Author Topic: Of Ambithology & Tomfoolery  (Read 2781 times)

Offline vooke

  • Moderator
  • Enigma
  • *
  • Posts: 5985
  • Reputation: 8906
Of Ambithology & Tomfoolery
« on: July 24, 2017, 10:47:44 AM »
Yesterday they pulled a stunt; showed up on both NTV and later Citizen. On NTV, it was just them and Larry while on Citizen they were joined by NASWA sympathizer Herman Manyora and Jubilee leaning Manwa Magoma. Manwa made his name n 2013 as the 'numbers guy' for his analysis and predictions. He reminds me of Pundit.

Start from 15:40

As usual the magicians defend their craft; Wolf sampled adults while Angela sampled registered voters who have every intent of voting. Wolf reckons there is an element of dishonesty as his sample turned out 90% registered voters instead of 75% ( we have 25M adults and only 19.6M registered voters). This could explain the differences in their polling not to mention timing; wolf did his at the beginning of the month while Angela a week later.

Funny that Wolf's defense of polling is the 2005 referendum. 2013 is carefully avoided.

The question of whether the country is heading in the right direction keeps coming up. Both pollsters got very different results. None could satisfactorily spin this. Angela was generous to Uhunye. I think the question is too complex and it should mirror the candidate popularities. 47% Approving of Uhuru while only 27% believe we are on course doesn't make sense. That means what, 20% don't like Uhuru but they'd rather survive under him as opposed to Baba?




Citizen is where Wolf and Angela's brains were picked. It was a shouting match


Turns out Angela polled 31 counties while Wolf did 47.
All claim it's very expensive increasing the sample size.
Manwa insists including or omitting some counties can turn the results on their head. I agree

Manyora thinks polls are shiet. How on earth does Uhunye score 29% in NBO?
Three reasons;
1. Polls may be cooked
2. Polls ignore Tribe which is just about all our elections are.
3. Our literacy levels may mean subjects misunderstand the poll questions

Manwa concluded that the two magicians can't agree on their tricks.
2 Timothy 2:4  No man that warreth entangleth himself with the affairs of this life; that he may please him who hath chosen him to be a soldier.

Offline RV Pundit

  • Moderator
  • Enigma
  • *
  • Posts: 37009
  • Reputation: 1074446
Re: Of Ambithology & Tomfoolery
« Reply #1 on: July 24, 2017, 10:53:56 AM »
I watched the NTV - I think both did fairly well considering Larry Madowo doesn't understand the subject matter.

Both polls look similar in most places except RV & Nairobi. I believe that without factoring tribe (the overriding factor) in their sampling strategy they are always going to get wrong Nairobi & RV.

The tribal data is available in 2009 Census. There is a reason why we do stratified sampling - because unless you're randomizing function is really good - you are better off starting with the knowns.

Angela Ambitho polls are of course cooked - you can see his history and she is always pro-Raila. I believe Tom polls but disagree with some of their figures in comspolitan counties.

Offline vooke

  • Moderator
  • Enigma
  • *
  • Posts: 5985
  • Reputation: 8906
Re: Of Ambithology & Tomfoolery
« Reply #2 on: July 24, 2017, 11:08:43 AM »
I watched the NTV - I think both did fairly well considering Larry Madowo doesn't understand the subject matter.

Both polls look similarly in most places except RV & Nairobi. I believe that without factoring tribe (the overriding factor) in their sampling strategy they are always going to get wrong Nairobi & RV.

The tribal data is available in 2009 Census. There is a reason why we do stratified sampling - because unless you're randomizing function is really good - you are better off starting with the knowns.

Angela Ambitho polls are of course cooked - you can see his history and she is always pro-Raila. I believe Tom polls but disagree with some of their figures in comspolitan counties.
I can't get a hold of Angela's latest poll but she mentioned she did 31 counties and not all counties in the Rift were covered.

Manwa' contention was 2013 patterns remain and omitting one where a candidate fared well would be disastrous. I want to see which she missed
2 Timothy 2:4  No man that warreth entangleth himself with the affairs of this life; that he may please him who hath chosen him to be a soldier.

