February 2008 Poll Average: Trends and Analyses

Overall Monthly Report

FOCUS numbers for February arrived just moments ago and thanks to new automated methods (and an hour of tweaks when the automatic methods proved not quite error-free), here are the most recent monthly averages of UVVM and FOCUS (and any other major pollster that announces its result during the period.

Since the FOCUS numbers do not significantly alter the UVVM trends for the same period, I will not go into overwhelming detail: Smer wins. SDKU and SNS form a distant second tier. SMK,KDH and HZDS remain stable in a low-but-electable third tier, and other parties flatline near zero.

Multiple-poll+average+ for +all+parties+ for the most recent +24+months+ in Slovakia

This graph averaging the last 4 months of polling results from multiple sources shows a slo, consistent rise for Smer and relative stability for almost every other party except for slight declines in SDKU and KSS.

Multiple-poll+average+ for +all+parties+ for the most recent +4+months+ in Slovakia

Multiple-poll+average+ for +coalition+support for the most recent +24+months+ in Slovakia

This long-term and short-term graphs of poll results for coalition and non-coalition parties shows the effect of Smer’s increase on the overall coalition-opposition relationship. The gap is as large as it has ever been 63:37, a 26 point gap.

Multiple-poll+average+ for +coalition+support for the most recent +24+months+ in Slovakia

Multiple-poll+average+ for +coalition+support for the most recent +4+months+ in Slovakia

This long-term and short-term graphs of poll results for (loosely defined) party “blocs” parties shows gains on the “Left” for Smer countered by losses for KSS (this grouping as “Left” probably has less meaning than it has in the past, as Smer has moved in other directions), losses on the “Right” for SDKU enhanced by slight losses for KDH (and we shall see in mid-March whether the departure of Palko and Miklosko hurts KDH significantly), slight gains among the Slovak Nationalists and among the Hungarian Nationals which are so slight as to be attributable to statistical noise).

Multiple-poll+average+ for +party+blocs+ for the most recent +4+months+ in Slovakia

Multiple-poll+average+ for +party+blocs+ for the most recent +24+months+ in Slovakia

Multiple-poll+average+ for +party+blocs+ for the most recent +4+months+ in Slovakia

Viewed simply as a question of parliamentary seats, this means, as in every recent month, that if Smer’s success in the polls translates into ballot box success, it can have its pick of coalition partners in a 2-party government.
Multiple-poll+average+ for +estimated+party+seat+distribution for the most recent +1+month+ in Slovakia

Multiple-poll+average+ for +estimated+party+seat+distribution for the most recent +24+months+ in Slovakia

The Trends and Comparisons Report for November-February should follow soon.

Popularity of Departing KDH Deputies

On the day after SDKU members began talking about publicly changing leadership, KDH members began publicly leaving the party after apparently failing to produce a leadership change. I will leave this to others to discuss in detail, but do here what I am prone to do: offer rather facile quantitative evidence of qualitative developments. In this case, I offer a numerical view of the popularity of the departing KDH members. Of course ballot position plays a role as well (as voters tend to tick off candidates in order going down)., and so any meaningful assessment requires a comparison to the baseline effects of ballot position. Full data is below but for clarity, I present the following two graphs which in different ways show the same thing: that the departing deputies were neither significantly more nor significantly less popular than those who stayed.

KDH Deputies’ Preference Vote, 2006
Source: http://www.statistics.sk/nrsr_2006/slov/index.jsp?subP=v

With the exception of Daniel Lipsic, that the KDH deputies who have not joined the departing four did tend to underperform (in terms of popularity) the deputies of other parties at similar ballot positions, particularly Hrusovsky. This is also true, however of two of the four departing candidates, particuarly Palko. Among the departing 4, only Miklosko demonstrated a popularity that was out of proportion with his position on the ballot, though this may have been the result of his under-placement there. In popularity terms, the outgoing four thus include 2 of the top 4, but their aggregate popularity would appear to be less than that of the two popular figures who remain (even as Hrusovsky underperformed and Lipsic overperformed).

