8

According to Newsweek1:

One in six French citizens sympathises with the Islamist militant group ISIS, also known as Islamic State, a poll released this week found.

Is this true?

116% of French Citizens Support ISIS, Poll Finds. Visited 09/07/2018. http://www.newsweek.com/16-french-citizens-support-isis-poll-finds-266795

Jordy
  • 3,846
  • 2
  • 24
  • 34
Probably
  • 887
  • 1
  • 6
  • 14
  • @Probably, [welcome to Skeptics!](https://skeptics.meta.stackexchange.com/q/1505/39853) I edited your question a bit, feel free to change it if you think I didn't do it correctly. – Jordy Jul 09 '18 at 09:48
  • 2
    16% of respondents to a poll indicated having sympathies towards IS. Without knowing the demographics of said respondents, and the questions asked, it's impossible to say how well a poll reflects the total population. The article itself suggests it may well correspond to the French MUSLIM population, which of course is far from the entire population. – jwenting Jul 09 '18 at 09:52
  • 3
    Also note the weasel wording of the article and its failure to mention the exact questions asked. For example, answering yes to "do you think homosexuality goes against the word of god?" could be interpreted as supporting ISIS policies of murdering homosexuals by far-right anti-Muslim groups. – dont_shog_me_bro Jul 09 '18 at 09:53
  • 1
    Not a full answer, but there are two big red flags in this claim: it was a poll made for a government owned and operated news agency, Rossiya Segodnya, and more than half of the 16% of French support would have to come from non-Muslims due to France's Muslim population being [7-9%](https://www.cia.gov/library/publications/the-world-factbook/geos/fr.html#People) of total population, which just does not seem realistic. If the original poll(which I haven't been able to find) was regarding just French Muslims, rather than all citizens, then there's a very slim chance it's possible. – Giter Jul 09 '18 at 12:38
  • 1
    @MichaelK I think it's incorrect to attribute the study to Russia Segodnia. [According to Vox](https://www.vox.com/2014/8/26/6067123/isis-poll) it was conducted by [ICM Research](https://en.wikipedia.org/wiki/ICM_Research). I doubt Rossiya Segodnya could have themselves conducted a study in UK, Germany and France. They likely contracted a Western firm for that. According to the same study, 7% in the UK had a favorable view of ISIS. – Fizz Jul 09 '18 at 20:19
  • @jwenting the question asks about "citizens" and you are speaking of "population." It would be good to know whether the poll even asked about citizenship. – phoog Dec 10 '18 at 16:07
  • 2
    "sympathizes with" and "supports" are two very different things, especially in cases like this. You can have all sorts of people who sympathize at least somewhat with an extremist group and/or the people in it, but who still disagree with its methods and/or goals, and would never support it. – Ben Barden Dec 10 '18 at 19:03
  • @phoog indeed, especially as the percentage of muslims under the citizenry is likely lower than under the population as a whole (the majority of new migrants over the last several years, maybe decades, into many European countries have been muslims after all, and many of those never gained citizenship yet still live there). – jwenting Dec 11 '18 at 04:48
  • @BenBarden moot point. They still would give moral support, propagandise their goals and targets, and often try to gloss over the methods the extremists use in my experience talking to them. "I would never do that but I can understand why they do it and share their goals" is the least such arguments. – jwenting Dec 11 '18 at 04:50
  • @jwenting It's really *not* a moot point. First of all, "occasionally speaks in favor of" is basically the high end of "sympathizes", and the low end of "supports". Second... your argument looks like it boils down to "acknowledging that there might be any validity to their complaints is equivalent to joining their crusade". That's a solid rhetorical position for attacking everyone who even starts to disagree with you, but it's lousy for actually letting yourself understand what's going on. – Ben Barden Dec 11 '18 at 14:15

2 Answers2

16

The French news magazine L'Express reported on this poll. They give information about the poll that complements the Newsweek article. In particular, the poll question was (my translation): ”According to your knowledge, tell us whether you have a very positive, somewhat positive, somewhat negative or very negative opinion of the Islamic State in Iraq and the Levant, also known under the name ‘ISIS’”. Of note, the question did not explicitly mention either Syria, and did not mention the name “Daech” which was most common in reporting in French media. Also note that the question asked how favorable an opinion respondents had, which is not quite the same thing as asking whether they supported ISIS.

Among French respondents¹, 3% declared that they had a “very positive” opinion of ISIS, 13% “somewhat positive” opinion, 31% “somewhat negative”, 31% “strongly negative”, and 23% hadn't heard of it. According to L'Express, the margin of error is around 3%.

