper on the internet” is (without telling anybody) basing its election infographics on the product of an untested polling firm with polls that are highly problematic.

Spot the Problematic Poll

How do we know that AVVM is problematic.  In my previous post on AVVM’s first poll, I showed that the firm’s results were far from the median polling numbers of other firms or even from the overall range of those results.  The second AVVM poll does not change this assessment in any meaningful way, as the following eight graphs show (AVVM is in yellow; the Presov University poll that I discuss below is in pink):

For the current government parties, AVVM’s second poll is not so out of line.  In its first poll AVVM produced exceptionally low results for Smer (orange), though by the second poll this had moved up (and other polls had moved down) until it lay in a more normal range.  Likewise AVVM is in the normal range for SNS and HZDS.

Among the Slovak opposition parties, AVVM produced fairly average results for SaS and KDH but quite high for SDKU.

It is for smaller parties that AVVM produces the oddest numbers (both in real and especially in percentage terms.  Most-Hid results are fairly average but numbers for SMK are exeptionally low–lower even than the low results produced by Median.

Of course the biggest variation from the norm, and the one that attracted Pravda‘s “strange” headline is the party’s results for AZEN, the new party formed by HZDS defectors Zdenka Kramplova and Milan Urbani.  I have not created a dashboard graph for this party because its numbers are actually so low that most polling firms do not report it.  To compensate for the lack of full data, I’ve created an approximation graph which gives AZEN the benefit of the doubt (assuming that it just barely misses the threshold for reporting).  Even giving it the benefit of the doubt, the results are striking:

AVVM results for AZEN are so far from the norm of the other four major firms (and from the Presov University poll as well) that they simply cannot be taken seriously.  It may be that AVVM is right and the others are wrong, but nothing in my experience would suggest that to be the case.

Furthermore, the explanations for this divergence given to SME by AVVM director Martin Palásek are so bad as to disqualify the firm from any further consideration:

  • The first explanation: “the party is first on the alphabetical list of parties” and may therefore draw additional support (To prove I’m not making this up: “Konateľ AVVM Martin Palásek výsledok AZEN vysvetľuje najmä tým, že sú prví v abecede, a tak ich uvádzali aj na anketových lístkoch”.)  If AVVM hasn’t figured out some way to control for this then either a) its polls do not deserved to be published anywhere or b) I will be able to win a significant share of Slovakia’s vote simply by registering a new party under the name “Aardvark Alliance”
  • The second explanation: Party chair Urbani “is among the most popular HZDS politicians” (“ponúkol tvrdenie, že Urbáni patril k populárnym politikom HZDS“).  This is in some ways even worse as it reveals a willingness to follow conventional wisdom rather than the hard data which is the only currency of pollsters (except of course that they accept money from parties to do polls).   In fact my I cannot find any evidence that Urbani has ever appeared among the lists of “most trusted” politicians conducted regularly by MVK, even though these sometimes contain as many as 30 names.  So much for AVVM.

Spot the Ambitious (but Still Problematic) Poll

In the same general category as AVVM but with important specific differences are the polls conducted by university students at Presov University.  These are fairly consistent with the other polls: just a bit high for Smer and SMK, a bit low for HZDS and SDKU and Most-Hid.  For three parties the Presov numbers are further from the norm: low but moving in the right direction for SaS, high and moving in the wrong direction for SNS and shockingly high for Unia.  For those who have forgotten, Unia is a merger of three economic and cultural liberal parties: Slobodne Forum (SF), which split from SDKU in the 2004, Civic Candidates (OK) which split from SDKU in 2008 (I think), and Liga -Civic Liberal Party (Liga-OLS) founded in 2008 by former officials of SDKU and ANO.  Merging three parties with extremely low preferences into a single party and then adopting an entirely new name seems in retrospect to have been poor tactics as since its formation Unia has regularly polled less than the previous totals for its component parties.  Except in the Presov University polls, as the graph below shows:

As with AVVM on AZEN, it is hard to take the Presov University numbers seriously on Unia since they are so far out of line even giving the party the benefit of the doubt on polls where its results are not reported.