KDH Deputies’ Preference Vote Against Expected Vote Based on Ballot Position
Source: http://www.statistics.sk/nrsr_2006/slov/index.jsp?subP=v

It is hard for me not to note here that the graphic presented by SME, while much more attractive, does not convey the same information, not only not placing the numbers in context (or even giving percentages) but not even allowing rapid comparison of where the departing deputies stood relative to those remaining (for more on this kind of visual analysis of statistical information, see http://junkcharts.typepad.com/and http://infosthetics.com/)


KDH graphic from SME

Meno a priezvisko kandidáta [Name] Počet platných prednostných hlasov [Number of preference votes] Poradie po zohľadnení prednostného hlasovania [ Order determined by preference voting] Poradie na kandidátnej listine [Position on original candidate list]
Pavol Hrušovský 95412 1 1
Daniel Lipšic 86536 2 3
Vladimír Palko 64026 3 2
František Mikloško 35841 4 8
Július Brocka 24932 5 4
Martin Fronc 10036 6 5
Rudolf Bauer 9867 7 6
Mária Sabolová 8612 8 19
Peter Gabura 8527 9 77
Monika Gibalová 6485 10 33
Stanislav Kahanec 5889 11 22
Pavol Minárik 3283 12 7

Source: http://www.statistics.sk/nrsr_2006/slov/index.jsp?subP=v

February 2008 UVVM: Smer near record high

It is hard to tell the same story in an interesting way every month (and maybe this is why Slovak papers, which unlike me have competing demands and limited space, do not tell the story at all). First the numbers:

UVVM+poll+data+ for +all+parties+ for the most recent +24+months+ in Slovakia

Politický subjekt: december 2007 január 2008 február 2008
SMER- sociálna demokracia (SMER) 45,2 43,9 45,4
Slovenská demokratická a kresťanská únia – DS (SDKÚ-DS) 13,5 14,7 12,5
Slovenská národná strana (SNS) 11,2 14,0 12,4
Strana maďarskej koalície-Magyar Koalíció Pártja (SMK-MKP) 7,6 8,0 9,8
Ľudová strana – Hnutie za demokratické Slovensko (ĽS-HZDS) 9,3 7,2 9,4
Kresťanskodemokratické hnutie (KDH) 8,4 8,2 7,4
Komunistická strana Slovenska (KSS) 2,0 2,0 0,8
Hnutie za demokraciu (HZD) 1,0 0,8 0,8
Slobodné fórum (SF) 1,1 0,3 0,7
Aliancia nového občana (ANO) 0,0 0,5 0,5
Iná politická strana, hnutie, koalícia 0,7 0,4 0,3

This overall long-term graph of poll results for UVVM shows things more or less back where they were two months ago, with a slight uptick for MK (perhaps simply because the poll may have included more Hungarians) and slight downtick for SDKU. HZDS and SNS in particular are at levels very close to late 2007. The differences are more apparent in the short-term graph without Smer.

UVVM+poll+data+ for +all+parties+except+Smer+ for the most recent +4+months+ in Slovakia

This short-term graph of poll results for UVVM suggests that the major overall change in the last four months in UVVM polling is the slide for SDKU (whether this has to do with their attempt to change the Media Law by withholding support for the Lisbon Treaty is obviously unclear) and the increase for MK back toward the historical norm (but this is so dependent on polling methodology that it is hard to distinguish this from noise). KSS falls below 1% for the first time since I started systematically to collect these numbers, (at least since 2002).

Overall, these changes have only a slight overall effect on overall support levels, so there is not much new to report there, but the current coalition does widen the gap despite the drop in SNS (more than compensated by rises for HZDS and Smer).

UVVM+poll+data+ for +coalition+support for the most recent +24+months+ in Slovakia

And the short-term graph:

UVVM+poll+data+ for +coalition+support for the most recent +4+months+ in Slovakia

In terms of bloc vote, the “right” (SDKU, KDH, ANO and SF) loses the slight lead it had acquired over the Slovak national bloc (HZDS, HZD, SNS) in the past 2 months, but there is no real change otherwise.

UVVM+poll+data+ for +party+'blocs'+ for the most recent +24+months+ in Slovakia

UVVM+poll+data+ for +party+'blocs'+ for the most recent +4+months+ in Slovakia

And none of this has any real effect on the potential distribution of parliamentary seats. If these numbers were to hold, Smer could form a coalition with any other single party. And it inches here toward at least the theoretical possibility of a one-party government of the sort not seen since Meciar got 74 seats in 1992. Of course there are reasons for thinking that such a result is highly unlikely (soft support, a likely leveling off of the current economic growth), but it is no less impressive, given the other things that have changed since 1992 (fewer small, sub-threshold parties, more institutionalized opposition). However it is calculated and however soft the support is, Smer is doing something that very few parties in similar circumstances have manged.

