<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Dappa ibialabo on Medium]]></title>
        <description><![CDATA[Stories by Dappa ibialabo on Medium]]></description>
        <link>https://medium.com/@dappaibialabo1234?source=rss-5e729454ef32------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*sYSOOrYIjcrEAPxZo8BolQ.jpeg</url>
            <title>Stories by Dappa ibialabo on Medium</title>
            <link>https://medium.com/@dappaibialabo1234?source=rss-5e729454ef32------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 16:21:19 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@dappaibialabo1234/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[As Bolt’s Nigerian Driver Numbers Jump 92%, 116 Port Harcourt Rides Reveal Why Drivers Charge Above…]]></title>
            <link>https://medium.com/@dappaibialabo1234/as-bolts-nigerian-driver-numbers-jump-92-116-port-harcourt-rides-reveal-why-drivers-charge-above-c1dfc2301509?source=rss-5e729454ef32------2</link>
            <guid isPermaLink="false">https://medium.com/p/c1dfc2301509</guid>
            <category><![CDATA[bolt]]></category>
            <category><![CDATA[research]]></category>
            <category><![CDATA[data-analysis]]></category>
            <category><![CDATA[e-hailing]]></category>
            <category><![CDATA[data-visualization]]></category>
            <dc:creator><![CDATA[Dappa ibialabo]]></dc:creator>
            <pubDate>Thu, 24 Sep 2026 02:18:58 GMT</pubDate>
            <atom:updated>2026-09-24T02:22:01.802Z</atom:updated>
            <content:encoded><![CDATA[<h3>As Bolt’s Nigerian Driver Numbers Jump 92%, 116 Port Harcourt Rides Reveal Why Drivers Charge Above the App, and Why Its Fares Still Ignore Drivers&#39; Fuel Costs</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*HLktsv7ZqB0pRhmURkMnDw.png" /><figcaption><em>With Bolt’s driver numbers surging, a study of 116 rides from Rukpokwu/SARS Road to UNIPORT finds drivers charge nearly 77% above the app fare, a gap that closely tracks the estimated cost of fuel.</em></figcaption></figure><p><strong><em>By</em></strong> <a href="https://medium.com/u/5e729454ef32">Dappa ibialabo</a> ,<a href="https://medium.com/u/62f643d660a0">Esther Nwodo</a> , Justice Ewelie, <a href="https://medium.com/u/71c3c7a220a7">Prosperiyowuna</a> ,Otti Christopher Favour, <a href="https://medium.com/u/c89b424c38d4">Godsonchigozie</a> , <a href="https://medium.com/u/2e2239c25d7f">Kumte Barivure</a> ,Mensanawaji Levi,<a href="https://medium.com/u/1bb51f01c12c">Franca Andy</a> <strong><em>with the Harvoxx Tech Hub Data Analytics Accelerator research team</em></strong></p><p>Anyone who has booked a Bolt in Port Harcourt knows the ritual. The app shows one price. The driver, politely or otherwise, names another. The passenger sighs, negotiates, or cancels and tries again.<br>That ritual is about to involve a lot more people. Bolt has seen a 92% surge in driver registrations in Nigeria following Uber’s exit [SOURCE AND DATE REQUIRED — see Editor’s Notes]. Every one of those new drivers signs up to the same fare structure, so the question of how well it works has become more urgent.<br>We set out to measure it. Our team recorded 116 real Bolt trips on one of the city’s busiest corridors: Rukpokwu/SARS Road to the University of Port Harcourt. For each ride we logged the time, distance, vehicle, the fare quoted by the app, the fare the driver asked for, the fuel price that day, and the car’s fuel efficiency. We cleaned the data in Excel, checked distances against Google Maps, and ran correlation and regression analysis.</p><h3>Why drivers charge more: the gap is large</h3><p>The average app-quoted fare in our sample was about ₦3,727. The average fare the driver actually asked for was about ₦6,610. That is a difference of nearly ₦2,900, or 77% above the quote.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NZxzPUXen709rxN6IBkJ2g.jpeg" /><figcaption><em>figure 1</em></figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0VmgZ-361yvYQtseHxrsjw.jpeg" /><figcaption><em>figure 2 Average quoted fare, driver-asked fare and estimated fuel cost per trip (n = 116). Fuel cost is estimated from each car’s rated efficiency under a worst-case assumption. Source: Harvoxx Tech Hub Data Analytics Accelerator.