on the use and abuse of statistics, Instrumentalizing Minorities edition

on the use and abuse of statistics, Instrumentalizing Minorities edition

This is what happens, Larry. This is what happens when you present data from one question from a two-year old survey in a sexy shaded map and let the Twitters have at it.

The tweet and article in question from Pew Research Global, here.

This sexy map and the corresponding article (which I’m sure considerably fewer people actually read) have got some rave and not-so-rave reviews on the one social network that rots my brain more than any other:

  • Ukraine Twitter? Happy. These numbers (apparently) prove that “Ukrainians are the least anti-Semitic [people] in eastern Europe,” the “most accepting of Jews,” and “more tolerant to minorities than any of its neighbours.”
  • The national news agency of Azerbaijan? Happy. Armenia, apparently, is “the most anti-Semitic country in central and eastern Europe,” they tweeted Thursday morning.
  • Armenian/Armenian-watcher Twitter? Unhappy, if not confused, with some suggesting the high number for Armenia could be rooted in lack of Israeli recognition of the Armenian Genocide and/or Israeli supply of weapons to Azerbaijan.
  • Polish nationalist Twitter? Unhappy. “Dirty manipulation,” since western European countries like France weren’t surveyed, where a Holocaust survivor was murdered last week in what French police are calling an anti-Semitic crime.
  • Islamophobes? Unhappy, because apparently all anti-Semitism is the result of Muslims and these numbers don’t jive with that theory.

This map – but, honestly, more the fevered reaction to it over the last day on social media – has got me pretty unhappy too, but for some different reasons

First (and relatively minor, in fairness) lot of people are suggesting this data’s new. It’s not. This report was released almost a year ago, based on surveys Pew did in central and eastern Europe in 2015 and 2016 on religious belief and national belonging. I still have a copy of this report on my mobile from last year.

Secondly, I’m not a fan of the question(s) and the way they’re presented in the survey.

This is the exact question that was asked. Identical questions were asked about Muslims, Catholics, Orthodox Christians and Roma, in that order.UntitledI don’t like this question. Yes, I understand what it’s basically drawn from (Bogardus social distance scale questions), but look at it in the context of the preceding survey questions and it’s just….weird. (Page 18, here).

The preceding questions are all about religious observance – here’s the three that precede Q59.

Untitled
So we jump from these right into a question that’s basically HEY WHAT DO YOU THINK ABOUT JEWS

There’s no seguing into or introduction of the battery of questions on Jews, Muslims, Catholics, Orthodox Christians and Roma; we move from “what do you think about your religion?” to “so tell me what you think of these minorities and/or people different from you.” It’s an odd jump, in my opinion, and one where more than one respondent must have wondered what the hell was going on.

Worse, though, is the question itself. Again, I get that these are somewhat standard questions – I’ve seen similar ones in other unrelated surveys – but, really, does “Would you be willing to accept Jews as: a. Members of your family, b. Neighbors, c. Citizens of our country” even make sense to an average respondent??

It makes me ask far, far too many questions about whether this question, and the data from it, is even reliable. How has this question been translated into the many languages used in actually administering the survey? Does it make any sense in them? Does it make less/more sense in some languages? Or, worst, does it mean different things to different respondents in different languages? If I was at the table discussing this survey in its design phases I’d have been asking a lot of questions about all this.

But what bothers me the most here is how these findings – or, more accurately, findings from *one question* on a two to three-year old survey – are being exploited.

As I’ve written about before, there’s a healthy crew of people out there who are more than happy to use Jews (or at least imagined representations of Jews) as a means to score political points (e.g., ‘look how well we treat our minorities,’ not like [insert enemy country here or Muslims in general]). Worse, though, are the ones who use The Jews In The Way That We Imagine Them as a cudgel to bash other minorities in their own countries (read: Muslims and Roma).

This is hypocritical, creepy and, if you’re a member of a majority and you’re doing this, what you are doing is bad and you should feel bad.

Advertisements

An International Anti-Corruption Day addendum/hangover

An International Anti-Corruption Day addendum/hangover

It was International Anti-Corruption Day on Saturday – a good day, as I wrote in the Balkanist, to raise a glass and remember that Bulgaria, the most corrupt country in the European Union, is taking over the rotating Presidency of the Council of the EU* in three weeks.