UVVM+poll+data+ for +estimated+party+seat+distribution for the most recent +1+month+ in Slovakia

UVVM+poll+data+ for +estimated+party+seat+distribution for the most recent +24+months+ in Slovakia

September-December 2007 Poll Comparison

Trends and Comparisons Monthly Report

Although I still do not have FOCUS data for January, it may be useful to post these individual-party graphs for the last 4 months of 2007. I will repost this as soon as I get the new data. In the charts below, Xs represent the firm UVVM, plusses represent FOCUS and diamonds represent MVK. The thick colored line represents an average of all three (or only UVVM and FOCUS of there is no monthly poll from MVK).

Multiple-poll+average+ for +Smer+ for the most recent +4+months+ in Slovakia

This graph of recent polling results for Smer from multiple sources shows the large difference among polls that has proven the norm in the last two years. UVVM’s estimates for Smer exceed FOCUS’s by an average of about 7 percentage points on a baseline of 35. Why this is so is a mystery, even to some of the pollsters involved, though I hope to find out more. MVK’s numbers (marked by diamonds with no connecting lines since they are not published at monthly intervals) stand conveniently in the middle, almost precisely at the average of the other two. Whether this is an accident or the reflection of a more broadly based sample is not an easy question to answer. MVK’s final poll results before the 2006 election were no more closer to the actual election numbers than UVVM’s (and by some calculations, actually slightly slightly farther away).

UVVM and MVK show an almost identical gain (about 5 points) for Smer between October and December. FOCUS shows no such overall rise, but does agree on rising numbers between Nov. and Dec.

Multiple-poll+average+ for +SDKU+ for the most recent +4+months+ in Slovakia

This graph of recent polling results for SDKU, to the same scale, shows more or less the inverse pattern for Smer, at least in terms of the relative support for the party: FOCUS numbers are higher than UVVM by an average of about 3-4 percentage points on a baseline of 15. MVK does not stand in themiddle here but in Oct. stands with FOCUS and in December with UVVM.

Multiple-poll+average+ for +SNS+ for the most recent +4+months+ in Slovakia

This graph of recent results for SNS shows a slightly narrower range of disagreement among polls–only about 2-3 percentage points on a baseline of 13. The differences are consistent (FOCUS shows higher overall preferences for the party than UVVM) and in this case so are the trends: every poll shows a general drop of about 1-2 percentage points. UVVM shows this drop in September; FOCUS shows it in December.

Multiple-poll+average+ for +MK+ for the most recent +4+months+ in Slovakia

This graph of recent results for MK shows occasional fairly wide disagreement among polling firms. FOCUS consistently shows higher numbers than UVVM, but sometimes the differences approaches zero while other times it approaches 4 points on a baseline of 9, quite a big gap. Here MVK stands in the middle in October and on the high side in December. If anything suggests a difference in the network of poll-takers it is this graph. Since the elasticity of voting for Hungarian parties is relatively low (at least lower than for other parties), differences here are more likely to suggest differences in interviewing patterns rather than changes in public opinion. UVVM’s numbers look a bit low here, but that is simply a guess based on census numbers an irrespective of other factors that may be at play here.

Multiple-poll+average+ for +HZDS+ for the most recent +4+months+ in Slovakia

This graph of recent results for HZDS shows a less clear difference among polling firms than the results for other parties. Here a big and consistent difference in September and October (UVVM on the high side, FOCUS on the low side) reverses in Nov. and re-emerges much smaller in December. Here it is MVK that is consistently lower than the other two. All firms show the same trend–a drop of approximately 2 percentage points on a baseline of 9 percent. This consistency in the trending–rare among these graphs–bodes ill for HZDS. At this rate the party would fall below the threshold of electability by the middle of this year. Even more restrained trendlines show the party hovering consistently between 4 and 7 percent around the time of the next scheduled elections in June 2010.

Multiple-poll+average+ for +KDH+ for the most recent +4+months+ in Slovakia

This graph of recent results for KDH shows strikingly little difference among polling firms and little difference over time.

Multiple-poll+average+ for +KSS+ for the most recent +4+months+ in Slovakia

This graph of recent results for KSS shows a difference among firms of about 2 percentage points. This is small in percentage terms but large in relative terms and has major significance for the party’s future as they reflect the difference between “in striking range” (FOCUS) and “no chance” (UVVM). MVK here splits the difference. In June 2006 UVVM estimates of KSS were about 0.5 percentage points higher than those of MVK suggesting perhaps some primacy for MVK’s results here. These polls show a slight–but very slight–positive trend.