</em></figcaption></figure><h3>The gap looks like a fuel bill</h3><p>The average estimated fuel cost per trip in our data was about ₦2,890, within ₦10 of the gap. That could be a coincidence, and we are not claiming it proves anything. But the arithmetic is uncomfortable.<br>Bolt takes a 25% commission, so a ₦3,727 fare leaves the driver about ₦2,795. That is slightly less than the estimated fuel cost of the average trip, before maintenance, tyres, or a single naira of profit.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*tM8nrIKk1OROdHgurdBjMw.jpeg" /><figcaption><em>figure 3 Where the average app-quoted fare goes. Driver receipts are calculated as 75% of the average quoted fare. Fuel cost is a worst-case estimate.</em></figcaption></figure><p>One caveat matters. Our fuel figure comes from published efficiency ratings, not metered readings, and the worst-case assumption likely overstates the bill on a good day. But Port Harcourt rarely offers good days. Traffic around the Rukpokwu market junction and the Eliozu flyover can push real consumption well beyond what a manufacturer’s brochure promises.<br>Read this way, the negotiation at the car window looks less like greed and more like a driver asking the passenger to cover what the app leaves out.</p><h3>Why hasn’t Bolt updated its fares?</h3><p>The honest answer is that we don’t know. Bolt’s pricing methodology is proprietary. What our data can do is show what the app’s fares do and do not respond to.<br>First, they do not respond to experience. We tested whether the number of rides a driver has completed affects the fare the app quotes. It does not: the relationship was statistically indistinguishable from zero. Some drivers in our sample had completed tens of thousands of trips, and the app quoted them the same fare as anyone else on the same route.<br>Second, fuel cost explains only about 26% of the variation in what drivers take home per ride (R² = 0.258). Distance and time of day mattered more.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NLGBsYXVkS4hnf5SpvqzRw.jpeg" /><figcaption>figure 4 Share of variation explained in two regressions on the 116-trip sample. Fuel cost was not a statistically significant predictor in our multiple regression (p = 0.452).</figcaption></figure><p>One reading, and it is an inference rather than a finding, is that Bolt&#39;s fare model may share the same weakness as our own. In our regression, estimated fuel cost was not a statistically significant predictor of what drivers take home (p = 0.452). That is largely because our fuel figures were built from each car&#39;s published efficiency range, using the worst-case scenario to estimate the cost of every trip. Such generalised numbers cannot capture what Port Harcourt drivers actually face: pump prices that shift within the same week, informal vendors when stations run dry, gridlock, detours and roads that add wear to every trip. If Bolt&#39;s pricing relies on similar assumptions, it would explain why the app&#39;s fares have not been updated for Nigerian drivers, even as the platform now takes on new drivers under the same structure. Bolt&#39;s methodology is proprietary, so this cannot be confirmed from outside. Driver-reported fuel data would test it.</p><p>The registration surge raises the stakes. If the gap we observed is structural, more drivers means more people absorbing it, and more passengers negotiating over it. Our study was not designed to test the effect of new registrations, but it describes the fare structure they are joining.</p><h3>Other patterns in the data</h3><p>Trips after 9 p.m. carried the highest fares, averaging roughly ₦4,300, about a fifth more than midday trips. Toyota Corollas and Camrys dominate the sample, favoured for fuel economy, cheap spare parts and mechanics who know them. That is a choice shaped by driver economics: Bolt does not price by vehicle class, so a driver in a fuel-efficient Corolla and one in a thirstier car are quoted the same fare for the same trip.</p><h3>What could change</h3><ul><li><strong>Ask drivers, not just passengers</strong>. A feedback system capturing fuel price, traffic and trip difficulty would give Bolt better data than star ratings alone.</li><li><strong>Review fares on a schedule</strong>. In a market where fuel prices shifted sharply after the removal of the subsidy, a review every six months is a reasonable starting point.</li><li><strong>Use local cost data</strong> instead of relying only on manufacturer-stated efficiency figures.