Although the poll was commissioned by an organization belonging to the Russian government, it was carried out by a British public opinion research company, ICM Research. Citing the pollsters:

respondents were selected by random digit dialling (80% landline, 20% mobile) and interviews were conducted by a live interviewer. Responses were weighted to age, gender, profession of head of household and Agglomeration category.

This is very strong evidence that the poll did take place. It is hard to get independent evidence that the poll results match what the polling institute reported, since all we have is their word for that, but it would generally not be in their business interest to lie since that could lead to being considered incompetent.

I could not find reports of an independent poll with a similar question in a similar time frame.

Keep in mind that the poll took place in July 2014, two to four weeks after the group had announced the establishment of a caliphate. If you'd like to refresh your memory, this is what Wikipedia had to say on the topic at the time (French version). In 2014, ISIS was something that happened far away, a faction in a civil war where the biggest other side (the Syrian government) was also evil. Be very wary if this poll is cited now without recalling the historical context: this was before ISIS committed terrorist acts in France.

¹ The pollsters seemingly did not ask whether the respondents were French citizens, so this is a poll of presumed (metropolitan) French residents who have a phone, weighted like any other poll according to the pollster's weighting methods and to a bias for ability and willingness to answer. About 6% of French residents are not citizens and about the same proportion of French citizens live abroad.

Gilles 'SO- stop being evil'
  • 3,182
  • 1
  • 26
  • 28
  • "“Daech” which was most common in reporting in French media". L'Express alas does not raise this point... or even mention the word Daech in that article. I've seen the argument however (in a less clear version, no mention of "Daech") in Washington Post. – Fizz Jul 10 '18 at 15:29
  • @Fizz True, I may be anachronistic too. “Daech” became a household name in France in 2014–2015 but at the time of the poll, it wasn't so widely known and possibly not yet under that name. I'll look through 2014 newspapers to see what they were using at the time. – Gilles 'SO- stop being evil' Jul 10 '18 at 16:11
  • I'm surprised that 77% of respondents had an opinion on this organization: in 2014, most French people wouldn't really know what ISIS is, so the correct answer for them should have been "I don't have enough elements to answer this question". – Cœur Dec 10 '18 at 10:33
  • The question asks about French citizens. If the poll did not ask about citizenship, which it seems not to have, then a word or two might be in order about the percentage of French telephone subscribers who are French citizens, and perhaps the percentage of French citizens who are not French telephone subscribers (I know several where I live in the US, for example). – phoog Dec 10 '18 at 16:10
  • @phoog I added a note, but this feels like nitpicking. Do you have any particular reason to think this would matter? I do expect a positive correlation between living abroad and being better informed about world news, and there must be some correlation between living abroad and certain political opinions but I have no idea which way it might swing the responses. However, this only concerns a small fraction of potential responders. – Gilles 'SO- stop being evil' Dec 10 '18 at 19:01
  • @Gilles I think it matters because misquoting and misunderstanding statistics is a common problem, and a major source of [pseudoscience and biased results](https://skeptics.stackexchange.com/help/on-topic). Since [about six percent of the French population does not have French citizenship](https://www.insee.fr/en/statistiques/1304052), claiming that a poll of French residents represents the opinion of French citizens could be well off the mark, especially if there's reason to suspect, as there is here, that foreign residents might have a different outlook on the question. – phoog Dec 11 '18 at 04:55
4

Support? No. Had a somewhat (13%) (or very, 3%) favorable opinion, according to that poll. The most detailed article about the poll I found is in French (big surprise), in l'Express.

The poll was indeed ordered by Rossia Segodnia (a Russian news agency). But it was conducted by ICM Research in France, Germany, and in the UK. Journalists from l'Express contacted ICM research to inquire about methodology. It was seemingly a representative sample of ~1,000 people in each country, interviewed by phone.

Les sondés ont été interrogés du 11 au 21 juillet 2014, par téléphone. 3007 personnes ont été contactée: 1006 en France, 1001 en Allemagne et 1000 au Royaume-Uni.

Comme la notice du sondage laisse plusieurs zones d'ombres, L'Express a contacté ICM Research qui nous indique que l'échantillon de sondés a bien été élaboré selon la méthode des quotas, en fonction de l'âge, du genre et des régions françaises.