Multiple-poll+average+ for +SF+ for the most recent +4+months+ in Slovakia

This graph of recent results for SF shows convergence around 1 percentage point, suggesting the party’s effective demise as an electoral party. How long it takes SF to respond to rumors of its own death are unclear but parties in Slovakia (as elsewhere) do tend to linger on the table. They are not alone in this. I do not include here charts for ANO and HZD because these have flatlined at around one as well.

One interesting side note in this regard, however. In contrast to existing parties like SF, ANO and HZD that receive almost no preferences, MVK surveys are consistently showing preferences for the Green Party–which to my understanding does not exist in a formal sense–at around 2%. I hope to find out more about this in coming meetings.

September-December 2007 Poll Average: Trends and Analyses

I offer the charts below to fill in gaps in previous reporting and to test the current blog setup. I hope to acquire data from FOCUS for January 2008 that will allow me to update this today. And UVVM data for February should already be available by next week.

As a side note, all of the data for this and previous posts is available in an on-line google spreadsheet posted at: http://spreadsheets.google.com/pub?key=pdhlCClsiyAPaeKoxMfAaXQ

Overall Monthly Report

Multiple-poll+average+ for +all+parties+ for the most recent +24+months+ in Slovakia

This graph averaging the last 24 months of polling results from multiple sources shows the current broad stability. Relative positions of parties have barely changed since the elections of 21 months ago.

Multiple-poll+average+ for +all+parties+ for the most recent +4+months+ in Slovakia

This graph averaging the last 4 months of polling results from multiple sources also shows little change other than a slight rise for Smer and a slight decline for HZDS, likely the result of the SPF controversy.

Multiple-poll+average+ for +coalition+support for the most recent +24+months+ in Slovakia

This long-term graph of poll results for coalition and non-coalition parties shows essentially no change.

Multiple-poll+average+ for +coalition+support for the most recent +4+months+ in Slovakia

This short-term graph of poll results for coalition and non-coalition parties shows that changes in party preferences essentially occurred within coalition boundaries.

Multiple-poll+average+ for +estimated+party+seat+distribution for the most recent +1+month+ in Slovakia

This month’s distribution of parliamentary seats shows no fundamental difference in coalition math. If these estimates are accurate reflections of voting (here there is considerable reason for caution) elections today would allow Smer to form a 2-party coalition with any party in parliament, and the only coalition that could be formed without Smer is a coalition that included all other parliamentary parties (not likely since it would need to include both the Slovak National Party and the Hungarian Coalition).

Multiple-poll+average+ for +estimated+party+seat+distribution for the most recent +24+months+ in Slovakia

As might be expected from the opinion data on which it is based, this long-term graph of distribution of parliamentary seats shows that the current expected distribution of parliamentary seats has changed little for since the 2006 election.

Still Tired From The Move

I used to wonder at blogs that dwelt on the minutiae of servers and hosts. No more. If you are reading this, then this blog (like Slovakia) has made a successful triple transition: a new platform (WordPress is proving worthy of the switch), a new host (many thanks to Joe Oravec at Wayne State for his remarkably able and generous help), and a new domain registrar (absolutely no thanks to Networksolutions for one of the worst customer service experiences I could imagine. It is one thing not to respond to queries. It is another, and far worse, to respond–as they did twice–with automated messages which suggested–incorrectly, of course–that the problem could be solved by purchasing additional Networksolutions products.

To test this new, I will post the poll average data from the last 1/3 of 2007 in a separate post. Average data from the current month may be available as soon as today.

Of course there may still be some glitches and so I would appreciate any comments here about anything here that does not work or just looks bad.

Slovakia Party Tree 2008

Once upon a time, Slovakia’s party politics was extremely difficult to follow, even for the trained observer. The comings and goings of parties and party members in and out of various groupings in various guises resembled a scene from a Marx Brothers movie. In order to keep track of these shifts and to explain them to others, I adapted and subsequently modified the work of a variety of other scholars (and the inspiration, if not the execution, of Edward Tufte) to create a “family tree” chart of Slovakia’s parties that would allow a relatively quick visual assessment of party changes over time. Since then the changes have slowed, but thanks to the intervention of Bela Kersagh, the chart has become extremely easy to update and so I present it here in its newly colored glory along with a table with more or less the same information. On reviewing the chart I realize that I have not always followed the same rules for inclusion and exclusion (in the shifts among small parties, for example) and so I ask forgiveness for the minor imprecision). It is my goal over the next several years to develop a standard scheme for displaying such a chart and working with other academics in the region to develop a standardized set that might find their way to public sites such as this one or Wikipedia.

