Category: public opinion

Presidential Election, First Round

Although the results were closer than almost anyone expected, the polls were not wildly off: only by a matter of 5-6% points and much less for the major candidates.  The closeness makes the second round interesting, though it’s much harder to see Radicova gaining 12 than Gasparovic gaining 5.

If the same voters were to turn out, it would be easy to make predictions, since we can guess fairly well about the 2nd choice preferences of voters for Martinakova, Bollova and Sidor.  For Miklosko and Melnik it is different (for related reasons):

  • Miklosko’s KDH fought bitterly with HZDS during the period when Gasparovic was the party’s chair of parliament and even though Miklosko will not back Radicova, many of his supporters may be willing to do so.  We shall see here whether past conflicts now rendered largely meaningless (ex-KDH v. ex-HZDS) will trump potentially emerging present conflicts over genuinely ideological questions (without looking more closely at the data, I would guess that Miklosko supporters are more similar to Gasparovic voters (culturally conservative, economically mixed) than to Radicova voters (economically mixed, culturally liberal).
  • Melnik’s HZDS fought bitterly after 2002 against Gasparovic and did so again in this campaign.  Here we shall see whether a current party conflict (Meciar v. Gasparovic) is stronger than a simultaneous ideological conflict (Melnik’s voters would unreservedly prefer somebody like Gasparovic over Radicova if it were not for the party ties.)  In this case I suspect ideology wins out and most Melnik voters go to Gasparovic.

This rough guess would produce something like the following.  Radicova gains more than Gasparovic but not enough to win:

Total 1st Round Share to Gasparovic in 2nd round Share to Radicova in 2nd round Total for Gasparovic Total for Radicova

Of course the same voters will not necessarily turn out and so if Radicova hopes to win, she will need a much bigger turnout in areas of her strength.  This might not actually be as hard as it seems since many of the areas of her greatest strength (particularly in the south) were in areas with rather low turnout, particularly along the country’s southern border.  Hungarians, it would appear from this aggregate data, may have supported Radicova at much higher than average levels but may have turned out in lower than average levels.  If this is true and if Radicova could mobilize Hungarian voter simply to turn out at average levels, she could come very close to Gasparovic’s totals :

Turnout

In the meantime, the daily papers have some reasonably good reportage (see: http://delicious.com/kdecay/pozorblog for articles I’ve tagged) and the Stasticky Urad has some other nice graphs and charts (see http://www.volbysr.sk).  For ease of comparision I’ve included some of the more interesting ones below.

Gasparovic’s regional totals:
Gasparovic Totals

Radicova’s  regional totals:

Radicova

Miklosko’s regional totals:

Regions “won” by Gasparovic versus those “won” by Radicova

Gasparovic v. Radicova

January 2009: FOCUS echos UVVM

January 2009 poll comparisions and averages

I should have known to wait for the FOCUS numbers to come in before writing yesterday’s post.  FOCUS is coming in much more quickly these days.  This month, however, one or the other would have been enough as the results are strikingly similar.  The two polls show the same trends for every major party.
For Smer the two moved in the same direction, almost perfectly.  Gone may be the day when FOCUS was much lower than UVVM on numbers for Smer.  Whether this is the result of a shift in network, methodology or the end of a long string of chance differences, I cannot say.  (Anyone know the answer?).  This month showed a decline from a peak in both.

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

For SDKU the resultsof the two polls have been shockingly close for three months now.  SDKU has stabilized, perhaps helped a bit by Radicova’s presidential campaign (perhaps not).

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

This graph of recent results for SNS shows a fairly wide divergence but similar trending–very slowly downward since December.

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

This graph of recent results for MK shows a slight rise in both polls, though the 2% gap is wide for such a small party.

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

The two polls’ numbers for HZDS are (as they have been) incredibly similar.  HZDS has reached its lowest recorded numbers in both polls (though with FOCUS this represents a tie with its numbers from August 2008).

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

This results for KDH show greater similarity between polls than during much of last year.  Both polls have shown essentially identical results for January and February and this time it is good news for KDH which has risen in both, even after its vice-chair Daniel Lipsic raised a storm by talking about his party’s participation in the “buying” of deputies during the previous parliamentary term.

