Those days, in FUTO, JAMBITES were warned of a mystical course codenamed “O.D.E” and trumpeted in Igbo as “O di egwu” [something mystical, dangerous and challenging]. That course was actually Ordinary Differential Equations (ODE).
Interestingly, it was the easiest course in the university because two smart men taught it: Dr Oyet and Dr Effiong. “Recall from equation 10 before the strike”…Effiong would begin. Effiong had no course notes and could remember the last thing he wrote on a blackboard four months later. He was simply amazing!
As you try to deal with these men, another great man – Dr. Ngochindo – who taught Organic Chemistry was memorable. His style was unusual: the bonds are there and you must see them as he backed the class talking to the blackboard!
Tekedia Mini-MBA edition 14 (June 3 – Sept 2, 2024) begins registrations; get massive discounts with early registration here.
Tekedia AI in Business Masterclass opens registrations here.
Join Tekedia Capital Syndicate and invest in Africa’s finest startups here.
These men are needed for a major service in Nigeria on a new poll: “A new poll conducted by a public advocacy group We2Geda Foundation said the presidential candidate of the Labour Party (LP), Peter Obi, is ahead of other candidates by 51 percent. The poll result was dated September 17, 2022, and signed by the Chairman, Board of Trustees of We2Geda, Ibrahim Abdulkareem, as well as a member of the BoT, Muna Obioha.”
The group said 51 percent of respondents suggested voting for Obi as their preferred candidate, 25 percent suggested voting for Atiku Abubakar of the Peoples Democratic Party (PDP) while Bola Tinubu of the All Progressives Congress (APC) ranked third in the voter preference poll with 19 percent.
“Further analysis also shows that Peter Obi remained a consistent favourite in the four major geopolitical zones including North-Central, South-South, South-West and South-East, while Alhaji Atiku Abubakar was the poll leader in the NorthEast and North-West zones,” the group said.
We2Geda explained that Nigerians between 26 and 50 years constituted the highest number of respondents in a random survey with a total of 15,438 registered voters in 36 states and the Federal Capital Territory.
Can someone solve the ODE of Nigeria and connect the bonds via these polls? It is #open
Comment on LinkedIn Feed
Comment : Its not about the chart or variance. For this poll to be considered at all, the volume of respondents is needed and if I may ask, how did you go about checking for uniqueness of responses?
My Response: The questions you are asking are things they provided in their report. Of course, I cannot add those in a LinkedIn post!
Comment 1b: 15,438 registered voters in 36 states too insignificant to matter for a country of over 200million. I’ve just read an extract of their report. Thanks and no thanks
My response: You need to revisit your statistics class notebook. They oversampled. 15,500 for 200 million in a scientific poll is oversampling! Typically, you need about 800-2,000. It is scientific but it is voodoo for those who do not understand the science of polling. In US, the average size is 1,000 and they use that to model for close to 330 million people . Look at the sample here https://projects.fivethirtyeight.com/biden-approval-rating/ .
Comment 1c: I do not need to revisit my statistics thanks. This is because I understand that the same techniques can not be applied as it was in the US to Nigeria whose elections are affected by various and different factors. This polls has used phones as a means of collecting data. Can this method effectively and uniquely cover 50% of the electorates in Nigeria? In the US maybe, in UK maybe, in Australia, maybe, in Canada maybe but in Nigeria, I ask again can it? Same can be said of online polls when it comes to Nigeria. So using a sample data that effectively omits a vast majority of the electorates should be ignored and politely made insignificant.
My Response: his polls has used phones as a means of collecting data. Can this method effectively and uniquely cover 50% of the electorates in Nigeria?” Yes, weight. If Americans can use 1,000 to model 330m people, 15,500 for 200m is just enough with appropriate weighing. Yet, I am not saying they are correct but I am saying that the sample size does not need to be 10 million to be accepted. I do not believe these polls because the models are nascent with limited historical data for validity. But the sample size is fair.
---
Register for Tekedia Mini-MBA (Jun 3 - Sep 2, 2024), and join Prof Ndubuisi Ekekwe and our global faculty; click here.
In 2015 when Buhari was leading the polls against Johnathan, many people did not query the sample size.