Martin Boon Profile Banner
Martin Boon Profile
Martin Boon

@martinboon

4,550
Followers
102
Following
72
Media
4,014
Statuses

“Battle hardened veteran of the polling industry”.

London
Joined April 2009
Don't wanna be here? Send us removal request.
@martinboon
Martin Boon
5 years
I conducted focus groups in the NE just before election, and Labour-to-the-core voters despised Jeremy Corbyn & what he’d done the the party so much they could not vote for it even if grandparents turned in their grave. They didn’t mention Tony Blair much, except positively.
@LauraPidcock
Laura Pidcock
5 years
Vital to learn the lessons of Labour’s #GE2019 defeat. However, the answers will not be found at the door of New Labour’s architects. Blair’s legacy still hangs around this party like a millstone, especially in the North East. I heard it time & time again:
5K
2K
7K
75
571
2K
@martinboon
Martin Boon
2 months
“Bring back national service” is a line I’ve heard countless times in focus groups over the years, and typically by the mad focus group member who I had to put a lid on because they spoiled it for everyone else. That’s who the Tories are clearly now listening to.
7
39
254
@martinboon
Martin Boon
7 years
@labourlewis Neither I, nor anyone at @ICMResearch has any knowledge of this poll. Please withdraw our name from it.
16
157
191
@martinboon
Martin Boon
3 years
I’d like to see a campaign for common sense in how to construct balanced batteries of statements.
@CommSenseUK
Campaign For Common Sense
3 years
We found: - culture is to Labour what the NHS was to the Conservatives for so long, its perceived values & priorities are out of step with the public - the Conservatives can confidently take a stand on these issues knowing that they are more in tune with the British public.
Tweet media one
28
80
205
6
12
101
@martinboon
Martin Boon
5 years
If my tired eyes have not betrayed me, this is a truly sensational performance from the pollsters, particularly @benatipsosmori and @AGKD123 @OpiniumResearch Congrats to basically everyone.
Tweet media one
7
37
99
@martinboon
Martin Boon
2 years
With the honorable exception of @OpiniumResearch I think there is almost uniformity in polling methods. If there were a spiral of silence/Shy Tory effect, right now it would be at its peak, and hardly anyone is controlling for it. Sorry to be a broken record.
5
29
87
@martinboon
Martin Boon
2 years
We're thrilled to announce that @LotteHargrave has joined us at @DeltapollUK as Head of Data Science after finishing up her PhD at UCL. Lotte will bring a sophisticated analytical function to our polling work that we've been seeking since @JoeTwyman and I set up four years ago.
Tweet media one
8
4
84
@martinboon
Martin Boon
7 years
So @UKLabour just 3-points shy of their historic 1997 victory and 1-point off their 2001 victory. Forgive me for being sceptical.
9
55
78
@martinboon
Martin Boon
5 years
We suspect @joswinson might be in trouble.
13
23
79
@martinboon
Martin Boon
5 years
Tories 15-points with Opinium.
9
46
58
@martinboon
Martin Boon
2 years
Ok, so a bunch of polls have come out at the same time, and one is an outlier that will grab the headlines and the others are around the 20-point lead mark. It’s a disaster for the Tories whatever way you look at it but do exercise some caution with the hot takes.
6
12
63
@martinboon
Martin Boon
3 years
A single focus group extensively reported in @thetimes is little better than a bunch of vox pops, especially if it is purposively recruited to reflect a specific group. You need 4 groups min to properly report. 8 people sitting round a table means little to nothing.
9
14
58
@martinboon
Martin Boon
5 years
The final @DeltapollUK call for #GE2019 is: Con 45% Lab 35% LD 10% BXP 3% Green 3% Oth 4% Data table at
3
34
59
@martinboon
Martin Boon
5 years
We're not massively confident, but right now we're predicting the Tories GAIN Canterbury by 2-points.
3
19
51
@martinboon
Martin Boon
3 years
Imagine doing a focus group tonight and nobody having heard anything about the Owen Paterson thing.