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

This graph of recent results for KSS shows little change from last month with the two polls placing the party in its traditional zone between 2 and 3%.

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

This graph of recent results for SF shows a small gain in both surveys moving it toward 3%, up notably from its nadir a year ago at less than 1%.  The party’s gains have been steady, so it is difficult to attribute this solely to the increased visibility due to party chair Martinakova’s presidential bid, but this may help.

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

The two other oft-polled parties, remain insignificant.  ANO broke 1% for the first time in 5 months in the FOCUS poll but was nearer to zero in the UVVM poll; after some respectable (for it) showings around 2% in the past months, HZD is back at around 1% in both polls.

Finally, FOCUS has decided to continue with its addition of small parties–KDS, ZS and LIGA.  Each of these parties did better than in the previous month of polling by FOCUS: ZS rose from 0.4% to 1.0%, almost identical to its 1.1% result in UVVM polls.  KDS rose from 0.3% to 0.6% (at the same time that KDH rose several points) in the FOCUS poll.  LIGA rose from 0.3% to 0.4%.  So far SaS is present on the polling list only of MVK and we do not have MVK numbers for January or February.

Since the two polls showed such similar results, averaged all together it is almost indistinguishable from yesterday’s report on UVVM.  Smer leads, SDKU stays stable, SNS and HZDS drop slightly, KDH and SMK rise.

2:||||||||||||||||||||||||

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

This means that the coalition stops its climb and the opposition stops its fall and so we are back almost exactly to mid-2008, though with HZDS a bit weaker and Smer a bit stronger.

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

The only really noteworthy result of analysis by bloc is that the “Slovak National” bloc falls to its lowest recorded levels (around 18%).  I doubt that these voters have lost their national sentiment, but they may have shifted to KSS or Smer (or even KDS?) which are not exclusively nationally oriented.

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

This month’s distribution of parliamentary seats shows that this all makes little difference so far in the big picture.  As always, Smer can form a government with only one partner (and since the numbers putting Smer with a simple majority last month were unlikely to be reached in an actual election, this is the way it has been essentially since the 2006 election.)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

As always, the actual polling numbers are available online at Google Docs:
http://spreadsheets.google.com/pub?key=pdhlCClsiyAMi39bLFpY_Zg

And the most recent three months are below in tabular format (using “iframe” which may not work on all browsers).

<br />

February 2008 UVVM: Coalition down slightly, new counting method for UVVM

UVVM Monthly Report for December 2008

If you set your frame narrowly enough, things can change quickly. In the big picture, nothing much has changed since last month, and the relative positions of parties are pretty much as they have been for the last two years.  Within that narrow frame, however, the current coalition has gone from near-highs for the post election period to near-lows thanks to drops in all three of the coalition parties, the first time all three have dropped simultaneously since late 2006.

Of the three, Smer’s drop was the largest but by far the smallest in percentage terms and the drop occurred after one of the party’s near-high scores, so some correction may be expected.  This month’s 46% is still one of the party’s eight highest scores over the last 30 months.
UVVM+poll+data+ for +all+parties+ for the most recent +24+months+ in Slovakia
UVVM+poll+data+ for +all+parties+except+Smer+ for the most recent +4+months+ in Slovakia

For the other coalition parties, the drop was more significant, with SNS sliding back toward the 11% mark after a high point late in 2008 and HZDS dropping to its lowest mark ever recorded in a major survey, the first time the party has ever received less than 6%.  Of course there is a lot of noise with smaller parties, but HZDS’s tendency to hit new lows every six months or so and its inability for its occasional highs to rise to the levels of previous highs cannot be happy news within the party.  The graph below suggests that the party is not yet in danger of dropping below the 5% threshold in the 2010 election, but the danger is getting closer, particularly since party regularly varies more than 1% from its trendline average meaning that with current trends there is a strong chance that the party will post at least one or two results below the 5% threshold in the months just before elections, a symbolic result that could further hurt the party’s chances.