It is hard to be as critical of this effort as it is of AVVM, first because it has not been used by a major newspaper without any effort to mention problems or discriminate among its strengths and weaknesses (though that is more the fault of Novy cas than AVVM itself), and second because from everything I can find out about it, it is a laudible educational effort that is open about its methodological limits, cites the geographical areas in which the poll-takers worked, and openly discusses its choice of question and the rationale for it.  I am still not sure that I understand how they actually went about the questioning or whether their decision to use party leader names rather than party names is a good way to measure preferences, but at least they are trying something new and explaining why they are trying it.  For that I give them great credit, even if I don’t feel like I can include their poll results in my overall average.

To Whom From Whom: Slovakia Electoral Shift Roundup

I had intended to post this tomorrow but SME just published an article on ebb and flow for the new parties SaS and Most-Hid based on a presentation by Olga Gyarfasova of IVO and so this is a good time to talk, if briefly, about where Slovakia’s voters are going and where they have come from.

First credit where it is due: thanks to Tim Haughton for finding and bringing to my attention an article in Pravda that presented data from an MVK about the stability of party electorates.  The graph made a decent point about the stability of party electorates (though I’ve argued above that these are not useful in predicting party results based on polls, making the statistic a rather academic one–not that there’s anything wrong with that) but also contained other data about where “non-stable” voters went that is in many ways more interesting because it helps to tell the story of where parties’ ebbs or flows are coming from.  That data, and the scale of the flows, was unfortunately buried in the graph and extremely hard to read.

So, with typical obsessiveness about graphics and a desire to avoid other work, I transformed this chart (more quickly than I thought I could) into something that told the story in a more complete way:

In this graph, I have roughly standardized the sizes of the images so that each pixel represents roughly the same number of voters.  The point is the same–KDH maintained the most loyalty, SNS the least–but from this (admittedly incomplete) data we begin to see the double flow that I have expected for some time: from former HZDS and SNS voters to Smer and from former Smer voters to SaS.  The graph only shows the data we have and so has lots of big residual categories (all other parties instead of individual parties and “undecided” instead of “will not vote” etc.).  It also fails to provide information on the flows to smaller parties or the share that flows from one party represent in the new electorate of the party to which it flows.

Thanks to Olga Gyarfasova, we now have some of that public for the two new parties, Most-Hid and SaS and here the pattern is again more or less what we would expect: SaS gains heavily from SDKU and new voters and disillusioned former voters; Most-Hid gains heavily from SMK-MKP.  There are a few surprises, however: SaS picks up a significant share of voters from Smer, and Most-Hid picks up voters from parties other than SMK-MKP (almost by definition “Slovak” parties) and from non-voters.

This is by no means the last word and all ebb and flow statistics need to pay closer attention to the degree to which flow is caused by people leaving and entering the electorate.  I hope to have more statistics on this in the near future and to improve the graphs above.  Until then, this will probably suffice.

Czech Election Update: Time for the Bigger Picture

A third post on the Czech election. There is a lot to say. I will begin by not quite rejecting the strong temptation to direct criticism to Dnes for celebrating the record high number of women in the new Czech parliament by adorning the article with a drawing of a topless woman and then inviting its readers to vote for “Miss Parliament” based on electoral headshots (I feel compelled to include the link just to prove I’m not making this up, but please don’t go there). It is enough to say, perhaps, that this is sexist even by Central European standards .

Instead I will focus on the question of volatility. In yesterday’s post I plotted Czech volatility over time but did not have time to use the right methods or comparison set. Now that vacation is over, I can fill in the gaps.

First, I can provide references to the articles I cited: the two best recent works I know on volatility:

Because Powell and Tucker provide time series data, I will focus on theirs here, but Mainwaring et al handle the question beautifully as well (particularly their focus on “young” as opposed to strictly “new” parties). My first task was to take the Czech electoral data and use it to distinguish between volatility among existing parties (Mainwaring et al call this intra-system; Powell and Tucker call it Type B) and volatility created by parties entering and leaving the party system (Mainwaring et al call this extra-system; Powell and Tucker call it Type A). Using Powell and Tucker’s method (slightly different from my rough cut yesterday because it excludes parties with less than 2%) produces the following graph which decomposes the red line into its
From this it becomes clear that with the exception of 1992 (when extra-system volatility was huge because of the breakup of Civic Forum and seversl other parties) and 2006 (when extra-system volatility was basically absent), the Czech Republic has had both types of volatility in roughly equal levels.

