If you are not itiboribo then you must be iti. You are yet to draw your own conclusions but you quarrel with my critique, not what you repeatedly misidentified as critic or person. We wait to read your own analysis line by line.
You are right that correlation is an inferential statistic that allows us to make an educated guess of the sort, if x then y. Now, the problem with The Guardian and their statistical consultants like you is that they are making inferences based on one variable, the registration of voters. Where is the other variable? If registration is the dependent variable that they wanted to explain, where is the independent variable in their correlation? What do you mean by descriptive statistics in my comment?
Spurious computation is not factual but you are entitled to your own opinion about that. Perhaps what they wanted to explain is better done with time series or analysis of variance and not with the correlation of x with x. Garbage in garbage out. I sabi small you said.
I have also taught statistics at undergraduate and graduate levels but you beat me with consultancy and professional development. Carry go. Just do not forget that there is a con in consultant.
Biko
-- On Mon, Feb 18, 2019 at 7:41 PM, Okechukwu Ukaga<ukaga001@umn.edu> wrote:Biko, my broda:You are doubling down on your misunderstanding and misinterpretation.And I am no itiboribo in statical analysis. Notably, I have not only taught this subject at the graduate, undergraduate and professional development levels, but have also consulted on it and written evaluation book much of which is on statistical data analysis.My sense is you know a little, enough to cause trouble, but perhaps not enough to fully understand and critic the given report. For starters you are conflating percentage (descriptive statistic) with correlation (an inferential statistic). As a famous America politician used to say, you are entitled to your own opinion, but not your own facts.I take no position as to the validity or reliability of the report because I have not given it enough consideration to reach any conclusion. But if you want to critic something, you must atleast get the basics right. You failed to do that in my opinion and that is what I wanted to call to your attention.Regards,Okey--On Feb 18, 2019 5:36 PM, "'Biko Agozino' via USA Africa Dialogue Series" <usaafricadialogue@googlegroups.com> wrote:Okey bros, you sound like itiboribo in statistical analysis. There is always room for error in statistics especially due to sampling errors.Just because something is reported by The Guardian, it does not follow that Bekee is Agbara and a native informant like me is only misunderstanding or misinterpreting or misreading the gospel truth.--Yes, there is always fear of mass rigging but not because of equality in increases in the registration of voters. Mass rigging has also been suspected in US elections and in the elections organized in Africa by colonizers in spite of the number of regustered voters.The task is for civil society organizations to mount voter registration campaings all year round and to force election officials to allow citizens to vote with drivers licence or other forms of govt issued ids rather than insist on the blind faith that the contractors with the monopoly over limited supplies of pvc should dictate who has the right to vote.A major factor in rigging is the low turnout of registered voters. This allows election officials to add a zero to the end of the tally of the cash and carry candidate.If one billion naira is allocated to each state to be awarded to 1000 voters at one million each if their voting number is selected by lottery after the elections, then we may get 0.99 correlation between registered voters and voter turnout. But anytime the r score is that high, statisticians caution us to check again for computational errors.Biko
On Mon, Feb 18, 2019 at 12:08 PM, Okechukwu Ukaga<ukaga001@umn.edu> wrote:--Biko,I think you are misreading/misunderstanding/ misinterpreting the report, as it talked about the number of registered voters increasing by about the same percentage in every state and not about correlation between voter registration and voter registration as you suggested. Here is what they reported:
The number of new voters registered in Nigeria since January 2018 has increased by almost exactly the same percentage in each of its states, according to documents seen and analysed by the Guardian, raising fears that the results of Saturday's presidential election could be open to mass rigging.On Feb 18, 2019 8:53 AM, "'Biko Agozino' via USA Africa Dialogue Series" <usaafricadialogue@ googlegroups.com> wrote:Thanks to CAO for sharing this. It sounds like fake news because the journalists reported that the source of their statistics vanished from the website. The snake seen by one person easily turns into a dancing python. The Guardian reporters also appear ignorant about statistical analysis and the fact that people do lie with statistics.Having taught statistical analysis a little in the past, I can say for sure that it is not only God who can produce a correlation of 0.9 or perfect correlation. Such an r score is often an indication of error in the computation. The statistical analysts that the reporters consulted probably did not know this.Simply put, you need two mutually exclusive variables to compute a valid correlation score. However, if you correlate one variable with the same variable, you will get 1:1 correlation or nearly so but that is junk statistics. The analysts were correlating the same variable, voter registration, with the same variable, voter registration. What did they expect to find? Mechew.On the error regarding the number of votes cast in 2015, it appears that someone entered the wrong total and this was deleted when the error was found out. Everything in Africa is not magic, magic, contrary to the expectations of Oyibo oracles.Statistical literacy should be made compulsory in African schools.Biko
On Sun, Feb 17, 2019 at 4:11 PM, Emeagwali, Gloria (History)<emeagwali@ccsu.edu> wrote:I wonder what Prof. Aluko thinks about the statistical analysis.
