Something in this WSJ story about Russian meddling in Bulgaria caught my eye – the bits about “exaggerated polling data” and this rather curiously-named polling firm:
“A Bulgarian polling company, Gallup International, which isn’t related to U.S. pollster Gallup Inc., accurately predicted Mr. Radev’s victory. The company, which is being sued by Gallup Inc. for using its name without authorization, also co-published a February poll that said citizens of four NATO members, including Bulgaria, would choose Russia, rather than NATO, to defend them if they were attacked. Those results were at odds with a similar poll by Gallup Inc., published a few days earlier, showing that most NATO members in Eastern Europe, including Bulgaria, see the alliance as protection.
Gallup International didn’t respond to requests for comment.
“This wrapped-in-secrecy poll had no details on methodology nor funding sources,” said Ilian Vassilev, Bulgaria’s former ambassador to Moscow. “Russian media strategists and their Bulgarian proxies used the Western name to fool people about its credibility and spread their message.””
So I decided to take a quick look at Gallup International’s website, here. Without digging into the actual polls they do themselves, not much to comment on here.
Things are a bit more interesting on the affiliations page.
First on their list of affiliations is the WIN/Gallup International, “made up of the 80 largest independent market research and polling firms in their respective countries.” It’s an affiliation I’ve never come across and has several members I’ve either never heard of or are minor players in their respective countries (not including Leger in Canada, who’s well established in Canada and, coincidentally, also part of ESOMAR), which I guess underscores the “independent” part of the description.
But it’s the ESOMAR bit which interests me most. ESOMAR is a large, well-respected international network of market research firms who abide by a code and guidelines on market research. Not everyone’s a member of ESOMAR (in fact, none of the big firms I worked for were members) but it’s a reassuring thing to see on a polling firm’s site if you know nothing about them.
But this is what you get when you do a search for Gallup International in the ESOMAR directory.
I double-checked the list of Bulgarian members to make sure I didn’t miss them. There are eight Bulgarian members of ESOMAR, according to the directory. Gallup International isn’t one of them.
(the title of this post explains it. there’s hockey on so I can’t be arsed with a preamble)
I did a bit of analysis using the most recent publicly available Ukraine data set I have – the May 2016 edition of the KIIS Omnibus survey of 2,000 Ukrainians in government-controlled parts of the country (as always, no Crimea, no unrecognized “DNR”/”LNR” statelets). The September 2016 data set will be available in February, they tell me.
I grouped together supporters of the three far-right parties that showed up in the May survey (Svoboda, Pravyi Sektor and Yarosh’s Governmental Initiative) and compared them to Radical Party supporters to see how similar they are.
The answer? Not very.
Age: Radical Party supporters tend to be older (average age 52.2 years old) than far-right party supporters (45.8 years old) [p≤.01].
Settlement size: Radical Party supporters are more likely to live in communities of less than 100,000 people (16.9%) than in cities with more than 100,000 people (10.3%) [p≤.01] – but there’s no significant difference for far-right party supporters.
Urban/rural: Obviously related to settlement size, Radical Party supporters were more likely to be from a rural area (18.6% compared to 11.1% urban; p≤.01). Again, there’s no significant difference for far-right party supporters, even if the numbers appear to slightly skew rural.
Attitudes towards Russia: Shocking no one, far-right party supporters are more likely to have bad/very bad views towards Russia (16.9% compared to 3.7% ‘good/very good’)[p≤.01]. Not so for Radical Party supporters – there’s no significant difference.
Will Ukraine be better/worse?: Far-right party supporters are more likely to think that the situation in Ukraine will be better (14.6%) or the same (15.7%) in a year’s time, compared to 6.9% ‘worse’. [p≤.01]. There’s no significant difference for Radical Party supporters.
Ukraine’s leaders moving country in right/wrong direction?: While Radical Party supporters are more likely to think the country’s going in the wrong direction (15.8% compared to 6.5% ‘right direction’; [p≤.01]), there’s no significant difference for far-right party supporters.
Perceived income: I’ve split the perceived income question in two (there’s five categories, with almost no one picking the fifth, ‘richest’ category) – think of it like ‘perceived lower income’ versus ‘perceived higher income’.