These levels are low not only with regard to the first electoral period but also with regard to the region as a whole. To convey that point, I steal figures from Powell and Tucker’s 2009 analysis, circle the Czech Republic, and insert the 2010 data. For comparison’s sake, I do the same for Hungary (and in two weeks will do the same for Slovakia). The results are as follows, with the Czech Republic in blue and Hungary in green.

By comparison with the region as a whole, both the Czech Republic and Hungary have exhibited low levels of both types of volatility; indeed in most election years during this period, these two countries have been at or near the lowest overall. On specific kinds of volatility, both Hungary and the Czech Republic have only occasionally been above average on intra-system volatility and never (since 1992 in the Czech Republic) above average on extra-system volatility. In 2010, however, their levels of intra-system volatility rise above the regional average, but so did the previously extra-system volatility, pushing the overall levels from the near the bottom to slightly above the regional trendlines.

In addition to the quantitative resemblance, there are also other similarities.

  • First, both elections show a clear exhaustion with those in power that is clearly motivated by a general disillusionment with large segments of the political elite. Both the Czechs and the Hungarians had, for the most part, avoided this for some time, with large numbers of disillusioned voters holding their noses and voting for old parties anyway or switching to the other big party (accounting for the not insignificant levels of intra-system volatility). The higher levels of extra-system volatility in both countries suggests that these tendencies have diminished and that these two countries have moved closer to the Central European norm of disillusioned voters shifting to new parties.
  • Second, related to the above, both countries saw the emergence of a culturally liberal party attractive to younger, educated voters making extensive use of social networking software: LMP in Hungary, VV in the Czech Republic. (It seems to my uneducated eye that the Hungarian variant is economically more statist, but this seems relatively unimportant to the overall profile.) It is significant that a nearly identical party–SaS in Slovakia–looks set to take a similar share in Slovakia, and that similar parties have done well in the Baltics. There is here something new, not exactly a new party family (though in their cultural liberalism and anti-corruption emphases they share significant elements) and not exactly a new party type (though their methods and organization do not fit fully with any of the currently hypothesized models, even cartel and firm models), but with strong and intersecting elements of both. Nor is it unique to Central Europe alone but elements of it have emerged also in the West. Expect to hear more on this question in this blog. This is something that will need explanation.

Despite these similarities, it is important to emphasize that the election results in these two countries also differ in ways that are quite instructive:

  • First, in Hungary the intra-system volatility represents the evisceration of one major party at the hands of the other–MSZP dropped while Fidesz rose–while in the Czech Republic, all of the established parties lost support compared to the last election.
  • Second, although both countries saw the emergence of new parties, there is a very big difference between the center right TOP09 in the Czech Republic the far more radical right Jobbik in Hungary. This difference is important not only for the political content of the two parliaments (something we see already in the legislative output of the Hungarian parliament) but also for future volatility: a party with a clear programmatic message such as Jobbik may be able, despite its internal factionalization, to survive multiple election cycles whereas the more diffuse TOP09 may have a harder time of it (though because I have just predicted that it will turn out not to be true).
  • Third, there is a significant difference in the role of these new parties: Jobbik will be able to remain outside of government whereas TOP09 and perhaps VV will likely be part of government. Both positions have their risks: Jobbik may be able to avoid responsibility but will also be able to claim few accomplishments and has already had the softer parts of its agenda on national questions pulled from underneath it by the current government. TOP09 and (probably) VV will be able to implement parts of their agenda but will also become responsible for any resulting problems (even those they merely inherit) and will face problems when their new (and relatively inexperienced) cabinet ministers succumb to the same clientelistic temptations as their predecessors. It is interesting to me that VV has openly contemplated staying out of government, suggesting that it has learned the lesson of past Czech “new” parties and those elsewhere in the region. (As Tim Haughton and I have argued elsewhere: new anti-corruption party + government participation = death)