HZDS trendline

For the opposition parties in parliament, this was a good month, especially for KDH which seemed to be facing a slide as serious as that of HZDS. Again there is a lot of noise and variation here but this gives the party a brief reprieve.  Overall, as the graph below shows, the opposition parties are back near their high point for the post election period.  Unfortunately for them, this is a paltry 33%.

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

And unfortunately for them, Smer is still within striking range of a one-party government.
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

As always, the actual polling numbers are available online at Google Docs:
http://spreadsheets.google.com/pub?key=pdhlCClsiyAMi39bLFpY_Zg  And the most recent three months are at the bottom in tabular format (using “iframe” which may not work on all browsers).

It is worth noting here, however that UVVM has changed its method for asking about voter participation and that this can have effects on the overall results.  As the table below shows, UVVM has changed its question three times in the last two years, switching to a “Do not sympathize with any” response after the 2006 election and then in 2007 changing back to the method it used before which allows a “will not vote” response.  This new option lowered the number of people who specified a party from an average 0f 74% to an average of 68% (though increasing disillusionment with all parties could explain some of the decline).  Now UVVM has split the question into two parts, first asking whether respondents will vote at all and then asking whether for whom they will vote. Asking first about voting has the immediate effect of increasing the share of “will not vote” responses by 8 percentage points.  In the second phase, the survey asks those who will vote to specify a party and here too it allows a “do not know” response.  Between the 8 point increase in “will not vote” and the 3 point increase in the combined “do not know” (respondents who do not know if they will vote and those who do not know for whom they will vote), this method produces an 11 point drop in those who actually specify a party.  Whether this is good or bad is hard to say.  It suggests a lower level of turnout–and I suspect the purpose of the shift is to fix UVVM’s overprediction of turnout in 2006–but whether it is a more accurate measure of results is hard to say.  It is notable that it does not substantially alter the overall party preferences, all of which stay within a few percentage points of their numbers using the old method.  Of course the only way to really test the effect is to ask both sets of questions at the same time, but this would double the expense.  More news on this as I find out any details.

2006 2007-2008 2009
For whom will you vote? For whom will you vote? Will you vote? For those answering “Will vote”: For whom will
you vote?
Does not sympatize with any 26.4          
    Will not vote 17.9 Will not vote 25.0
Does not know who to vote for 14.2    
    Does not know whether to vote 11.4
Will vote 63.6
Specifies party 73.6 Specifies party 67.9   Does not know whom to vote for 6.2
Specifies party 57.4

Finally, I should note that like MVK and FOCUS, UVVM appears to have added the Party of Greens (SZ) to its list of parties on the official roster.  At some point I will rework my software to include them here, but for the moment they remain at a single percentage point (close to the results obtained by FOCUS in January) and so not much is lost by their omission.  It is noteworthy that only one in a thousand respondents (here it can only be one out of the 600 asked) specified a party not on the list, suggesting that there is no spontaneous support for KDS, Liga, or SaS.

<br />

January 2009 Poll Averages: Smer gains, KDH loses, new parties become visible

January 2009 Poll Averages

Overall Monthly Report

Numbers for January are in and while there is not too much to report, there are a few striking findings.  Smer had an excellent month, tying its previous high score in UVVM and achieving a new high in FOCUS (by more than 2 percentage points over its previous high in October 2008).  As the graphs below show, this means that for the first time the poll /average/ gives it a straightforward parliamentary majority of 76 seats.  The big open question is whether a sinking economy will pull these numbers down, but every time it seems that the party has now hit its peak, it sags a bit and goes on to a new peak, so this is truly an open question.

2:||||||||||||||||||||||||

This short-term graph of poll results for averages of major parties other than Smer shows a sharp–two percentage point–drop in KDH not quite compensated for by a rise in SDKU, while within the coalition, SNS’s loss was HZDS’s gain (perhaps literally).  And SMK continues its slow slide to places well below its demographic base.  In general the nationali parties have remained stable while the major opposition parties have all dropped by small margins over the past few months. Among the smaller parties, KSS stayed stable and the 4% result for Slobodne Forum in last month’s FOCUS poll is even more clearly apparent as an outlier.  HZD (not shown here) has been on a very slight upward trend, perhaps related to the Gasparovic presidential campaign, but shows nothing like the bounce that it saw in Gasparovic’s previous campaign.