With that in mind, Radical Party supporters were more likely to describe themselves as part of that lower-income group; their household situations tend to be “lacking money for food” or “enough money for food but not for clothes” compared to having enough money for clothes or to buy expensive things (16.6% compared to 7.9%)[p≤.01]. As for far-right party supporters, there were no significant differences here.
Reported income: Radical Party supporters were more likely to report they earned less than 3000 UAH a month (16.6% compared to 10.1% more than 3,000 UAH/month)[p≤.05], while there was no significant difference for far-right party supporters. This question, FWIW, doesn’t appear on all KIIS surveys.
Region: Far-right support tends to come from western Ukraine (21.7%), compared to 7.0% in central Ukraine, 5.3% in southern Ukraine and 3.1% in eastern Ukraine [p≤.01]. Radical Party support, on the other hand, is actually pretty even across western, central and southern Ukraine (14%-16%).
Education: Like with perceived income I’ve had to split education into two broad categories – ‘lower-educated’ and ‘higher-educated’. Using those categories Radical Party supporters tend to be lower-educated (19.3% compared to 11.8%)[p≤.01], while no significant difference appears for far-right party supporters.
Gender: Far-right party supporters are more likely to be men (14.7% compared to 7.2%) – no such significant difference appears for Radical Party supporters.
Based on this data, far-right party supporters and Radical Party supporters don’t look too much alike.
Far-right party supporters, relatively speaking, are young, predominantly male and concentrated in western Ukraine and have much more negative attitudes towards Russia.
Are they more concentrated in rural areas? They may well be, but the stats from this survey alone don’t allow me to draw that conclusion. If they are I suspect it’s a weaker relationship than for Radical Party supporters.
Radical Party supporters, on the other hand, tend to be older Ukrainians who live predominantly in rural areas across different regions of Ukraine, have lower levels of income and education and feel more pessimistic about where Ukraine’s headed.
1. This is one poll, taken at one point in time more than seven months ago. I want to repeat this with more recent data to see if these trends hold or whether new ones emerge (or with different data if someone wants to give it to me).
2. The small sample size of decided voters (less than half of the original sample of 2,000 Ukrainians) really inhibits the amount of analysis I can run – thus why you see some of these oversimplified ‘higher/lower’ categories. This means some of the possible nuances between the cracks don’t get captured (e.g., between four levels of perceived income or education, etc.). This also so means that some differences that weren’t statistically significant here could show up as significant in different, larger surveys.
3. (2a?) The sample size isn’t remotely big enough to try and do more complex analysis (e.g., logistic regression) to determine what variables (e.g., gender, age, etc.) make the biggest impact on far-right or Radical Party support. Bah.
[usual preamble: polling done from December 16 through 20th in only government-controlled areas of Ukraine (i.e., no Crimea, no “DNR/LNR”), 2,000 respondents 18+, MoE 2.3%. Graphs below are mine, which mean any errors are likely mine]
1. Trust for some, not-particularly-high-levels-of-trust for others
Razumkov/DIF has asked Ukrainians for while now how much they trust a number of social institutions, from the President to the armed forces to NGOs (pages 13 through 15 of the PDF, in the back, for anyone following along at home). I’ve picked out a few from the 25-or-so list below.
These figures should (should) be worrying to anyone at the Bankova or the Rada – trust in the president (49% in 2014 to 24% in 2016) and in the Rada (31% to 12%) has tanked while trust in political parties (11% in 2016) is even lower. I haven’t graphed it out here but there’s obviously also been a corresponding increase in those who say they distrust the President (44% 2014: 69% 2016), the Rada (57% 2014: 81% 2016) and political parties (71% 2014: 78% in 2016). Keep in mind too that not a single individual Ukrainian politician is more trusted than distrusted (pages 5 and 6, question 7), so, ouch.
But who do Ukrainians trust more than their politicians? The armed forces, volunteer battalions, the National Guard, volunteer organizations and churches, not to mention Ukrainian media. Hmm.