Only after writing that piece did I see new survey numbers that had just come out from a recent Special Eurobarometer survey on corruption in the EU. Ah well.

As you might have guessed, the numbers for Bulgaria aren’t good.

18693092_403
“Kill corruption” (Credit: BGNES)

There’s a full report and a 20-page summary report available, so I’ll just rehash a few of the key (damning) findings about perceptions of corruption in Bulgaria, but still enough to make me feel like I’m suffering from some post-International Anti-Corruption Day makhmurluk:

  • It’s not just that Bulgarians think there’s corruption in their public institutions – many feel that it’s actually getting worse.
    • When asked whether there was corruption in local or regional public institutions, 86% of Bulgarians said there was. This isn’t just one of the highest figures in the EU – it’s increased 7% since the 2013 survey asking the same questions, which is the biggest increase in the EU.
    • As for corruption in national public institutions, 87% of Bulgarians said there was, again one of highest – and, again, the biggest increase since 2013 (5%), which is also especially striking given that this figure increased in just four of the 28 EU countries.
  • Only 14% of Bulgarians think there’s sufficient transparency and supervision of the financing of political parties, the lowest in the EU (but still risen from 9% in 2013).
  • Only 13% think there are enough successful prosecutions in the country to deter people from corrupt practices. Yes, the lowest.
  • 83% of Bulgarians, second only to Greece, think that high-level corruption cases are not pursued sufficiently in the country.
  • Are measures against corruption applied impartially and without ulterior motives? Only 15% of Bulgarians say so – the lowest in the EU.
  • In 20 of the 28 EU countries, there’s been an increase in the proportion of respondents who think government efforts to combat corruption are effective. Bulgaria’s not one of them – only 15% agree, the second worst in the EU behind Latvia.
  • Bulgarians don’t know where to report corruption: only 28% say they know where they’d report an act of corruption, with only Hungary being lower. Worse, this figure’s declined 15% (!!) since 2013, the biggest decrease in any EU country.
  • Bulgarians aren’t particularly trusting of the police and/or customs, and they’re getting more distrustful over time:
    • When asked who they’d trust most to deal with a case of corruption they’d complained about, 25% of Bulgarians identified the police, compared to a 60% average across the EU;  that figure’s dropped 11% since 2013.
    • Worse, when asked whether they felt the giving and taking of bribes and abuse of power for personal gain was widespread across a number of institutions, 71% of Bulgarians named police/customs. This was the highest in the EU and the most common response in Bulgaria – only Latvia had police/customs as a most common response. For reference, the EU average is 31%.

These figures don’t make one feel all that confident about Bulgaria’s fight against corruption. Bulgarians themselves certainly don’t seem to be.

 

*not the “European Council” as I wrote in my piece, and I think I’ve inadvertently written before. That president is Donald Tusk. Having confusingly similarly-named institutions is a great idea, guys. 

 

a few choice words from Valeri Simeonov

a few choice words from Valeri Simeonov

As Bulgaria gets ready to take up the Presidency of the Council of the European Union in January, it’s time to hear from Valeri Simeonov – one of Bulgaria’s Deputy Prime Ministers (in charge of economy and demographical policies), co-spokesman of the far-right United Patriots and head of the Bulgarian Council on Ethnic Minority Integration.