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

Not pictured here–because they are mere single data points, is new polling data for other small parties, many themselves new. FOCUS has apparently begun to follow the lead of MVK and include these parties on their response sheet, making it easier for voters to choose them.  The psychological effect of this choice is immediately apparent, increasing by a factor of 10 the number of people specifying a party other than those above (albeit from 0.1% to 1.0%).  The Party of Greens (SZ) netted 0.4% from this choice while the KDH-splinter Conservative Democratic Party (KDS) emerged at 0.3% and the liberal LIGA emerged at 0.3% as well.  The 0.3% for KDS may have some bearing on the drop in KDH, but the 0.3% represents only one seventh of the KDH decline.

It is notable that FOCUS numbers for these small parties are much lower than those of MVK which regularly reports SZ figures above 2% and in September reported KDS at 0.8% (but has not reported KDS numbers since).  Interestingly MVK in December put the numbers of yet another new party, the liberal Freedom and Solidarity (SAS) at 2.0%.  FOCUS did not include this party in its list.

Now back to results for bigger parties. Even with weakness in SNS, the Smer increase (and small HZDS increase as well) still pulls the coalition to a new high of 66%: two out of three voters in Slovakia.

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

This long-term graph of poll results for (loosely defined) party “blocs” parties shows the coalition’s increase coming from Smer while SNS and HZDS continue to split their 20%, each month differing only in who gets how much of it. The drop in KDH hurts the opposition significantly.

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

This month’s distribution of parliamentary seats shows the first “poll average” Smer majority.  It’s coalition partners bring an additional 29 seats.  The opposition by this standard would muster 45, which is not that much more than the current coalition without the Smer majority.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

As always, the actual polling numbers are available online at Google Docs:
http://spreadsheets.google.com/pub?key=pdhlCClsiyAMi39bLFpY_Zg

And the most recent three months are below in tabular format (using “iframe” which may not work on all browsers).

<br />

December 2008: Poll Averages and Comparisons

Overall Monthly Report:
Record highs for Smer and disagreements elsewhere

For the first time I can remember we have both FOCUS and UVVM numbers by mid-month (FOCUS getting there first this month) and so we can do a quick analysis already.  If time allows over the holiday season, I may try to do a few more year-on-year comparisons.

This graph averaging the last 24 months of polling results from multiple sources shows a number for Smer that is higher than any previous average.  Although the party is not at is record high for either UVVM or FOCUS, this is the first time it has been at a near-record high for both in the same month.  SDKU is at a correspondingly record low, though since these parties do not share the same voter pool and any correspondence is likely a coincidence.  (Unlike Smer, furthermore, SDKU is not seeing numbers that are unprecedented for the party: what is here a record low would have been for the party a near record high between 2003 and 2006, though then the party was in the tougher position of bearing responsibility for government actions)

2:||||||||||||||||||||||||

These numbers do not give the party a majority of seats but they do bring it within 3 of a majority, again the highest numbers recorded for a poll average.  The opposition, by this calculation, would have a mere 46, less than 2/3 of Smer alone.

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

It is worthwhile to note the degree to which these numbers reflect disagreement among pollsters, however.  In the first place, the “Smer-gap” has re-emerged.  Smer numbers for FOCUS have recently notched upward, but now so have its numbers for UVVM, so we are back at a 6 point difference which is smaller than before only in the sense that it must be measured against a higher baseline for Smer so that it accounts for a smaller overall percentage.
Multiple-poll+average+ for +Smer+ for the most recent +4+months+ in Slovakia

For SDKU, however, the gap has now disappeared and both polls show the party at the exact same level, just above 11%, though not quite in need of the shift in axis that SNS suffered in last month’s poll (from the 10%-20% range to 5-15%).  Still, the news for SDKU is not good.

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

This graph of recent results for SNS shows the return of divergence after long months of near-agreement. FOCUS shows a continuing slide while UVVM shows a sharp rise.  Is politicization of conflicts with Hungary working for SNS?  It depends on your pollster.