What happens in Hungary and the Czech Republic as the result of these elections will not, I think, have much impact on the broader world (and may not even have much impact on the quality of the daily lives of Hungarians and Czechs) but they are worth this degree of close scrutiny (and more) because what is going on there is indicative, I think, of significant transformations in the relationship between who people are (demographically), what they think (attitudinally) and who they vote for (politically). Demographic patterns and attitudinal patterns still exist but their relationships to political behavior have changed as perceptions of corruption have risen to the top of the list of concerns (and “endurance” becomes shorthand for “corruption”) and as political entrepreneurs take advantage of this change and of new organizational technologies to provide an ever changing menu of new parties (themselves organized primarily for short term gains). For awhile the Czech Republic and Hungary offered some evidence that the trend was not inevitable. With the most recent elections in these countries, it is more difficult to see any alternative.

Finally, from the broader perspective it is interesting that in Hungary and the Czech Republic the election has been regarded as a fundamental shift, a major change in the game of politics despite the fact that the degree of shift was, by regional standards, only about average. Perhaps earthquakes only seem major if you are not used to them, but they still shake buildings. Of course people and institutions figure out ways to survive even where earthquakes are a regular occurance, but their lives are different than they would be in areas with less seismic activity (money is spent differently, personal habits follow different patterns). I hope to spend a good portion of the next few years thinking about how political life is different when every election is an earthquake.

FOLLOW-UP ON PREVIOUS POSTS:

Ben Stanley commented via Facebook: “It’s great being a CEE specialist. Instead of accumulating common wisdoms, we get to throw them all away and make up new ones!”

Cas Mudde comented via Facebook, “Would be interesting to see the same for seats in parliament.”
Here it is, though it is actually rather uninteresting compared to the same calculations with votes.  It suggests, however, that one of the reasons that new parties do not survive is that they never really get started.  US, VV and TOP09 are the only new parties actually to have made it over the threshold since 1992, but this rather technically excludes the Green Party, SZ whose 2006 incarnation is difficult not to describe as a new party despite a certain degree of legal and organizational continuity with the one established shortly after the revolution of 1989.  It will also be interesting to see if, as Sean Hanley wonders, Suverenita manages to use its stronger-than-expected performance to chip away at CSSD in the next election cycle, thus enhancing (or if VV or TOP09 fare badly, maintaining) the red “2010” bar.

Czech Election Update: The (Slightly) Bigger Picture

Another quick and ugly post from vacation (thanks to my family for tolerating my obsession even as we drive from city to city to visit loved ones).  I wanted briefly to put the 2010 Czech election into the context of the Czech party system over time (the next quick and ugly will, I hope, put it into broader regional perspective, if Josh Tucker of the Monkey Cage doesn’t do it first.

For now, all I wanted to do was to post a few pictures about what the most recent elections say in raw quantitative terms about the Czech party system circa 2010.

The big news is that thanks to this election cycle the Czech Republic’s party system looks significantly different today than it did ten years ago (indeed it is closer to 1992 or 1996) and signs are that the current changes will presage more change (or, to put it in a different and more awkward way, a period of stable change as opposed to stable stability)

Let’s start with the number of parties:

While the actual number of parties in parliament (the green line) did not change from last election to this one other measures suggest a substantially increased number of parties.  The red and blue lines show calculations of party system size based on Taagipera and Laakso’s method which show more significant parties in the voting than any time since 1992 and more even distribution of seats in parliament since 1992 as well (and it actually comes quite close to reaching 1992 levels.

The second major difference is in volatility–the number of seats changing hands from one election to the next.  This is actually quite a complicated question because it depends on how we consider succession from one party to the next, but in this election in the Czech Republic the lines of succession are fairly clear (not so in previous elections).  Without going into too much detail, volatility in Czech elections looks like this:

While not approaching 1990-1992 levels (and there is a good argument that even 1990-1992 was not that high), this is the highest volatility the Czech Republic has seen since, with only 61% of seats remaining in the hands of parties that held them before the election.  (As I will try to discuss tomorrow, while this is unusual for the Czech Republic it actually brings it more within the “normal” range for Central and Eastern Europe.)