4
5
52
@martinboon
Martin Boon
4 years
In a Coronavirus age where just about MR company is flailing around trying to jump on the bandwagon, got to say that Ipsos-MORI are knocking it out of the park. Truly tremendous polling, public info, data viz and industry leadership.
4
5
49
@martinboon
Martin Boon
5 years
There was a time, not so long ago, when online panels were chock full of UKIP middle aged males bed blocking quota cells with immediate responses that prevented non-ukip equivalents participating. 1/?
3
33
50
@martinboon
Martin Boon
7 years
No point glossing it over. The turnout model backfired. My call and it was the wrong one.
12
8
47
@martinboon
Martin Boon
2 years
Fascinating thread from Chris but stunned by this. I have never been told by anyone not to release a poll. This goes against every principle of objective and independent polling.
0
20
46
@martinboon
Martin Boon
3 years
@Dannythefink @GavinBarwell A long time ago I was off to do some of *your* focus groups, and I left my bag on the train. It contained £2000 in cash incentives, £1000 of high end recording equipment and a new hot water bottle for my wife. On being found next day, only the hot water bottle was missing.
6
2
48
@martinboon
Martin Boon
5 years
Tonight of all nights, let‘s briefly pay homage to Cleggmania, an almost entirely illusionary but dominant feature of the 2010 campaign.
4
4
42
@martinboon
Martin Boon
6 years
By looks of things, @Ed_Miliband participation in the #SuperBowl is in jeopardy.
@ProFootballTalk
ProFootballTalk
6 years
Illness making its way around the Eagles locker room
180
96
195
0
6
42
@martinboon
Martin Boon
9 months
There is not a reputable pollster in the land who would agree.
@038Degrees
Nasim Chowdhury 💥
9 months
@robfordmancs @Keir_Starmer Nobody said that it was a super-precise gauge of Muslim opinion. If you bothered to do a little bit more than Google you'd see that @SadiqDorasat 's work has been very useful in highlighting Muslim opinion and forming action plans based on it.
3
1
1
0
5
39
@martinboon
Martin Boon
7 years
Feels like the Tory wheels are coming off. @ICMResearch poll might be out of date before it's published.
3
24
36
@martinboon
Martin Boon
5 years
I understand that the @DeltapollUK constituency polls will be released at 7pm. Once placed in public domain we will try and get tables on our website straight away but I'll be in a car at that time so don't @ me.
2
9
35
@martinboon
Martin Boon
5 years
Tonight’s polls will be important. Two polling groups have emerged - one with higher Con leads and one with lower. Tonight should see the ‘higher’ group publish - @YouGov @OpiniumResearch and everyone’s favourite @DeltapollUK
@robfordmancs
Rob Ford
5 years
Cons and Labour both up by almost exactly the same amount since election was called on Oct 31st. But Cons have little BXP vote left to squeeze, and have stalled this week. Lab have advanced further and have more LD/Grn etc vote left to squeeze. Will race tighten further?
4
29
37
2
14
36
@martinboon
Martin Boon
5 years
Big mystery here. Polling companies coalesced around methodological consistency more than ever. And yet the polls are ‘all over the place’. My great fear: it’s all about what convenience sample is used, and we’re blindly trusting something we know less about than ever before.
6
26
35
@martinboon
Martin Boon
5 years
Tell you what - with all this expectation about the YouGov model tonight... I think it will be pretty mundane and I'd be surprised if it doesn't track their regular poll outputs.
7
6
35
@martinboon
Martin Boon
1 year
I helped run focus groups for the Tories before and after ‘97 and the hostility and invective toward them now feels much worse than my recollections of it then.
@robfordmancs
Rob Ford
1 year
This certainly feels more run up to 1997 than 1992. Ahead of 1992, Major's popularity helped drive a rebound for his party. Ahead of 1997, entrenched public hostility to Cons dragged Major's ratings down.
9
26
114
2
10
34
@martinboon
Martin Boon
4 years
If a pollster really told Rod Liddle this, which I doubt, then shame on you for this pathetic spinelessness. #contempt “There are no opinion polls available. I asked one company why this was the case, and the reply was: “Because we were worried we wouldn’t like the results.”