Offline RV Pundit

  • Moderator
  • Enigma
  • *
  • Posts: 37009
  • Reputation: 1074446
Re: Of Ambithology & Tomfoolery
« Reply #3 on: July 24, 2017, 11:25:48 AM »
How would that be representative in a prov with 14 counties with different dynamics. That is why she is getting it wrong..deliberately of course. Tom polls number for Nairobi are obviously wrong.
I can't get a hold of Angela's latest poll but she mentioned she did 31 counties and not all counties in the Rift were covered.

Manwa' contention was 2013 patterns remain and omitting one where a candidate fared well would be disastrous. I want to see which she missed

Offline vooke

  • Moderator
  • Enigma
  • *
  • Posts: 5985
  • Reputation: 8906
Re: Of Ambithology & Tomfoolery
« Reply #4 on: July 24, 2017, 01:13:46 PM »
Both magicians have another poll up their sleeve before August 1st.

They should go really deep and wide. Do at least 100 per county. If they religiously stick to 2K, it will be utter rubbish in nonhomogeneous regions, which is where all eyes are on
2 Timothy 2:4  No man that warreth entangleth himself with the affairs of this life; that he may please him who hath chosen him to be a soldier.

Offline RV Pundit

  • Moderator
  • Enigma
  • *
  • Posts: 37009
  • Reputation: 1074446
Re: Of Ambithology & Tomfoolery
« Reply #5 on: July 24, 2017, 01:39:35 PM »
Let see. It's a week to 1st August. I don't know how Tom's Ipsos will do household survey in that short period of time - do all the back-checks -but Ambitho with her telephone interview should be good to go - of course I doubt she did any back-checks in her bid to release the poll on the same day as Ipsos.
Both magicians have another poll up their sleeve before August 1st.

They should go really deep and wide. Do at least 100 per county. If they religiously stick to 2K, it will be utter rubbish in nonhomogeneous regions, which is where all eyes are on

Offline gout

  • VIP
  • Enigma
  • *
  • Posts: 3764
  • Reputation: 1374
Re: Of Ambithology & Tomfoolery
« Reply #6 on: July 24, 2017, 04:24:54 PM »
Manyora - a lecturer- being dismissive of opinion polls is unfathomable. A lecturer who thinks basic research is shiet?? Utter nonsense!
I underestimated the heartbreaks visited by hasla revolution

Offline vooke

  • Moderator
  • Enigma
  • *
  • Posts: 5985
  • Reputation: 8906
Re: Of Ambithology & Tomfoolery
« Reply #7 on: July 24, 2017, 05:05:05 PM »
Manyora - a lecturer- being dismissive of opinion polls is unfathomable. A lecturer who thinks basic research is shiet?? Utter nonsense!

He is all for research, just that political polls are so heavily tainted by tribal bias that they don't work except in homogeneous groups/regions
2 Timothy 2:4  No man that warreth entangleth himself with the affairs of this life; that he may please him who hath chosen him to be a soldier.

Offline Kim Jong-Un's Pajama Pants

  • Moderator
  • Enigma
  • *
  • Posts: 8728
  • Reputation: 106254
  • An oryctolagus cuniculus is feeding on my couch
Re: Of Ambithology & Tomfoolery
« Reply #8 on: July 24, 2017, 05:39:18 PM »
Just by reading this thread, it's not difficult to guess what the polls say.  Even without looking at them.  It would be interesting if they shared some of their sampling strategies.
"I freed a thousand slaves.  I could have freed a thousand more if only they knew they were slaves."

Harriet Tubman

Offline gout

  • VIP
  • Enigma
  • *
  • Posts: 3764
  • Reputation: 1374
Re: Of Ambithology & Tomfoolery
« Reply #9 on: July 24, 2017, 07:43:13 PM »
Aihhh??? Science/research is all about taking into consideration all sort of factors - gender, sex, poverty, race, tribe, dialect, name it.

Manyora - a lecturer- being dismissive of opinion polls is unfathomable. A lecturer who thinks basic research is shiet?? Utter nonsense!