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

For MK it does not depend on the pollster–we are back to agreement after a month of disagreement–but MK numbers have been so erratic (given its base) that it is hard to know what to think.

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

As with SNS, it was either a good month or a bad month for HZDS depending on your pollster.  The party’s numbers from FOCUS this month exceed its numbers from UVVM, something that has happened only 4 times in the 30 months since the previous election.  Regardless of relative position, the two are quite close suggesting that the party’s polling level lies around 8% at present, not good for a party that regularly commanded twice that level four years ago, but not in danger of extinction.  Long term trends show the party falling below 5% but not in the next year.

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

This graph of recent results for KDH shows the emergence of a gap that had, for a while, disappeared.  Through most of 2007 the party polled better with FOCUS than with UVVM and it does so again, with a nearly 4 percentage point gap, one of the highest in percentage terms for any party.  Both, however, show a slight rise for the party from November to December.

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

This graph of recent results for KSS shows little change: the party usually polls better with FOCUS than with UVVM and it does so again.

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

If all we had were FOCUS numbers, it would be easy to accuse this post of burying the lead: “SF nears threshold of electability.”  That may be true, especially as dissatisfaction with SDKU appears to grow, but here we find it’s only true of one poll:  The party scores 4.5% with FOCUS but remains mired at 1% with UVVM.  There is a story here but it is hard to know which way it is going.

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

Other parties remain low.  The November bump for HZD has receded in December and ANO is at at near-record lows with both FOCUS and UVVM for the party’s lowest-ever result in poll averages (0.2%).  Its decline in the FOCUS poll may help to explain some of the rise for SF which is its alternative on the non-parliamentary “right.”  The category of “Other” party is also extremely low, suggesting that respondents do not spontaneously offer the Green Party, the Conservative Democrats or Liga-OLS.  Of course this depends on the nature of the questioning: MVK appears to include the Green Party on its list and regularly receives percentages of about 2% for that party.  The same might be true of KDS or Liga if their names were included.

As always, the actual polling numbers are available online at Google Docs:


http://spreadsheets.google.com/pub?key=pdhlCClsiyAMi39bLFpY_Zg

And the most recent three months are below in tabular format (using “iframe” which may not work on all browsers).

<br />

December 2008 UVVM: The High Plateau

UVVM Monthly Report

This overall long-term graph of poll results for UVVM shows Smer back up to a near majority in preferences and with a majority of parliamentary seats while the opposition fades back toward the 30% mark it reached in July and previously in January of 2007.  With the exception of Smer’s rise, however, these numbers are not reflected in this month’s FOCUS poll (which for the first time in my memory beat UVVM to the punch) and which I will discuss in the next post.  This month the FOCUS and UVVM data are at odds in many areas, with one of the highest levels of disagreement between the two (excluding Smer) in the past 18 months, but more on that later.

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

This short-term graph of poll results for UVVM minus Smer shows a reversal in the positions of SDKU and SNS, suggesting that the polarization on Hungary-related issues may be working (FOCUS does not show quite the same trend, but again more on that later).  HZDS continues a minor slide with its second lowest score ever in a UVVM poll (though in FOCUS the party holds its own this month).

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

Adding these up, the strong month for Smer and SNS minus a slight decline in HZDS gives the coalition a near-record high, just shy of 7 out 0f every 10 voters in Slovakia expressing a preference.  The mirror image non-coalition parties drop to 3 in 10.

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

The same basic trends apply to blocs.  The rise for SNS counters the drop for HZDS for stability in the “Slovak National” bloc while the drop in SDKU hurts the “Right.”UVVM+poll+data+ for +party+'blocs'+ for the most recent +24+months+ in Slovakia

And a few percentage points here or there makes a huge difference in the nature Slovakia’s potential future government: were these election results, Smer would need no other partners, a development that could significantly reshape the nature of Slovakia’s politics.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

As always, the actual polling numbers are available online at Google Docs:


http://spreadsheets.google.com/pub?key=pdhlCClsiyAMi39bLFpY_Zg

And the most recent three months are below in tabular format (using “iframe” which may not work on all browsers).