6
9
33
@martinboon
Martin Boon
8 years
Best team to run economy. May/Hammond 53%. Corbyn/McDonnell 15%. @ICMResearch
4
69
31
@martinboon
Martin Boon
5 years
Five constituency polls from @DeltapollUK are now live in The Observer: Esher & Walton, Beaconsfield, Herts SW, Portsmouth South and Berwick Upon Tweed. All are now published on our website here:
6
8
30
@martinboon
Martin Boon
9 years
Oh shit.
6
10
28
@martinboon
Martin Boon
2 years
How to demonstrate you know next to nothing about statistical significance by citing statistical significance.
@saylor
Michael Saylor⚡️
2 years
@elonmusk With 116.6 million followers, your polls are starting to become statistically significant. What if Twitter had an "All Users" poll that you could push to every single twitter account to find out what the entire network is thinking, with no particular adverse selection? 🤔
3K
4K
102K
1
2
29
@martinboon
Martin Boon
7 years
For some horrifying reason, at this moment of all moments I have been drawn back to this from 10.01pm 7th May 2015.
@martinboon
Martin Boon
9 years
Oh shit.
6
10
28
3
7
28
@martinboon
Martin Boon
5 years
The polling industry looks to have had a very, very good night. It's about fractions/chance who is closest.
0
9
28
@martinboon
Martin Boon
5 years
This weekend in The Observer there will be a constituency poll bonanza from @DeltapollUK @PeterKellner1
6
8
28
@martinboon
Martin Boon
5 years
I think the Brexit Party will storm the EU elections, but translating that to 20%+ in Westminster VI mid term is another thing entirely, and caution is all that I think should apply.
2
4
25
@martinboon
Martin Boon
5 years
Wimbledon too close to call and Labour drift into third in Kensington. Tories looking ok in Finchley.
2
19
24
@martinboon
Martin Boon
5 years
Three more constituency polls from @deltapolluk coming this evening in The Observer, scheduled for 7pm release. Full tables will be on our website.
1
14
27
@martinboon
Martin Boon
5 years
In future, I’m not going to bother tweeting about opinion poll methods (but someone really has to), I’m just going to hurl out a couple of focus group truths. Much more fun.
@martinboon
Martin Boon
5 years
I conducted focus groups in the NE just before election, and Labour-to-the-core voters despised Jeremy Corbyn & what he’d done the the party so much they could not vote for it even if grandparents turned in their grave. They didn’t mention Tony Blair much, except positively.
75
571
2K
2
4
24
@martinboon
Martin Boon
2 months
“Gotten”. Shoot me now.
@OpiniumResearch
Opinium
2 months
A majority (54%) of those in the UK think that the economy has gotten worse in the past 12 months. A plurality (35%) are pessimistic about the economy in the next 12 months.
Tweet media one
3
6
16
6
3
26
@martinboon
Martin Boon
7 years
So, if I used my 2015 online method I get most accurate 2017 outcome. Unfortunately 2015 method had EM winning by 4-points = worst. #ffs
5
16
26
@martinboon
Martin Boon
8 years
The effectiveness of of great @ICMResearch #polling . @ianakirby reminds me it was 7 years ago today. I will happily take this to my grave.
Tweet media one
5
12
24
@martinboon
Martin Boon
5 years
Latest @DeltapollUK for the @DavidWooding @TheSun (vs last Sunday): Con 31% (-4) Lab 28% (+4) LD 17% (-1) BXP 13% (-1) Oth 10% (+4) 2049 adults 5-7th Sept.
4
33
21
@martinboon
Martin Boon
7 years
Splashing this on the front page is even braver than @YouGov doing it in the first place.
@DCAJason
Jason Clark
7 years
Tweet media one
0
4
3
12
25
23
@martinboon
Martin Boon
1 year
@jamesjohnson252 @Mutteroo @PeoplePolling This is basic introduction to question wording stuff James. The MRS has an excellent course on it.
1
0
24
@martinboon
Martin Boon
6 years
@rcolvile The precise answer to your precise question Rob is this: at the 95% confidence interval 2000 interviews are correct to within +/-2.2% while 25,000 interviews are correct to within +/-0.6%.