He is all for research, just that political polls are so heavily tainted by tribal bias that they don't work except in homogeneous groups/regions
I underestimated the heartbreaks visited by hasla revolution

Offline RV Pundit

  • Moderator
  • Enigma
  • *
  • Posts: 37009
  • Reputation: 1074446
Re: Of Ambithology & Tomfoolery
« Reply #10 on: July 27, 2017, 11:15:30 AM »
This is why she is getting RV wrong...the sampling is skewed to NASA leaning counties...

Offline vooke

  • Moderator
  • Enigma
  • *
  • Posts: 5985
  • Reputation: 8906
Re: Of Ambithology & Tomfoolery
« Reply #11 on: August 01, 2017, 08:03:59 PM »
Something I had missed. Look carefully at the Battleground counties.

Uasin Gishu where Babu had 21% vs Uhunye's 74% in 2013 is now a battleground
2 Timothy 2:4  No man that warreth entangleth himself with the affairs of this life; that he may please him who hath chosen him to be a soldier.

Offline RV Pundit

  • Moderator
  • Enigma
  • *
  • Posts: 37009
  • Reputation: 1074446
Re: Of Ambithology & Tomfoolery
« Reply #12 on: August 01, 2017, 08:07:06 PM »
Uasin gishu wsr home ground is only battleground in cooked stat.

Offline vooke

  • Moderator
  • Enigma
  • *
  • Posts: 5985
  • Reputation: 8906
Re: Of Ambithology & Tomfoolery
« Reply #13 on: August 01, 2017, 11:58:53 PM »
Something I had missed. Look carefully at the Battleground counties.

Uasin Gishu where Babu had 21% vs Uhunye's 74% in 2013 is now a battleground

Curious enough, Bomet is a Jubilee stronghold. If correct,what exactly has Rutto added to NASWA other than a nationalistic semblance? Where's the Rift in the Rift?
2 Timothy 2:4  No man that warreth entangleth himself with the affairs of this life; that he may please him who hath chosen him to be a soldier.

Offline Kichwa

  • Moderator
  • Enigma
  • *
  • Posts: 2886
  • Reputation: 2697
Re: Of Ambithology & Tomfoolery
« Reply #14 on: August 02, 2017, 01:08:57 AM »
One thing for sure, NASA has the big MO. Towards the end of a marathon, this is the most reliable indicator of the winner. Pundits MOAS assumes the race is static from the get go. I am liking NASA's chances more and more.
"I have done my job and I will not change anything dead or a live" Malonza

Offline vooke

  • Moderator
  • Enigma
  • *
  • Posts: 5985
  • Reputation: 8906
Re: Of Ambithology & Tomfoolery
« Reply #15 on: August 02, 2017, 06:44:02 AM »
One thing for sure, NASA has the big MO. Towards the end of a marathon, this is the most reliable indicator of the winner. Pundits MOAS assumes the race is static from the get go. I am liking NASA's chances more and more.

You don't find it curious that Bomet is reported as a Jubilee stronghold while Using Gishu is a battlefront?

If Rutto is or was ever going to Rift the Rift, starting point would be his home turf.


Come on, it may tickle your biases but call it what it is; bullshiet
2 Timothy 2:4  No man that warreth entangleth himself with the affairs of this life; that he may please him who hath chosen him to be a soldier.

Offline vooke

  • Moderator
  • Enigma
  • *
  • Posts: 5985
  • Reputation: 8906
Re: Of Ambithology & Tomfoolery
« Reply #16 on: August 09, 2017, 09:26:41 AM »
Ambitho made tons of cash from NASA peddling their fantasies, she must be an Okoyu at heart
2 Timothy 2:4  No man that warreth entangleth himself with the affairs of this life; that he may please him who hath chosen him to be a soldier.

Offline einstein_g

  • Superstar
  • *
  • Posts: 114
  • Reputation: 68
Re: Of Ambithology & Tomfoolery
« Reply #17 on: August 09, 2017, 09:32:40 AM »
Honestly it's not ok for these opinion pollsters to keep peddling bullshit. It's a complete lack of ethics.