2. We shall overcome, eventually
Ukrainians are less likely now than in 2014 to think the country’s “difficulties” (“труднощі” – native speakers, yell at me if there’s a better translation) will be overcome in the next few years, but I’d chalk this up more to realism than downright pessimism – the percentage of Ukrainians who say the country won’t be able to overcome its difficulties at all hasn’t significantly changed since 2014.
3. Optimists, pessimists and everyone in between
Buried way in back on page 14 of the PDF are some stats on whether a series of situations (24 of them) have got better, worse or stayed the same over the past year. I’ve picked a few below from both times DIF/Razumkov have asked the question.
Not too many Ukrainians felt things have got better over 2015 and 2016, but what’s most striking to me here is the percentage of Ukrainians who think the country’s international image hasn’t got better over 2016.
(I’ve drawn up the “worse” and “stayed the same” graphs separately, largely so I didn’t have to produce one graph with an absolute rainbow of shit in it.)
What do people think has got worse? Ukraine’s international image, for one, alongside social protection and the situation in Ukraine as a whole.
The “stayed the same” graph makes sense once you’ve seen the better/worse graphs but on its own here is pretty much useless. Whatever.
I’m not too surprised by much here, except the “Ukraine’s international image” question. What were respondents thinking exactly when they answered this?
4. Party time!
I’ve drawn up some graphs of party ratings over 2016 in DIF/Razumkov’s polls as well as KIIS’s polls for comparison.
I’ve included major (generally +5%) parties plus those scary right-wing parties that inherently must accompany any discussion of Ukrainian politics in English regardless of whether it feels like some of these parties could fit all their voters on a couple of school buses.
Batkivshchyna’s gone up some, BPP and Samopomich have gone down some, the Opposition Bloc may have dropped a bit, Lyashko’s Radical Pitchforker Party’s generally stayed the same and Svoboda and the other three Peoples’ Front of Judaea sects are wheezing after participation ribbons. That’s about it.
I think the bigger story here – and what I’d personally like to focus on more as I nerd around with polls like these and raw data sets in 2017 – is that more Ukrainians say they don’t know who they’re going to vote for, and even more (depending on the survey) say they won’t vote at all. Yeah, looking at the jockeying and horseracing from all these parties-spun-off-from-other parties and renamed-and-renamed-again parties tells us something, but it doesn’t tell us enough. Who isn’t voting? Why? Who doesn’t know who they’re going to vote for? Why? Do their social/political attitudes differ from decided voters? And so on.
I’ve had some of this data kicking around for a few days, in some cases a few weeks. Enjoy.
In the May 2016 wave of the Kyiv International Institute of Sociology’s Omnibus (the most recent publicly-available wave), respondents were asked (KIIS’ translation from Ukrainian/Russian versions):
If to speak about Ukraine in general, how do you think, in one year from now the situation in Ukraine will be better or worse than now?
Only 10.6% of Ukrainians think things will be better; 31.3% think it’ll be the same and 47.5% think it’ll be worse.
People who think Ukraine will be better were more likely to say they’d vote for Bloc Petro Poroshenko (14.5% compared to 7.1% who said the situation will be the same and 3.2% who said it’ll be worse) and Samopomich (8.8% compared to 4.4% who said the situation will be the same and 3.2% who said it’ll be worse).
Conversely, people who think the situation in Ukraine will be worse were more likely to say they’d vote for the Opposition Bloc (9.1% of those who said the situation’ll be worse, compared to 3.7% who said it’ll be the same and a whopping 0.9% [!!] who said it’ll be better).
People who think the situation in Ukraine will be worse are less likely to say they’ll vote (28.1% compared to 21.0% same and 13.4% better).
Young Ukrainians (18 to 29 years old) are more likely than older Ukrainians (13.9%, compared to 8.0% of 60 to 69 year olds and 8.6% of those aged 70+) to think the country will be better.
The sausage fest on the far right
In the same May 2016 Omnibus just 2.9% of women said they’d vote for a far right party (i.e., Svoboda, Pravyi Sektor or Yarosh’s national whatever), compared to 6.9% of men (4.7% overall). Even with such a relatively small sample size for far-right supporters, this is a statistically significant relationship.