  • Speaking in an interview on October 31 with a cable television channel, Simeonov said that “bTV turned out to be like an infantile fatty with certain mental abnormalities put into a porcelain shop, not knowing what he is doing.”
  • After it emerged that a deputy minister from the United Patriots had been photographed giving a Hitler salute, Simeonov reportedly made comments to the effect that in the 1970s, he had been taken on a visit to Buchenwald and “Come to think of it, who knows what kind of joke photos we took there…can anyone say now, submit your resignation and go back to the village”.
    • Also, “Is that really a Nazi salute? On what basis should he be withdrawn – that he’s a Hitlerite or a member of the Nazi party? Nonsense.”
  • “I cannot allow a handful of Sorosoids to badger us while we are trying to solve important problems.”
  • “…it is indisputed that a large part of the Gypsy ethnicity lives beyond any laws, rules and general human norms of behaviour. For them, the laws do not apply, taxes and charges are incomprehensible concepts, electricity, water, social and health insurance bills have been replaced by the belief that they have only rights, but not obligations and responsibilities. What has created the belief in our swarthy compatriots that everything is allowed… and that everyone is obliged to feed, dress and treat them for free?”
    • Roma are also, in his opinion, “naked, self-confident and ferocious humanoids ready to murder, to steal a few leva.”
    • …and Roma also want “sickness benefits without being ill, child care for children who play with the pigs in the streets and maternal benefits for women with the instincts of street bitches”.
  • In 2013 his National Front for the Salvation of Bulgaria (NFSB), in its platform, said it “initially envisages the abolition of all illegal structures in Gypsy ghettos and the creation with minimum resources of individual settlements outside the large settlements. Promotion of voluntary birth limitation through free contraceptives. The isolated settlements can be turned into a tourist attraction, which is a mass practice in the most developed democracies (Indian Reserves in the US, Aboriginal Settlements in Australia, Gypsy Settlements in the Czech Republic and Hungary, etc.).”

He also assaulted an elderly woman during a protest to stop Turks from crossing the border to vote in March’s elections. Oh, and he also just got convicted of breaking anti-discrimination legislation with some of his comments above about Roma.  Happy Friday!

 

The story behind that “terror attacks” map you keep seeing

The story behind that “terror attacks” map you keep seeing

You’ve seen this map somewhere on social media the last few weeks, haven’t you?

Here it is from Poland’s deputy justice minister, because somehow this is how you show solidarity with the citizens of a country millions of your own people live and work in.

2017-06-04 18.10.49

I’ve also seen it from Polish MEP and unsuccessful candidate for the presidency of the European Council Jacek Saryusz-Wolski, though my favourite version is the one tweeted out by that one guy who managed to get fired from The Rebel.

IMG_cyjezr
OOOOH this one has different colours!!

And last, a version I saw on Instagram this week.

IMG_3d1kpb
Cool.

So where the hell is this data even from? Turns out, as someone from the right-wing Polish Twitterati told me, it’s data from the reputable Global Terrorism Database (GTD) at the University of Maryland, which is an “open-source database including information on terrorist events around the world from 1970 through 2015.” (Notice right now that says “events,” not “attacks.” This will be important). I was further informed that, for some reason, the data on this map that’s been making the rounds is only from 2001 on. OK.

So I took a look through the GTD data on some of the countries (including Poland) on this map.  There are certainly no terror “incidents” (read, “incidents”) listed in Poland from 2001 on. OK, so that seems (seems) accurate.

But what about terror in other countries? I’m particularly interested in these apparent incidents in Iceland, which shows up in some versions of the map and, having been there, doesn’t exactly strike me as a terror hotbed.

Since 2001, there have apparently been two terror incidents in Iceland that explain the two Icelandic dots on the map:

  • In 2012, “An explosive device detonated near government offices in Reykjavik city, Reykjavik North Constituency, Iceland. The explosive device was partially detonated by a robot meant to deactivate it. No group claimed responsibility for the incident.” Property damage was listed as unknown.
  • In 2014, “Assailants attempted to set a Lutheran Church on fire in Akureyri city, Northeast constituency, Iceland. No one was injured in the attack; however, the building was damaged. No group claimed responsibility for the incident.” Property damage is listed as “minor.”

No one was killed or injured in these two incidents.

Again, “incidents” is the key word. These two big red Icelandic points, and many others on the map, don’t represent terror attacks at allMany of them, including these two in Iceland, represent vague criminal acts that may not actually have anything to do with terrorism (let alone jihadist terrorism), that have barely caused any property damage and, more importantly, haven’t killed or injured anyone.

Why no Polish incidents in the GTD since 2001? Surely there’s been at least one shitty attempt at something like a pipe bomb in a car that never went off (there was one in the Czech Republic database, as I discovered) that would merit a mention in this database, though presumably this will make it into 2017’s list for Poland, given the criteria for inclusion.