1
5
24
@martinboon
Martin Boon
5 years
I’m one of the people in the industry. And this is bollocks. Shame on Bastani.
@AaronBastani
Aaron Bastani
5 years
Kellner is widely viewed as an unusually wrong pollster by people in polling community. Consistently. This is not a leftist thing @paulmasonnews - ask people in industry!
13
2
10
1
6
23
@martinboon
Martin Boon
5 years
Don’t forget to check in for your final set of @PeterKellner1 @michaelsavage @DeltapollUK Observer constituency polls tonight. Probably released at 7pm but that’s not my call.
3
8
23
@martinboon
Martin Boon
5 years
Gotta be honest. I’m bemused by the polling variation. There’s nowhere near the methodological diffusion of 2017 (I don’t think), which means most pollsters are mostly doing the same things. This implies variation is largely down to panel used to generate the interviews. Scary.
5
15
21
@martinboon
Martin Boon
15 days
I put all the audiences in, both in 2010 and 2015. Easily the most difficult and stressful thing I’ve ever done. But polling companies do not select the questions, they collect them, and that’s different. The broadcasters choose.
@IainDale
Iain Dale
15 days
No. It was puerile, rude and could never have produced good answers. It’s the sort of thing that happens when you contract out the selection of questions to a sodding polling company,
399
293
4K
3
1
23
@martinboon
Martin Boon
2 years
Major report on social class in Britain by @InstituteGC and @PeterKellner1 published this morning, featuring 8,000 interviews from @DeltapollUK This was a joy to be associated with.
0
8
22
@martinboon
Martin Boon
7 years
Check our new marginals xbreak in @guardian tables. Lab doing MUCH BETTER in their <10% marginals than Tories doing in their own.
6
11
21
@martinboon
Martin Boon
7 years
The only day when the word "brisk" is ever used.
0
8
22
@martinboon
Martin Boon
7 years
I guess you can rule us out of any herding now @DPJHodges ?!
3
3
22
@martinboon
Martin Boon
8 years
Full tables for @ICMResearch @Channel4 #muslimPoll (with control and summary topline) can be found here
4
27
20
@martinboon
Martin Boon
5 years
Write up from @PeterKellner1 in The Observer on @DeltapollUK constituency polls in three London seats. #GE2019
2
11
19
@martinboon
Martin Boon
2 years
Whenever I fret about the accuracy of opinion polls, I always turn to economic indicators for reassurance.
@rcolvile
Robert Colvile
2 years
Wow. These were the Bank of England's inflation forecasts as of 13 months ago (via @stevetoy )
Tweet media one
12
18
53
0
1
21
@martinboon
Martin Boon
5 years
Caller: are you doing a poll in constituency x? Me: I'm not saying, you'll have to wait for publication. Caller: I know you are, you polled me. Me: Why are you calling me then? Caller: You're obliged to tell me everything under BPC rules. Me: Sigh. Yes, when it's PUBLISHED.
0
1
21
@martinboon
Martin Boon
2 years
Just imagine being in the middle of a focus group on the economy when news of a certain resignation reaches you. There goes the discussion guide.....
1
1
20
@martinboon
Martin Boon
10 months
Completely not buying any part of this.
@patrickkmaguire
Patrick Maguire
10 months
Jeremy Corbyn would deny Sadiq Khan a third term if he runs for London mayor next year, new polling suggests Without Corbyn, Khan beats Hall 33 per cent to 32 per cent When included, JC polls 15 per cent - Khan 25 per cent, Hall 30 per cent Hear more on @TimesRadio shortly
Tweet media one
353
159
454
2
2
21
@martinboon
Martin Boon
4 years
Well done @MattSingh_ If true, any survey looking at Brexit numbers not controlling for educ attainment about as far from ‘peeerless’ as it gets. I mean, it’s not as if that wasn’t the biggest polling lesson from 2016.