<br />

Poll Comparisons: August-November 2008

August-November 2008 Poll Comparisions

Trends and comparisons monthly report

Although the overall figures for the last four months show little change, there have been some curious shifts in the polling numbers of various polling firms.  A graph of recent polling results for Smer from multiple sources shows in particular a sharp increase among Smer supporters on FOCUS polls, especially in October, when FOCUS and UVVM produced the same result.  (For context, the last time that Smer numbers were higher in FOCUS polls than in UVVM polls was April 2006, and since the June 2006 elections Smer’s numbers from UVVM have exceed its numbers from FOCUS by an average of 6 percentage points.  Smer’s preferences were not higher in MVK polls, however, and the party’s record highs in FOCUS polls were counteracted by record lows in MVK, especially in November.  (A side note: because UVVM and FOCUS produced identical results for Smer in October, the overlapping of their + and x symbols misleadingly produces a near-circle that resembles the indicator used here for MVK).

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

This graph of recent polling results for SDKU shows a reorientation of polling numbers but a stable aggregate.  Numbers for UVVM dropped sharply between October and November but rose sharply for MVK.  They stayed the same for FOCUS and the overall result is an average identical to the previous month.

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

This graph of recent results for SNS for the entire season shows tight clustering around the mean, with no more than 1 percentage point difference between any party and the mean.  That mean appears to have settled between 10-11% in the last three months and it is intriguing that this has happened despite an escallation in tensions between Slovakia and Hungary and heightened rhetoric and visibility for Slota.  December’s UVVM numbers may help figure out if there is any delayed effect but if not, it bodes ill for SNS which should be seeing some boost from the conflict.  A side note, I found it necessary this month to shift the range of the graph for SNS from a 10%-20% scale (like SDKU) to a 5%-15% scale (like SMK, HZDS and KDH).

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

This graph of recent results for MK again show unexpected volatility for what should be a stable population.  FOCUS numbers (hidden under the MVK circle) rebounded from near record lows in the October poll to something closer to the recent FOCUS averages for MK.  The inclusion in November of numbers from MVK (which tends to produce higher numbers for MK than other firms) brought the party’s average back up above 9% but I now have little doubt that MK’s numbers have begun to trail the party’s demographic base in a way they have not done in the past.  Whether this means they will poll lower in elections is an open question.  It will also be interesting to see if December numbers show the conflict between Slovakia and Hungary producing higher MK numbers.

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

This graph of recent results for HZDS shows more stability than for other parties.  The relative positions of pollsters remains unchanged and the party’s average has not changed by 0.5% percentage points from month to month.

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

This graph of recent results for KDH shows a bit more volatility than has been the recent norm, with a sharp drop by UVVM countered by a sharp rise by FOCUS.  MVK hugs the mean for this party, as it has often done.  The KDH-splinter party KDS registered for the first time above random noise on MVK’s September survey (though at a mere 0.8%), but the (as usual) haphazard reporting of results by major Slovak dailies does give any information about that party’s performance in November.

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

This graph of recent results for KSS shows stability.  Good MVK numbers bring the party up near 3% for the first time in half a year but little else seems to have changed.

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

This graph of recent results for SF shows bigger changes.  After October’s consensus at the 2% mark, FOCUS now shows the party at 3% while the UVVM numbers are back at 1%.  This is always difficult to gauge since the party’s total support is below the margin of error, even assuming no sampling problems.

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

Finally, while I did not prepare the graphs, HZD is at a post election high of 1.9, perhaps because of former party chair Gasparovic’s presidential election bid (the party saw a significant but fleeting gain during Gasparovic’s last effort) while ANO hovers on the threshold of non-existence with another 0.5%.