So the next time you see this map, you’ve got a few options. If it’s got no legend or title, you can always tell whoever shared it that the points represent vague definitions of criminal acts that don’t always seem to be reported consistently. If it says something about “terror attacks,” tell them they’re completely, 100% wrong, and tell them there’s more than enough data on the GTD website for them to make a proper map of actual terror attacks that isn’t just a cute meme for people who don’t like Muslims.

Donbas, elections and visa-free travel: latest Razumkov Centre poll

Donbas, elections and visa-free travel: latest Razumkov Centre poll

New Ukraine survey/poll data, new blog post.

The Razumkov Centre just released findings from a survey. Boring stuff: more than 2,000 respondents, interviewed between Sept 9-14, 2016 across all of Ukraine except Crimea and the “LNR”/”DNR”, 2.3% MoE.

A few findings I found particularly interesting:

Special status

Half (50%) of respondents did not support granting special status to Donbas, with 23% agreeing and 27% being unsure (“difficult to say”).

The regional numbers don’t throw up any surprises, with respondents from western Ukrainian being the least supportive and respondents in eastern Ukraine in Donbas being the most supportive.

graph-1

I’d like to see breakdowns by a few other factors, particularly age and level of education (presumably done behind the scenes already? Not significant, possibly), but the Razumkov Centre does provide us with a comparison of how people responded to these questions six months ago.

graph-2

It’s not the decline in those saying ‘no’ over the two surveys (56% down to 50%) that grabs my attention here – it’s the increase in those respondents who aren’t sure (20% up to 27%) when it comes to special status for Donbas. Based just on these numbers here (i.e., without a bunch of data to nerd around with, whether the sample sizes are large enough to show regional trends over time, etc.) I’d say that an increasing number of Ukrainians aren’t sure what ‘special status’ means at all.

Elections, elections

The numbers aren’t all that different when it comes to elections in the occupied territories in Donbas (the “DNR”/”LNR”) – just under a quarter (24%) support them, half (51%) don’t support them and a quarter (25%) aren’t sure at all.

Again, the regional breakdowns are no surprise (though I’ll highlight, as per the Razumkov report, that the differences for eastern Ukraine and the Donbas aren’t statistically significant).

graph3

Want to see the most useless table ever? It’s right there. There’s basically been no significant change over six months on this one.

graph4

Visa-free regime with Europe

In the same survey (presented in a different document), people were asked what they thought about the still unrealized visa-free travel regime with the EU. Just over a third (35%) said visa-free travel with the EU was important or very important to them and, not surprisingly, it’s a lot more important to people in western Ukraine.

image-7
Yes, the colour scheme was a deliberate choice.

Fortunately the Razumkov Centre points out something about age before I can ask about it – younger respondents (aged 18-29) were more likely to say that a visa-free regime was important or very important (61%) than respondents 60 years and older (18%). Not particularly shocking.

But most Ukrainians aren’t expecting visa-free travel to land anytime soon. Only 7% expect it by the end of this year, 45% expect it in 2017 (either early or later in 2017) while one in five (20%) say it’ll never come. That pessimism is much stronger in eastern and southern Ukraine than anywhere else.

image-8

Comments/interpretations welcome.

Why EU attitudes in Ukraine are more than just east versus west

Why EU attitudes in Ukraine are more than just east versus west

To celebrate Europe Day in Ukraine and to drown out all that noise coming from Mikhailivsky Square I decided to go back and revisit something I wrote a few weeks ago.

I analyzed data from the most recent Omnibus Survey run by the Kiev International Institute of Sociology. Every three months KIIS surveys 2,000 Ukrainians, the most recent data being from February.

Their ‘attitudes towards the EU’ question:

Please, imagine, that now is a referendum on whether Ukraine should join the European Union. You can vote for, against or abstain from voting. What would you choose?”

Just under half (49%) of Ukrainians said they’d vote in favour of EU accession, and 28% would vote no. A few (9%) said they’d abstain, but about one in seven (14%) said they didn’t know how they’d vote.

And yes, there are regional differences, which is the point of all this:

  • Western Ukraine: 77% yes
  • Central Ukraine: 52% yes
  • South Ukraine: 36% yes
  • East Ukraine: 25% yes

If you feel like you can stop here and convince yourself that it’s all a matter of east vs west, ‘pro-Russian’ versus ‘pro-Ukrainian’ regions – please, for the love of all things holy, read on.