Same goes for the February 2016 Omnibus, when 2.8% of women and 5.3% of men would vote for a far-right party (3.9% overall)
The May survey also asked a number of questions about which social issues troubled people the most, asking respondents to choose the top three from a list. “Revival of the Ukrainian nation” was chosen by 4.2% of people –5.2% of men and 3.4% of women.
No, that’s not a huge difference (a Cramér’s V of .05 for anyone who cares) but it’s still a big enough one to be statistically significant.
Relevant/unsurprising too: 13.9% of supporters of far-right parties chose “revival of the Ukrainian nation” in their top three compared to 4.2% of all Ukrainians….
Women were more likely than men (27.9% compared to 21.5% in May 2016; 27.7% compared to 20.3% in February 2016) to say they were undecided, that they didn’t know who they were going to vote for.
There’s a separate post or article or two buried just in those numbers and others like it. Interpretations and perspectives more than welcomed, particularly from members of the human race who are a) Ukrainian and b) not men.
Depression in Ukraine
Lastly, some data from the 2012 European Social Survey (ESS), the last wave Ukraine took part in. Despite the name, fieldwork actually took place in July and August 2013 for anyone keen to place this data within a very specific pre-Maidan time frame.
Friday they held their founding congress where, among other things, they agreed on policies to expand presidential powers, sever ties with Russia, expand the right to bear arms and “[restore] the death penalty…for treason and the embezzlement of government funds by officials in excessive amounts” (uhh wut), capping off the day with a torchlitmarch through central Kyiv.
If they’re really, properly serious about becoming an electoral force (which is doubtful when their new leader says things like “[there] are several ways of coming to power, but we are trying something through elections, but we have all sorts of possibilities”[my emphasis]), they’re stepping into a crowded and not-particularly-profitable far-right marketplace.
Just look at recent KIIS poll numbers for the far-right/nationalist parties, below, shaded in boring grey (and that, as I’ve noted before, are far below those for other European far-right parties).
It’s the same story across all the polls over the last year – there’s “an already fragmented radical electorate,” and one that, even hypothetically (i.e., impossibly) combined, can barely toe past Ukraine’s 5 percent electoral threshold. And now this National Corps wants to squeeze itself in? Bonne chance, gentlemen. They might be able to fare better than some of their comrades in the short term – I predict they’ll poll reasonably well in the first few polls they show up in – but I imagine they’ll find their <5% ceiling in due course.
Yes, I’m far from Kyiv right now and have more familiarity with a bunch of numbers than anyone on the ground there more up to speed with the ins and outs of Ukraine’s factionalist far-right. Still, I don’t think this has much to do with proper, democratic party politics. To me this move seems to be about little more than trying to lend a veneer of legitimacy to a very small minority of football hooligans and hardcore hangers-on — a minority who remind me more and more of a gaggle of wannabe Montagnardsthan a genuine political party.
I found some interesting numbers in one of the past waves of the Kyiv International Institute of Sociology (KIIS) Omnibus survey.
In the September 2015 wave (specifically from September 9 to September 24, for anyone wanting that level of detail), KIIS asked more than 2,000 Ukrainians in face-to-face interviews:
Would you like to see your oblast secede from Ukraine and become an independent state?
Would you like to see your oblast secede from Ukraine and join another state?
KIIS interviewed in all Ukrainian-government controlled areas of Ukraine; they didn’t interview in Crimea or in the “LNR,” though they did interview in the “DNR” (they no longer do). For the overall Ukraine and regional numbers I’m only including Ukrainian government-controlled areas – I talk about the “DNR” numbers towards the end of this piece.
The resounding answer to both these questions? No – fewer than 2% of Ukrainians are interested in their oblast seceding and becoming an independent state or joining another state.
The regional splits are pretty interesting, with western Ukrainians leading the very-small-number separatist brigade (4% for both questions) and eastern Ukrainians at the head of the slightly-larger-but-still-not-very-big crew of those who are ambivalent about separatism (11% for both questions). Boring graphs below.
I’ll draw your attention to the fact that the overwhelming majority of people in eastern Ukraine (that is, Kharkiv oblast and the Ukrainian government-controlled parts of Luhansk and Donetsk oblasts) state that they don’t want their oblast to secede and become either an independent state (88%) or join another state (85%).