This shift is even more interesting because of the nature of the volatility.   Both Tucker and Powell and Mainwaring et al have done fantastic work in the past two years distinguishing between types of volatility and suggesting that it makes a difference if the shift is between parties already in parliament or between parties in parliament and new parties.  I do not have time to recreate the calculations of the authors above, but there is another method for displaying it that is perhaps even more provocative.  The graph below shows the share of vote going to parties depending on when they first appeared on the ballot (more or less corresponding to when they were created):

What this graph says to me is two things:

  • Until this election, nearly all of the Czech vote went to parties that were created in the first two years after the fall of communism.
  • New parties appeared, but they almost never survived.  It is remarkable that even though dozens of new parties appeared in the Czech Republic between 1992 and 2006, the combined vote totals for those parties in 2010 was less than 1%.  The Czech Republic’s political scene today contains parties that are (by Czech standards of 20 years of democracy) rather old (60%) or entirely new (39%) and almost none in between.
  • Given that, pattern, the question is what the Czech Republic looks like in 5 or 10 years.  If the current new parties show the same survival patterns as their “new” predecessors, they will not exist in one or two election cycles (this is the pattern elsewhere in the region).  The old parties may recover some of their voters but probably not all of them and the rest will go on to other new parties which will be equally short lived.  The larger this space gets, the larger space it may create for the next election and the more likely the Czech Republic is to find itself with the same patterns as the Baltics and other countries in the region.  The Czech party system dog has stopped not barking.

Guest Blogger: Sean Hanley on Czech Electoral Politics

Editor’s note:  For the first time on this blog I welcome a guest blogger, and there is no better way to start than with Sean Hanley of University College London, School of Slavonic and East European Studies, Centre for the Study of Central Europe.  It is Sean’s Blog, http://drseansdiary.blogspot.com/that made me realize how useful blogs were for academic communication and gave me the inspiration to take mine much more seriously.  Sean writes:

“I think that we are seeing the rise of new parties – and actually if you track other things including the world of Czech extra-parliamentary parties there is a clear ‘below the radar’ trend that has been underway for several years. I tend to see the challengers as still very much (as they have always been) on the centre-right.  On the left the challenges are not so serious. SPO(Z)  given their current inability to go beyond 3% in the polls, are not, and I think will not become, a serious challenger to CSSD as TOP09 and VV are for ODS and KDU. Crucially, SPO(Z) lack the element of novelty.  Suverenita, if they could get the resources and pick up the political momentum needed for lift off, would be a better bet for real a centre-left populist breakthrough: populist message, well known leader, element of novelty including having a female leader (Veci verejne have also traded on the same formula and have done well despite being led by the bluff Radek John).  On the other hand, this type of formula was tried in 2006 by Zelezny and the Indepedent Democrats and flopped. There is also the possibility of a recurrence of the 1999-2002 surge of KSCM, but I wonder about how likely that would be to repeat since KSCM lacks the broad appeal and tactical flexibility to capitalise on such opportunities.

It would be interesting to compile a dashboard window for support for non-parliamentary parties (including TOP09, which has not made it independently into parliament) to see what trends there are As regards, disillusionment/new party cycles, I would be a little cautious about the viewing this as a very general model to which the CR — after an odd and exceptional period of historically conditioned party system stability — will tend gravitate.  An underlying question is the nature of the disillusionment cycles and demand for new parties and specifically whether there is a certain cyclical/fixed background layer of frustration with established parties held in check by diminishing stock of organizational and political robustness or whether we are seeing eruptions of discontent, which fit with some longer term unfolding pattern linked to social change or the changing nature of reform politics in the region.

To return to the short term the election looks to be shaping up for the predicted battering for ODS and Pyrrhic victory for CSSD.  It will be interesting to see if Klaus blocks a minority CSSD government as promised to avoiding a CSSD-KSCM entente, although if KDU-CSL make it into parliament (and their apparent loss of all but their core support has the positive electoral implication that such voters will tend to turn out, whereas I think TOP09 and VV might poll less than polls suggest) it could leave us again with the familiar deadlocked situation. Grand Coalition anyone?”