South

Point form.

  • There’s a divide between young and old. Almost half of 18-29 year olds (45%) and 30-39 year olds (46%) support EU accession in southern Ukraine, compared to 23% of 60-69 year olds and 12% of those older than 70.
  • There are some pretty big differences when it comes to levels of education. Those who had the highest levels of education (some level of higher education/degree) were more likely to support EU accession (46%) than those with less than ten years’ education (a whopping 6%).
  • People who indicated they were on some sort of state pension were less likely to support EU accession (22%).
  • People in the lowest of five socioeconomic brackets (those who said they ‘lack money for food’) were much less likely than those in the third (and most common) bracket (those who said they ‘have enough money for food and clothes’) to support EU accession (15% compared to 46%).
  • Self-identity as (more) Ukrainian or Russian also played a role.
    • Self-identity as Ukrainian: 41% yes
    • Self-identity as equally Russian and Ukrainian: 17% yes
    • Self-identity as Russian: 0% yes. Yes, zero – meaning that no one surveyed self-identified as Russian in south Ukraine and also supported EU accession
    • Affiliation with Kyiv Patriarchate (45% yes) compared to Moscow Patriarchate (19% yes)

East

More point form.

  • There’s also a divide between young and old. More than a third of 18-29 year olds (35%) and 30-39 year olds (41%) support EU accession in eastern Ukraine, compared to 17% of 60-69 year olds and 12% of those older than 70.
  • Education differences here too. Those who had the highest levels of education (some level of higher education/degree) were more likely to support EU accession (41%) than those with less than ten years’ education (11%).
  • People who indicated they were on some sort of state pension were less likely to support EU accession (12%).
  • Interestingly, the socioeconomic questions didn’t show any significant differences. I’ll get to that.
  • Self-identity as (more) Ukrainian or Russian again.
    • Self-identity as Ukrainian: 37% yes
    • Self-identity as equally Russian and Ukrainian: 2% yes
    • Self-identity as Russian: 7% yes.
    • Affiliation with Kyiv Patriarchate (41% yes) compared to Moscow Patriarchate (8% yes)

The survey data from eastern Ukraine also includes respondents from the ‘DNR.’ As I talked about a few weeks ago, results from the ‘DNR’ stand out pretty starkly against the rest of eastern Ukraine.

So what do the results look like if we just look at Ukrainian government-controlled eastern Ukraine?

East, sans ‘DNR’

The last bit of point form.

  • Age, yeah. Almost half of 18-29 year olds (44%) and more than half of 30-39 year olds (52%) support EU accession in eastern Ukraine, compared to 22% of 60-69 year olds and 13% of those older than 70.
  • Education. Those who had the highest levels of education (some level of higher education/degree) were more likely to support EU accession (50%) than those with less than ten years’ education (11%).
  • People who indicated they were on some sort of state pension were less likely to support EU accession (15%).
  • Take out the ‘DNR’ and socioeconomic status become significant, because statistics is fun. People in the second lowest of five socioeconomic brackets (those who said they ‘didn’t have enough money for clothes’) were much less likely than those in the third (and most common) bracket (those who said they ‘have enough money for food and clothes’) to support EU accession (24% compared to 47%).
  • Self-identity as (more) Ukrainian or Russian again.
    • Self-identity as Ukrainian: 38% yes
    • Self-identity as equally Russian and Ukrainian: 6% yes
    • Self-identity as Russian: 7% yes
    • Affiliation with Kyiv Patriarchate (44% yes) compared to Moscow Patriarchate (17% yes)

What do all these numbers mean?

There’s a lot more to Ukrainians’ attitudes towards the EU than just what part of the country they happen to be from. When almost half of young people in eastern Ukraine support EU accession, engaging in this ‘OMG pro-Russian east versus pro-Ukrainian west blah blah’ is more than just lazy oversimplification. It’s wrong.

And the factors that are associated in eastern and southern Ukraine with being opposed to EU accession? Being older. Being less educated. Being a pensioner. Being from a lower socioeconomic bracket. In other words: class, with a healthy dose of (self-) identity thrown into the mix. You know, just like that other country having a wee chat about the EU right now.  Ukraine’s not always that different, guys.