GE
From: 'Biko Agozino' via USA Africa Dialogue Series <usaafricadialogue@ googlegroups.com>
Sent: Sunday, February 17, 2019 9:15:10 AM
To: usaafricadialogue@ googlegroups.com
Subject: Re: USA Africa Dialogue Series - UK's Guardian Newspaper Indicts Nigeria's Electoral CommissionAny independent analysis of this in the Naija press?
Biko
On Sun, Feb 17, 2019 at 5:48 AM, Chidi Anthony Opara, FIIM<chidi.opara@gmail.com> wrote:Courtesy: Guardian (UK)
The number of new voters registered in Nigeria since January 2018 has increased by almost exactly the same percentage in each of its states, according to documents seen and analysed by the Guardian, raising fears that the results of Saturday's presidential election could be open to mass rigging.
Voters in Africa's biggest country by population will choose between the incumbent Muhammadu Buhari, his main rival Atiku Abubakar, and more than 70 other candidates.
Since the last presidential election in 2015, many more people have become eligible to vote, and many others have registered to take part in the polls for the first time. About 10 million new voters signed up between January 2018 and early 2019 – according to data released by the electoral commission (INEC) – twice the number that signed up in the first nine months of registration, between April 2017 and January 2018.
But analysis of the data for each of the country's 36 states and its capital shows that INEC has increased the number of new registered voters by almost exactly the same percentage across all states. The correlation is a "statistical impossibility" and does not reflect Nigeria's demographic changes, according to data analysts working with the Guardian. Additional data seen by the Guardian also shows irregularities in registration for the 2015 election, until now considered to have been free and fair.
On average, voter registration in each state increased by 2.2% between April 2017 and January 2018, and by 7.7% for the whole registration period ahead of Saturday's election.
Plotted on a scatter line graph, there is a 0.99 correlation across all the states, without a single outlier. According to three separate data analysts, the parity cannot be a coincidence. "Only God works that closely," one analyst said. If some of the new voters registered are fake it would imply meddling at the electoral commission, though it is unclear whether it would be the ruling party or the opposition that would stand to benefit.
Saturday's election is seen as a referendum on Buhari's first term, which has been marred by his prolonged absence due to illness, a weak economy, and the government's failure to effectively tackle corruption and insecurity.
A faction of Boko Haram attacked a state governor's convoy on Tuesday, killing four people and stealing vehicles; on the same day, 15 people were crushed to death at a ruling party rally in eastern Port Harcourt. On Thursday, 14 sacks of ballot papers were intercepted in Kano state – though police said they were merely "specimen" papers to educate voters.
There have also been reports that the privacy of citizens may have been compromised after INEC and the Nigerian communications commission allegedly allowed the ruling party to access personal data.
The 2015 election in which Buhari came to power was widely held to be free and fair. However, an analysis of separate figures shows that manipulation may have happened in favour of Buhari's party, which was running in opposition to Goodluck Jonathan's People's Democratic party.
A clue may have been dropped last July when the INEC, perhaps inadvertently, publicly referenced a different set of results to the one on which Buhari's victory was based.
Both documents showed 29.4 million votes were cast. But according to the original results, 31.7 million accredited voters participated in the election, whereas in the second set of results that figure dropped to 23.6 million.
The discrepancy suggests an additional 6 million accredited voters, far more than the APC's winning margin – as per the original result set – of 2.6 million votes.
Smart-card readers were used for the first time in 2015 and the second set of results was released in response to widespread criticism after the new technology malfunctioned, forcing millions of voters including Jonathan to use the manual process. The second result set appears to have disappeared from INEC's website two months ago, along with all others relating to the 2015 poll.
There is frequently rigging in Nigerian elections, and it is not usually limited to whichever side happens to be in power at the time. Powerful politicians move fluidly between the two main parties, taking support and rigging mechanisms with them.
Buhari's party chief may have revealed the truth with a slip of the tongue at a press conference in September.
"For democracy to flourish, only people who can accept the pain of rigging – sorry, defeat – should participate in an election," said Adams Oshiomhole.
--
Chidi Anthony Opara is a "Life Time Achievement" Awardee, Registered Freight Forwarder, Professional Fellow Of Institute Of Information Managerment, Africa, Poet and Publisher of PublicInformationProjects
--
Listserv moderated by Toyin Falola, University of Texas at Austin
To post to this group, send an email to USAAfricaDialogue@ googlegroups.com
To subscribe to this group, send an email to USAAfricaDialogue+subscribe@ googlegroups.com
Current archives at http://groups.google.com/ group/USAAfricaDialogue
Early archives at http://www.utexas.edu/ conferences/africa/ads/index. html
---
You received this message because you are subscribed to the Google Groups "USA Africa Dialogue Series" group.