November 2008: Monthly Poll Averages

November 2008 Poll Averages

Overall Monthly Report: Continuity continued

How many ways to write the same story.  While there are some interesting variations among polls (to be covered in the next post) the overall story of Slovakia’s party support remains astoundingly consistent.  This month takes the edge off of some extremes in last month, most notably last month’s peak performance by Smer (the result of a major rise in its FOCUS numbers which this month have receded somewhat, though they remain higher than before) and valley numbers for SMK (also thanks to unusual FOCUS numbers).  This simply puts every party back at their overall recent averages:

2:||||||||||||||||||||||||

This long-term graph of poll results for coalition and non-coalition parties shows a slight drop in the coalitions performance because (unlike in October) there was no Smer increase to balance weakness in SNS and HZDS:

Multiple-poll+average+ for +coalition+support for the most recent +24+months+ in Slovakia
The long-term graph of poll results for (loosely defined) party “blocs” parties shows very little change over time.

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

And as always (of late), the month’s distribution of parliamentary seats shows a near majority by Smer.  If the election were held today, it would have its choice of a single partner.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

As always, the actual polling numbers are available online at Google Docs:
http://spreadsheets.google.com/pub?key=pdhlCClsiyAMi39bLFpY_Zg

And the most recent three months are below in tabular format (using “iframe” which may not work on all browsers).

<br />

Poll Analysis: UVVM Volatility

Also available in Slovak, here

I have occasionally expressed concern about the recent numbers coming out of UVVM, particularly those for the Party of the Hungarian Coalition.  Something is wrong there. In the spirit of my heroes at http://www.fivethirtyeight.com, I decided to take a look at the data.  The chart below calculates the total month-to-month differences in party results (up, or down) and then measures these as a percentage of a party’s average support during that period (i.e. a party with an average support of 20 percentage points and average monthly volatility of 1 percentage points would be listed at 5%).  I have done the calculations for the overall average (all of the polls within a given month) and for UVVM and for the two electoral periods for which we have good data: 2002-2006 and 2006-present.

Party 2002-2006 2006-2008
  Overall UVVM Difference Overall UVVM Difference
Smer 7.0% 9.2% 2.2% 8.9% 7.8% -1.1%
SDKU 12.5% 16.4% 3.8% 13.2% 16.2% 3.1%
SNS 14.7% 18% 3.3% 12.1% 12.1% 0.0%
MK 8.0% 12.0% 4.0% 10.1% 16.1% 6.0%
HZDS 9.9% 12.4% 2.5% 12.2% 14.1% 2.0%
KDH 10.4% 16.1% 5.8% 10.2% 10.9% 0.7%
Average 10.4% 14% 3.6% 11.1% 12.9% 1.8%

What is not surprising here is that UVVM alone has much higher levels of volatility than the overall average which includes multiple polls that smooth out the monthly variability.  What is surprising here is the fact that the the difference in volatility of SMK between UVVM and the overall average grew (from 4% to 6%) even as the differences between UVVM and other surveys declined.  Between 2002 and 2006 the volatility of SMK was second to the bottom (after Smer) in both UVVM and overall averages.  Between 2006 and 2008, the SMK volatility stayed the second lowest in the overall averages but rose to highest in UVVM by a wide margin.

This is all particularly surprising since the overall volatility of the Hungarian Coalition’s electorate have remained remarkably stable over time, as the following table demonstrates:

Party (elections in sample) Volatility as a share of average party support Raw volatility
ANO (2002,2006) 103.2% 4.9%
SNS (1994,1998,2002,2006) 80.5% 5.9%
Smer (2002,2006) 73.6% 15.7%
KSS (1994,1998,2002,2006) 51.2% 2%
SDKU (2002,2006) 45.1% 7.5%
HZDS (1994,1998,2002,2006) 38.7% 8.7%
MK (1994,1998,2002,2006) 11.4% 1.2%
KDH (1994,2002,2006) 10.7% 0.9%

There is an easy explanation for this:  UVVM has had difficulty maintaining its Hungarian sample.  This is understandable–this is a difficult task–but it is important to keep this potential problem in mind rather than to assume that all polls actually reflect representative samples.  This should be a question of analysis rather than assumption.  It also raises questions about the representativeness of this sample (and that of other pollsters) for other parties.

I am deeply grateful to UVVM for all the support they have provided to me over time. This is a problem that needs attention but I am hopeful that the institute can return to its previous levels of excellence.