If there’s anything to take away from how I decided to spend my Saturday evening, it’s a point that’s been made before, repeated before and which, unfortunately, will need to be repeated again:

There is no magical giant line down the middle of this country that divides it into ‘pro-Russian’ and ‘pro-Ukrainian’ parts.

 Please, no more.

What do Ukrainians think about the EU?

What do Ukrainians think about the EU?

When people give me access to lovely nationally representative data sets, I am a happy man.

I got my hands on the data set from the most recent Omnibus Survey run by the Kiev International Institute of Sociology. Every three months KIIS surveys 2,000 Ukrainians about everything from who they’d vote for to what kind of presents they’d like to get for New Year’s. The most recent data is from February 5-16, and fieldwork for the next round is beginning in just a few days.

One of the first questions they asked was about attitudes towards the EU (KIIS’s translation from the Ukrainian and Russian versions of the questionnaire):

Please, imagine, that now is a referendum on whether Ukraine should join the European Union. You can vote for, against or abstain from voting. What would you choose?”

Just under half (49%) of Ukrainians said they’d vote in favour of EU accession, and 28% would vote no. A few (9%) said they’d abstain, but about one in seven (14%) said they didn’t know how they’d vote – a figure I’ll talk more about below.

UA and EU accession

Not surprisingly, there are some pretty stark regional differences, with more western Ukrainians (77%) inclined to support EU accession than anyone else, particularly eastern Ukrainians (25%).

EU UA region

I’ve tried to unpack those figures from southern and eastern Ukraine, to the extent that the sample size allows me to make confident comparisons – and for some of the larger oblasts, it does:

  • Donetsk oblast (including DNR): 57% of those surveyed said there weren’t in favour of EU accession; only 19% said yes
    • Just in the DNR, 74% said no, and only 6% said yes
  • Luhansk oblast (not including LNR): 56% said no; only 23% said yes
  • Odessa oblast: 38% said no; only 26% said yes and, interestingly, 21% said they didn’t know how they’d vote
  • Kharkiv oblast: 28% said no; 37% said yes

A few other differences of note:

  • Women were more likely than men to say they didn’t know how they’d vote (17% of women compared to 11% of men)
  • Younger people (between the ages of 18-29 and 30-39) were more likely to say they’d vote yes (57% and 58%, respectively) than those aged 60-69 (39%)
  • People with the lowest levels of education (less than ten years of education) were least likely to vote yes (23%), but at the same time most likely to say they didn’t know how they’d vote (31%)

Do these differences matter?

Well, some matter more than others.

I ran a few logistic regression analyses, because a) this is something I know how to do despite never really ever wanting to, and b) it’s a good way to parse out what matters most in yes/no survey questions.

What did I find?

Yes: Living in western Ukraine was by far the strongest predictor of voting Yes. This is probably the least surprising finding I’ve ever found running one of these things and I assume no one even remotely connected to the Ukraine-watching universe is shocked by this. The other differences I mentioned above (age, etc.) come up insignificant in the regression analysis.

No: Living in southern and eastern Ukraine were strong predictors of voting No. Interestingly, living in southern Ukraine was a stronger predictor than living in eastern Ukraine – I’m curious to see whether this holds in up in future surveys. Also, self-identification as Russian or ‘mostly Russian’, as well as being a member of the Moscow Patriarchate, predicted voting No. Not particularly shocking.

Don’t know: The only predictor of saying Don’t Know was being a woman.

What does this all mean?

Most of this is obvious and confirms what we all already know, but what I’m intrigued by are the relatively high numbers of Don’t Knows in the overall figures (14%).

I’m curious about the high Don’t Knows in Odessa oblast (21%), and whether this speaks to something below the surface there or whether it’s just something that’ll fall out in the wash in future surveys.

What I’m most curious about though is so many of the Don’t Knows being women – almost one in six women don’t know how they’d vote, compared to just over one in ten men.

Are they less willing to give a yes/no answer, (i.e., is there some sort of gender-related social desirability bias here?) or do they genuinely not know how they’d vote? If so, why?

And, most importantly, what would convince them to say Yes or No?