To unsubscribe from this group and stop receiving emails from it, send an email to usaafricadialogue+unsubscribe@ googlegroups.com.
For more options, visit https://groups.google.com/d/ optout.
--
Listserv moderated by Toyin Falola, University of Texas at Austin
To post to this group, send an email to USAAfricaDialogue@ googlegroups.com
To subscribe to this group, send an email to USAAfricaDialogue+subscribe@ googlegroups.com
Current archives at http://groups.google.com/ group/USAAfricaDialogue
Early archives at http://www.utexas.edu/ conferences/africa/ads/index. html
---
You received this message because you are subscribed to the Google Groups "USA Africa Dialogue Series" group.
To unsubscribe from this group and stop receiving emails from it, send an email to usaafricadialogue+unsubscribe@ googlegroups.com.
For more options, visit https://groups.google.com/d/ optout.
--
Listserv moderated by Toyin Falola, University of Texas at Austin
To post to this group, send an email to USAAfricaDialogue@ googlegroups.com
To subscribe to this group, send an email to USAAfricaDialogue+subscribe@ googlegroups.com
Current archives at http://groups.google.com/ group/USAAfricaDialogue
Early archives at http://www.utexas.edu/ conferences/africa/ads/index. html
---
You received this message because you are subscribed to the Google Groups "USA Africa Dialogue Series" group.
To unsubscribe from this group and stop receiving emails from it, send an email to usaafricadialogue+unsubscribe@ googlegroups.com.
For more options, visit https://groups.google.com/d/ optout.--
Listserv moderated by Toyin Falola, University of Texas at Austin
To post to this group, send an email to USAAfricaDialogue@ googlegroups.com
To subscribe to this group, send an email to USAAfricaDialogue+subscribe@ googlegroups.com
Current archives at http://groups.google.com/ group/USAAfricaDialogue
Early archives at http://www.utexas.edu/ conferences/africa/ads/index. html
---
You received this message because you are subscribed to the Google Groups "USA Africa Dialogue Series" group.
To unsubscribe from this group and stop receiving emails from it, send an email to usaafricadialogue+unsubscribe@ googlegroups.com.
For more options, visit https://groups.google.com/d/ optout.
Listserv moderated by Toyin Falola, University of Texas at Austin
To post to this group, send an email to USAAfricaDialogue@ googlegroups.com
To subscribe to this group, send an email to USAAfricaDialogue+subscribe@ googlegroups.com
Current archives at http://groups.google.com/ group/USAAfricaDialogue
Early archives at http://www.utexas.edu/ conferences/africa/ads/index. html
---
You received this message because you are subscribed to the Google Groups "USA Africa Dialogue Series" group.
To unsubscribe from this group and stop receiving emails from it, send an email to usaafricadialogue+unsubscribe@ googlegroups.com.
For more options, visit https://groups.google.com/d/ optout.
Listserv moderated by Toyin Falola, University of Texas at Austin
To post to this group, send an email to USAAfricaDialogue@ googlegroups.com
To subscribe to this group, send an email to USAAfricaDialogue+subscribe@ googlegroups.com
Current archives at http://groups.google.com/ group/USAAfricaDialogue
Early archives at http://www.utexas.edu/ conferences/africa/ads/index. html
---
You received this message because you are subscribed to the Google Groups "USA Africa Dialogue Series" group.
To unsubscribe from this group and stop receiving emails from it, send an email to usaafricadialogue+unsubscribe@ googlegroups.com.
For more options, visit https://groups.google.com/d/ optout.
Listserv moderated by Toyin Falola, University of Texas at Austin
To post to this group, send an email to USAAfricaDialogue@googlegroups.com
To subscribe to this group, send an email to USAAfricaDialogue+subscribe@googlegroups.com
Current archives at http://groups.google.com/group/USAAfricaDialogue
Early archives at http://www.utexas.edu/conferences/africa/ads/index.html
---
You received this message because you are subscribed to the Google Groups "USA Africa Dialogue Series" group.
To unsubscribe from this group and stop receiving emails from it, send an email to usaafricadialogue+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Listserv moderated by Toyin Falola, University of Texas at Austin
To post to this group, send an email to USAAfricaDialogue@googlegroups.com
To subscribe to this group, send an email to USAAfricaDialogue+subscribe@googlegroups.com
Current archives at http://groups.google.com/group/USAAfricaDialogue
Early archives at http://www.utexas.edu/conferences/africa/ads/index.html
---
You received this message because you are subscribed to the Google Groups "USA Africa Dialogue Series" group.
To unsubscribe from this group and stop receiving emails from it, send an email to usaafricadialogue+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
No comments:
Post a Comment