Looking at some of the numbers by oblast is more striking, even when taking the small sample sizes into consideration.
Re: becoming an independent state:
Just under 9% of people in Zakarpattia (far western Ukraine, bordering Poland, Slovakia, Hungary and Romania) said they’d like to see their oblast secede from Ukraine and become an independent state – the most of any oblast. There isn’t a particularly large sample size here but given what I know about the history, politics and demographics of Zakarpattia I’m not surprised to see higher than other oblasts.
Only 3% of people in Odessa oblast and in both Luhansk and Donetsk oblasts (reminder, only the Ukrainian government-controlled parts) said they’d like to see their oblast secede from Ukraine and become an independent state.
Exactly 0.00% of people in Kharkiv, Chernihiv, Zaporizhia and Sumy oblasts said they wanted to see their oblast become an independent state. Anyone wondering about the digits past the first two decimal points there should be advised it’s an infinite string of zeroes.
Re: joining another state:
Like above, just under 12% of people in Zakarpattia said they’d like to see their oblast secede from Ukraine and join another state. Again, the most of any oblast and not surprising.
Just over 6% of people in Donetsk oblast and 5% in Luhansk oblast (again, only the Ukrainian government-controlled parts) said they’d like to secede from Ukraine and join another state. Remember, the other parts of these oblasts are occupied by ostensibly separatist statelets…
More than 3% of people in Zaporizhia oblast, less than one percent of people in Dnipro and Kharkiv oblasts and a round 0.00% of people in Chernihiv and Sumy oblasts said they wanted to see their oblast join another state.
But what of separatism in the apparently separatist “DNR”? As I pointed out a few months ago, polling and surveying in a place the “DNR” can’t possibly be easy. I can only imagine how bad social desirability bias (“the tendency of [respondents] to give socially desirable responses instead of choosing responses that are reflective of their true feelings”) might be in an globally-unrecognized puppet state with a laundry list of human rights abuses to its name. It’s likely one of the main reasons no one, as far as I know, tries to do surveys there anymore.
So what did people in the “DNR” have to say about separatism compared to their Ukrainian government-controlled Donetsk oblast brethren across the “ceasefire” line?
As you can see below, the numbers are pretty different; they suggest a plurality in the “DNR” is in favour of becoming an independent state, but are equally split when it comes to joining another state (can’t imagine what other state they were thinking of).
Even if – if – these numbers actually reflect how residents of the “DNR” feel, it’s far from a ringing endorsement of separatism, especially when you consider the level of Ukraine-crucifies-children-level propaganda people there have been hit with over the past two years. And the same percentages of people say yes and no when they’re asked if they want to secede from Ukraine and join another state? Hardly confident numbers for an entity that already considers itself a “separatist” state.
Of course, this all assumes that there are next to no issues with social desirability bias there at all, that respondents definitely wouldn’t feel pressured to say certain things to a stranger in a fiefdom where they could be thrown “in the cellar” for being critical of the regime. OK.
Looking for a popular secessionist movement? You’re better off coming to see me than going to Ukraine.
The Kyiv International Institute of Sociology (KIIS) released results from a new poll this week (link in Ukrainian).
They talked to 2,040 people between September 16 and 26, 2016 across Ukraine except for a) Crimea and b) areas of Donetsk and Luhansk oblasts not controlled by the Ukrainian government (i.e., the useless-passport-offering “LNR”/”DNR”).
KIIS’ key findings, in brief:
Over the past six months support for Batkivshchyna has declined slightly, while support for Bloc Petro Poroshenko has increased slightly.
Support for the Opposition Bloc, Samopomich, Oleh Lyashko’s Radical Party and Hromadyanska Posytsia has remained largely the same.
KIIS’ data over time is below. If you have some moral objection to the quality of my jpeg you can go look at the original link.
Keep in mind this table above only includes Ukrainians who a) said they’d vote and b) knew who they were going to vote for – which is only 41% of people who actually took part in the survey. The table below includes regional breakdowns as well as figures for non-voters, undecided voters and those pesky ballot spoilers (though no Ukrainian Edible Ballot Society yet).