Sean makes his electoral predictions on his blog today!  Check it out: http://drseansdiary.blogspot.com/2010/05/czech-elections-more-new-old-politics.html#links

Tomorrow: guest blogging from Tim Haughton of the University of Birmingham

Slovakia Dashboard News: May Polis Poll Closely Follows Trends

Today’s Czech election (and tomorrow’s Eurovision song contest) are the big news (except that Slovakia and the Czech Republic failed to make it into the final yet again) but there’s actually no actual news just yet, so there’s time for a brief comment on reports of today’s Polis poll in Slovakia.  I still wonder about the viability of telephone polls like that of Polis, but to the extent that the firm did fairly well last time in its prediction and tends to follow the same trends as all of the other pollsters (except Median), I am willing to give it a certain amount of credit and build it into my averages.  Polis just released a poll conducted between May 15 and May 20 with a fairly large sample.  Here’s what we see:

Polis’s 34% for Smer is the lowest score the party has received in any major poll in the last two years, but Polis tends to under-poll for Smer, at least compared to the others.  The patterns, however, are extremely consistent:  Polis, MVK and FOCUS all show nearly identical seven point drops for FOCUS since January and all, including Median, show approximately ten point drops since a year ago.  Why this should be is open to question:  the emergence of SaS (I accidentally wrote VV; Freudian slip) and perhaps a few points to the new SDL and perhaps some last minute hesitation by Smer’s many soft supporters who, as election nears, are now forced actually to consider their vote.    It is important to remember that in the 2006 election Smer actually outperformed final polls, but these are never a good indicator (or more precisely they are sometimes a good indicator but it is difficult to predict when).

Polis also tends to underpoll for SNS–by quite a significant margin–but its trends tend to follow, and it is interesting that after a rise in April in both FOCUS and Polis (large in FOCUS, small in Polis) SNS is in both polls back to levels below its March result, suggesting that the Fidesz boost was short lived.  Of course since the Polis and FOCUS polls were taken, we have a new, and perhaps more significant dual-citizenship question which may help SNS, but Smer is also hitting this hard and may pick up much of that reservoir.

Polis also underpolls HZDS–interesting that it underpolls all three coalition parties, and strong reason never to consider its numbers alone, especially for predicting government composotion–but as with the other parties, Polis numbers do follow basic trends and here they show the same slow slide as all of the others except Median (about which I have commented frequently elsewhere).  There’s not much to see here, but Slovakia’s next goverment may depend on whether its shrinking voting base turns out to vote more than others’.

Overall the current coalition dropped two points in this month’s Polis poll, a smaller amount than in FOCUS (4) or Median (3) but this is perhaps to be expected since Polis starts with a lower baseline for the coalition parties (and so they have less to lose).

Even discounting SDKU’s jump in recent Median polls, both FOCUS and Polis show increases for Polis in April-May, though Polis’s is much larger and to levels not normally seen for this party.  I have some doubts about the magnitude, but (like Smer above) SDKU also tends to perform better than polls in the final election (though this time they have SaS to contend with)

Every single poll shows KDH dropping from a March peak to a lower (but still relatively high) level in May.  The question is not the trend but the magnitude.  Polis, MVK (and Median) show the party between 11 and 12.  FOCUS, by contrast, shows it between 8 and 9.  No data that I have tells me which side of this range the party’s scores actually fall.

SaS follows the same trend in every poll: continued rise in May.  Polis’ jump here is slightly smaller than that of FOCUS (which shows a correspondingly smaller jump for SDKU, so the difference may lie in different measures of supporters of those two parties, or it may not).  The party continues to rise and is not facing the sort of “switcher’s remorse” that we saw for SOP in 1998, for ANO, Smer and HZD in 2002 and for SF in 2006.  At least not yet.

Overall the right shows a point and a half gain in this most recent Polis poll, more or less the same that we saw in FOCUS a week ago and slightly smaller than that of Median.  This trend, too, continues.

Amazingly both Hungarian parties cross the threshold in this poll even though the total Hungarian electorate as measured by this poll is only 10.4.  The actual electorate is probably somewhat larger, so this is good news for Hungarians in Slovakia.  It is amazing that these two parties continue in their head-t0-head duel which maximizes both the potential gain–if both do well–and the potential risk–if one drops only slightly below the threshold.  I frankly never expected the parties to maintain this kind of parity, and whether they can keep it up is the thing that most intrigues me about the upcoming election results.