@MattSingh_
Matt Singh
4 years
I may be doing something wrong here... But it looks has though this sample has rather a lot of graduates. If we look at the working age (16-64) population (to compare with the APS), it looks like 50.6% have level 4 qualifications or higher. In the actual 2018 APS it was 39.2%
Tweet media one
2
1
21
1
3
21
@martinboon
Martin Boon
5 years
Recently a brand new client wanted to ask a question which I refused to do, on the basis it was leading, emotionally driven and one of those questions we include in our polling joke question compilations. Unsurprised to see a competitor went and asked it.
4
1
21
@martinboon
Martin Boon
5 years
Quite brilliant from @StephenDFisher and Dan Snow. I have been banging on for a while about panel used being much more likely to explain pollster outcomes this time around, so great to have some evidence on the subject Also this is a worry for me.
1
12
20
@martinboon
Martin Boon
3 years
I think we should poll what % of the population think we can grow concrete.
4
1
19
@martinboon
Martin Boon
5 years
I'm not sure I was expecting this, but I'm really pleased with how @DeltapollUK phone polls are matching up with MrP models. At times like these, don't underestimate how little crumbs of comfort give huge adrenaline bursts to worried pollsters.
6
0
18
@martinboon
Martin Boon
8 years
Everyone needs to calm down a bit. Please leave me alone so I can get the data to @guardian_clark
7
10
18
@martinboon
Martin Boon
7 years
@ICMResearch @guardian tables has revised break for marginal seats. @Conservatives lead 48% vs 31% in Lab held seats <15% majority in E&W.
4
27
17
@martinboon
Martin Boon
7 years
Spot on, this.
@CJTerry
Chris Terry-Enescu
7 years
The Exit Poll: Chucking large amounts of money at a bunch of glorious nerds to find out the result 6 hours early. It is the best thing.
1
14
44
0
5
19
@martinboon
Martin Boon
5 years
Thrilled that so far, the @DeltapollUK @PeterKellner1 @michaelsavage constituency polls appear to be really good. Only Kensington caused us trouble. Maybe not the BS that one news anchor suggested.
3
2
18
@martinboon
Martin Boon
7 years
This is fundamentally, completely, and without ambiguity, wrong.
@DPJHodges
(((Dan Hodges)))
7 years
@David_K_Clark @stephenkb No, it's not about competitive advantage at this stage, as we saw last time, it's about safety in numbers.
4
0
1
1
2
18
@martinboon
Martin Boon
8 years
#brexit polls were "wrong" is now universal narrative. During campaign there were more published polls saying Leave than saying Remain IIRC.
5
13
16
@martinboon
Martin Boon
5 years
"Look at the shares, not the lead". Sir Bob Worcester, about a zillion times. And he was right you know.
1
5
18
@martinboon
Martin Boon
5 years
Good opinion research leaves little to no room for respondent or indeed analytical ambiguity.
@robfordmancs
Rob Ford
5 years
@MarinaHyde Bit concerned the question may not quite show this. The wording may encourage people to see agreement as “I am really committed to my preferred Brexit outcome” rather than “I support political violence”
9
11
85
2
2
18
@martinboon
Martin Boon
5 years
@DeltapollUK @PeterKellner1 @michaelsavage Can I take this back? We got Kensington right, and now Finchley too. Stunning.
3
4
17
@martinboon
Martin Boon
5 years
We are now into the phase where pundits chime in: ‘I hate the polls but as they now reinforce my own anecdotes/feel on the ground’ I’m ok to feed them in‘ (but if they’re wrong was I not clear I hated them all along)?
0
4
16
@martinboon
Martin Boon
9 years
Let's stop with the anti @LordAshcroft stuff. His techniques same as @ICMResearch +/-, & right not to name candidates until last minute.
5
12
15
@martinboon
Martin Boon
1 year
Wisdom.
@hugorifkind
Hugo Rifkind
1 year
@stephenkb North London is North of Euston Road. South London is South of the river. I will not stand for this notion that Central London is in North London. That way madness lies.
13
0
49
1
2
16
@martinboon
Martin Boon
4 years
Twitter won't like this, but: it can be simultaneously true that there is uproar over Dom Cummings while focus groups participants sit back & wonder who he is & what the fuss is about. Polls may tank but groups can be a very different animal. Some people just don't give a shit.