[an update says:] les catégories socioprofessionnelles et les emplois des personnes interrogées, qui n'étaient pas mentionnées mardi, ont été ajoutées. L'institut s'est excusé sur Twitter d'avoir publié une notice incomplète dans un premier temps

The actual results:

  • Français: 3% très favorables, 13% assez favorables, 31% assez défavorables, 31% très opposés, 23% ne connaissent pas le groupe
  • Britanniques: 2%, 5%, 20%, 44% (29% ne connaissent pas)
  • Allemands: 0%, 2%, 28%, 54% (16% ne connaissent pas)

A fairly high number of respondents didn't even hear/know of the group. Far more than those who expressed any kind of sympathy. This suggest that the poll came at a time when the group was not very prominent. Some commentary later that year in HuffPo confirms that:

The timing of the surveys may go some way to explaining the gulf in support. The ICM survey took place this summer, and was conducted prior to some of the group's most high-profile atrocities like the beheading of Western journalists James Foley and Steven Sotloff, and aid workers David Haines and Alan Henning.

And since that July 2014, another one in August, conducted only in the UK (and by a different company) found:

15-17 August 2014

A ComRes poll of 2,042 adults for ITV News asked how the British government should respond to IS, taking into account the level of military action necessary to achieve a particular outcome. In reply, 20% suggested that we should attempt to defeat IS in its entirety, 29% that we should seek to prevent IS making further gains, and 30% that Britain should not get involved and leave the situation to run its course (the remaining 21% were undecided). Just over one-third believed the British government should arm Kurdish forces who were fighting IS.

So the strong opinions one would get today about ISIS weren't so strong at the time. Incidentally, the 49% of British who did support military action against IS in the latter poll come reasonably close to the 44% who had a strong negative opinion of the group a month before.

There's some more commentary in Washington Post about the ICM poll:

ICM Research hasn't provided a lot of detail about the methodology in their press release, but they were able to offer some more to The Post. Our resident pollster, Scott Clement, says that while the methodology isn't perfect (the survey was largely conducted with calls over land lines, meaning that cellphone-reliant adults could be undersampled), it wasn't terrible either. Clement suggests that respondents could be misinterpreting the question or simply ignorant of what the Islamic State is: Hearing "The Islamic State of Iraq and the Levant" and thinking it is just talking about the country of Iraq. Given the continues naming issues that Islamic State/ISIS/ISIL has had, such confusion is understandable to a degree.

I'm not sure how valid it is... since the confusion would have to have been mainly in the minds of the French (why weren't Germans affected, for instance?). The French sources I found don't discuss this issue, but then the French translation/version of the survey isn't available either.

There's a news story from September 2014 on the naming issue:

The French government has decided to refer to the brutal militant Islamic caliphate Isis as 'Daesh' saying that the other acronyms used for the terror group 'blur the lines between Islam, Muslims and Islamists'.

French foreign minister Laurent Fabius asked journalists and media organisations to follow their decision, saying to France 24 yesterday: “This is a terrorist group and not a state.

“I do not recommend using the term Islamic State because it blurs the lines between Islam, Muslims and Islamists. The Arabs call it ‘Daesh’ and I will be calling them the ‘Daesh cutthroats’.”

But it's unclear to me if this naming issue was widespread enough in July to confuse the French in that poll.

Community
  • 1
Fizz
  • 57,051
  • 18
  • 175
  • 291
  • Although the Wikipedia article on ICM Research lists a lot of plaudits for the company, there was one other poll of theirs (this one ordered by Channel 4) relating to Muslims that raised eyebrows: https://www.nytimes.com/2016/04/15/world/europe/poll-british-muslims.html – Fizz Jul 09 '18 at 20:55
  • Daesh is just short for al-Dawlah al-Islamīyah fī l-ʻIrāq wa-sh-Shām. As one might expect from the name, it pretty much is just calling it the Islamic State of Iraq and the Levant. Like, if US politicians called it L'etat islamique de l'irak et du levant, they'd still be calling it the Islamic State. – Obie 2.0 Jul 16 '18 at 08:27
  • @Obie2.0 while that may be literally true, using the acronyms avoids reinforcing the misperception that some have that the group in question is somehow representative of Islam. It also avoids possible ambiguity between "the Islamic State" and "an Islamic state." – phoog Dec 10 '18 at 16:16
  • @phoog - I agree to an extent, but that potential ambiguity still exists for anyone who knows a smattering of Arabic. Besides, while this reason might seem a bit silly, I think it's the name they chose in English media, and I personally believe in respecting that choice even for "evil" groups like ISIS. – Obie 2.0 Dec 10 '18 at 17:14