The chart and table are below. They are also available as in pdf format: http://www.la.wayne.edu/polisci/kdk/pozorblog/Slovakia Party Tree and Table 2008.pdf

Slovakia Party Table 2008

Slovakia Party Table 2008

Postscript: I have recently discovered that this kind of diagram is really nothing new, nothing but a modified Sankey Diagram, commonly used in the energy industry and materials industry to track flows in a closed system. This works because parliamentary seats represent a similarly closed system, though, unlike the energy diagrams, we cannot say exactly where the seats “go,” since voting is secret. Hence our Sankey has abrupt and opaque reorganizations of overall allocation during every election but otherwise closely resembles the overall Sankey approach.

Surprise: Meet the New Boss

This week’s discovery presents the perfect opportunity to express my ongoing gratitude to my dissertation adviser, A. James McAdams.

Among the many other ways in which he has shaped my work, Jim sent me off to my first year of fieldwork in 1993 with one of the most productive tools in my research I have ever encountered: a question. “When you get home at night,” he suggested, “write just a few sentences on the question, “What surprised me today.”

In other words, what did I encounter that I did not expect? What looked different than I would have guessed? What did not fit the model? I have not done this as regularly as I would have liked, but it is a question that has consistently called me to look for the holes in my models, the limits of my understanding, the places where I see what I want to see and disregard the rest. That’s where the interesting stories emerge. And if I cannot think of something that answers the question, then I know I am doing something wrong, because I am blinded by my almost assuredly limited, if not utterly wrong, presuppositions.

I was delighted this week, therefore, to encounter something that made my job all-too-easy. While walking through the Slovak town of Samorin I passed an attractive park with a stone walk and a series of stone plinths, on each of which was a bronze bas relief of a human face.

Memory Park, Samorin, Slovakia

This is the kind of thing I am always curious about. What I expected was plaques of obscure (to me) local figures from the communist era (national-level figures such as Husak or Jakes would probably not have survived past 1990), or perhaps, given the ethnic composition of the region, even figures from 19th century Hungarian history (with signs that these replaced had earlier plaques of Husak and Jakes). What I did not expect were these figures:

Image Park, Samorin, SlovakiaImage Park, Samorin, Slovakia
Image Park, Samorin, SlovakiaImage Park, Samorin, Slovakia

Freddie Mercury, John Lennon, Jim Morrison and Curt Cobain. Not pictured here–because they are unfortunately not yet on the park’s website–are additional plinths dedicated to Jimi Hendrix and Bon Scott. The website explains that the plinths and plaques were erected by a civic group called “Immortal” founded in 1992 to “provide financial aid to anti AIDS and drug publications, moreover we help financially and provide concert opportunities to the beginner local and regional bands.” The Slovak Spectator has a bit more information here.

So having found my surprise, I must wonder what to make of it. As a student of politics, I am paradoxically delighted to see an emphasis on the non-political. While more than a few fights have erupted over musical taste (see below), the choice of rock stars here does not have same potentially divisive character as a set of political representations, and so it is good to see. (It is reminiscent of a comment by a Czech friend who pointed out how much simpler life would had been if local mayors had insisted on streets named after trees and flowers instead of political figures. Between 1918 and 1992 the name of street of my institute in Plzen changed from Franz Josef, to Woodrow Wilson, to Hermann Goering, to Victory, to Stalin, to Moscow, to Svoboda, back to Moscow, and finally to America).

And yet, politics remains an issue, particularly when it involves rival ethnic groups. The same group that built the music park–Immortal–also lists among its activities the creation of a statue of (Hungarian) King Istvan. There is no lack of ethnic symbolism in such a choice and no absence of politics. The issue of ethnicity does not go away, even though members of ethnic communities may sometimes actively focus on uniting rather than dividing. Through the work of Immortal in Samorin, “Imagine” meets “We Are The Champions” in more ways than one.

Samorin, Slovakia

Finally, as a postscript, there is the question of music, about which I have no expertise but lots of opinions. The recent elevation of Bon Scott of AC/DC raises questions about the current direction of the foundation. Its members does not appear to have rejected candidates on the basis of lifestyle or cause of death. The only common denominators appear to be “rock star” and “died young” and so it is rather a shock that Scott precedes Bob Marley and Keith Moon and a variety of others. Unfortunately the current trend points instead toward Jeff Pocaro of Toto and Steve Clark of Def Leppard. Fortunately this may lead at long last to memorials for the deceased drummers of Spinal Tap (if they can find enough room in the park).