5
7
16
@martinboon
Martin Boon
4 years
If a junior researcher put together a slide deck for the nation this bad, they would be ashamed (and probably fired without significant and fast paced signs of improvement).
0
2
16
@martinboon
Martin Boon
5 years
We're estimating a 14-point Con win.
2
4
16
@martinboon
Martin Boon
5 years
One for @EnglandInEurope : The UK government should encourage a United Ireland as a solution to the Irish Border? Yes 38% No 25% DK 27% The RoI should leave the EU as a solution to the Irish Border? Yes 24% No 40% DK 36%
4
3
14
@martinboon
Martin Boon
5 years
My timeline: give me polls, give me polls, now! More! Also my timeline: yeah but polls are a bit rubbish, innit.
3
1
16
@martinboon
Martin Boon
4 years
Completely my experience of him too. At the 1st ever TV leaders debate, he immediately bought me and my team a round of drinks as soon as he found out who we were. A true professional.
@ProfJaneGreen
Jane Green
4 years
Alastair Stewart is a kind, generous gent, and takes huge pride in getting it right. He has been very kind to me, and I'm so upset to see his ITV News career ending this way. Surely we can all make mistakes - unintended - apologise, and crack on...
5
7
60
0
2
16
@martinboon
Martin Boon
5 years
Do you think Brexit will ever end? @DeltapollUK 7-9sept, 2049 adults. Yes 41% No 35% DK 24%
@IsabelOakeshott
Isabel Oakeshott
5 years
You clearly don’t get out enough @labourleave @brexitparty_uk @Nigel_Farage
134
257
2K
1
9
15
@martinboon
Martin Boon
5 years
No doubt there will be a slew of polls tonight,and @DeltapollUK will be front and centre. Look out from us for national polling and the first of a big slate of constituency polls.....
2
6
14
@martinboon
Martin Boon
4 years
Sadness.
Tweet media one
3
3
15
@martinboon
Martin Boon
2 months
Most of the @DeltapollUK office will have to cancel holidays. If ever I were tempted to fix a poll just out of spite, this would be the time.
@gabyhinsliff
gabyhinsliff
2 months
Calling a snap summer election shortly after the lobby & most of your own MPs have booked summer hols on the understanding there wouldn't be a summer election would make for such a furiously bad-tempered campaign launch I'm almost curious to see it
58
185
2K
2
4
15
@martinboon
Martin Boon
5 years
Humbly, few people know more about audiences and questions selected for live leaders’ debates than me, as I was involved right from the off before 2010 election. Everyone should back off the BBC, because it’s miles harder than you think.
4
2
14
@martinboon
Martin Boon
8 years
I want to be clear: in my view this is the most rigorous Muslim survey I've seen in the public domain, incl prior @ICMResearch surveys.
5
10
14
@martinboon
Martin Boon
5 years
BUT, just remember that overstating of a anti-European right of centre party in poll after poll is not unheard of, nor misunderstood as to why it might be the case.
1
4
14
@martinboon
Martin Boon
5 years
Busting this bubble a little. Past vote recall numbers for Labour are consistently lower than their actual share in 2017 i.e people are embarrassed to say they are voting Labour. Historically, this implies polls are underscoring Labour, if this phenom exists today.
@rcolvile
Robert Colvile
5 years
I haven’t run this past any pollsters, and it’s no more than a hunch. But I have a feeling Shy Tories may be a thing again this election. Been told so consistently that Tories are evil that they don’t want to admit it, but in the silence of the voting both...
86
88
587
3
9
13
@martinboon
Martin Boon
5 years
Maybe Scotland should consider suing RWC for allowing this game to go ahead....
0
7
14
@martinboon
Martin Boon
5 years
Anthony has carved out a really valuable niche here on Twitter, to the benefit of all of us in the research industry. I for one really appreciate it. Thanks Anthony!
@anthonybmasters
Anthony B. Masters
5 years
It has been claimed a recent poll implied there was a 10 point Conservative lead in 2017. This criticism is ill-founded: Opinium use the earliest available recall in their weightings. The article discusses false recall. #GE2019 #GeneralElection2019
5
12
17
1
1
14