A few thoughts and speculations of mine from across the sea. Comments and whatnot welcome.
What of Batkivshchyna and Bloc Petro Poroshenko?
The numbers for Batkivshchyna and Bloc Petro Poroshenko in this poll are a bit different from Rating’s poll that I talked about in August, despite fieldwork only being about a month apart.
Batkivshchyna’s 15.4% here versus 18.3% in Rating’s poll (2.9% difference)
Bloc Petro Poroshenko’s 14.5% here versus 9.1% in Rating’s poll (5.4% difference)
The other parties over 5% here (Opposition Bloc, Samopomich, Radical Party and Hromadyanska Posytsia) are largely the same. As for Za Zhyttia I’ve just learned they came out of the Opposition Bloc in May.
But why the difference for Batkivshchyna/Bloc Petro Poroshenko? The “difficult to answer” might have something to do with this – 31.1% (!!) of people in this poll said they didn’t know who they were going to vote for, compared to 16.3% in Rating’s poll. (I go on at length about this below)
That’s….a big difference, so big it makes me think something’s up methodologically here. Are interviewers from these different firms following up with people who immediately respond “don’t know” in the same way? For example, are Rating’s interviewers prompted by their script to ask again, which leads some of those “don’t knows” to respond with a party whose name they know – which could often be Batkivshvchyna? Are KIIS’s prompted to do the opposite, to accept the “don’t know” answer without prompting for a different answer? I know from experience that the way interviewers are led and trained to follow up with respondents can make a difference.
I’m purely speculating – and I’m not saying either one of these is right or wrong – but to see that big a difference between two different polls by different firms using the same method (in-person interviewing) makes me think it could be down to something like this.
Of course, another explanation could be that the apparent increase for Bloc Petro Poroshenko and the apparent decline of Batkivshchyna are reflective of an actual, significant change in what people think over the course of a month. I doubt this.
Region, region, region
No Ukrainian political party (as I understand it) has ever really been able to capture significant, meaningful support from all parts of the country. That doesn’t look like it’s about to change.
Parties like the PoR-remnant Opposition Bloc pick up almost all their support in southern and eastern Ukraine, while Lviv mayor Andriy Sadovyi’s Samopomich, not surprisingly, gets most of its support in western Ukraine. Even the two ‘big’ parties of Batkivshchyna and Bloc Petro Poroshenko can’t pull in consistent support across Ukraine, as you can see below in the absolute most cluttered PoS chart you’ll ever see.
When the bottom of the table’s more interesting than the top
I’m most interested in what’s going on down here in the relegation zone, which I haven’t bothered to translate because I’m a) too lazy and b) if you’re reading this you’re probably Ukrainian/speak at least some Ukrainian and can read it better than me anyways.
First, the 5.8% of people who said they’d vote ‘against all’ or spoil the ballot. That’s high, even just for a poll; by way of comparison, in that debacle of a Hungarian referendum last week 4% of voters spoiled their ballot in protest, a figure I still can’t get over.
Second, the 22.3% of people who said they wouldn’t take part in voting. I don’t have much to say about this – I don’t actually think that’s a remarkably high or low number, though the difference from west to east (17% compared to 28%) is worth noting. Is this down to more western Ukrainians heading their own more recognizable parties (e.g., Sadovyi, Lyashko, Hrytsenko, Tyahnybok), or is it a matter of different regional political cultures and attitudes towards democracy and (non) voting showing through? Both?
Lastly, the 31.1% of people who said “difficult to say” (don’t know, basically). As someone schooled in the arcana of Canadian, British and American electoral politics I have to admit I don’t think I’ll ever be able to wrap my mind around the idea of almost a third of people not sure who they’re going to vote for.
This 31.1% figure says more to me about Ukrainian party politics and the state of Ukrainian democracy than any other figure here. No party, whether small, not-as-small or in-theory-biggest, seems able to capture the interest and imagination of post-Maidan Ukraine at all. To quote Sonic Youth, fragmentation is (still) the rule.
I’ve got my hands on the most recent KIIS Omnibus data (that’s publicly available – from May 2016). Been running some numbers. Will have some reasonably interesting analysis of my own in a few days.