</li><li><strong>Reward experience with a loyalty</strong> system for drivers with long ride histories, since ride count currently has no bearing on the fare.</li></ul><h3>What we can’t say</h3><p>This is a snapshot, not a census. It covers 116 trips, one route and one city, and some vehicle models appear only a handful of times. Fuel costs were estimated, not metered. And because Bolt’s pricing methodology is proprietary, any explanation of why its fares have not moved is inference. We can show the pattern, but only Bolt can explain it.<br>What we can say is that the ritual at the car window is not random. In our data, the gap between the app and the street follows the cost of running a car in Port Harcourt closely enough that it deserves a real answer, from the platform, from regulators, and from the rest of us who keep haggling over it.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c1dfc2301509" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[30 Built-In Python Functions Every Data Professional Should Know]]></title>
            <link>https://medium.com/@dappaibialabo1234/30-built-in-python-functions-every-data-professional-should-know-7541cceb54d3?source=rss-5e729454ef32------2</link>
            <guid isPermaLink="false">https://medium.com/p/7541cceb54d3</guid>
            <category><![CDATA[data]]></category>
            <category><![CDATA[software-engineering]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[data-analysis]]></category>
            <category><![CDATA[python-programming]]></category>
            <dc:creator><![CDATA[Dappa ibialabo]]></dc:creator>
            <pubDate>Sun, 21 Jun 2026 22:22:08 GMT</pubDate>
            <atom:updated>2026-07-22T17:59:37.635Z</atom:updated>
            <content:encoded><![CDATA[<p>30 Built-In Python Functions Every Data Professional Should Know</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XLBX7cWCVH0-eDikw8Tiqw.png" /></figure><p>A practical reference for beginners and aspiring data analysts</p><p>When I first started learning Python for data analysis, I assumed I&#39;d need to memorize hundreds of functions before I could call myself proficient. It didn&#39;t take long to realize that wasn&#39;t true. A solid grasp of a relatively small set of built-in functions is enough to carry you through the vast majority of everyday programming and data tasks — from cleaning a messy dataset to writing a quick, readable script. The rest, you pick up as you go.</p><p>That realization is what inspired this piece. I put together a list of thirty built-in Python functions that I find myself reaching for again and again, and I wanted to write about not just what they do, but why they matter and how they fit together in practice.</p><p>Let&#39;s start with the basics — how a program talks to the people using it. Every Python developer&#39;s first function is almost always `print()`, the simple command that displays output to the console. Just as essential is `input()`, which lets a program capture whatever a user types in. When you need to work with files rather than direct input, `open()` handles that, giving you access to read, write, or append to a document. And whenever you&#39;re exploring a function or library you&#39;ve never used before, `help()` is there to pull up its documentation right inside your environment, saving you a trip to Google.</p><p>Data, of course, rarely arrives in exactly the form you need it. This is where type conversion comes in. `int()` and `float()` convert values into whole numbers and decimals respectively, while `str()` does the reverse, turning numbers or other objects into text. Before processing a piece of data, it&#39;s often worth checking what you&#39;re actually dealing with, and that&#39;s the job of `isinstance()`, which confirms whether an object belongs to a particular type. These four functions might seem small, but they quietly prevent a huge number of bugs.</p><p>Once your data is in the right format, you&#39;ll usually want to organize it into a structure. Python gives you four core ones: `list()` for an ordered, changeable collection; `tuple()` for an ordered collection that can&#39;t be altered once created; `dict()` for pairing keys with values; and `set()` for storing unique items without worrying about order or duplicates. Knowing which structure to reach for in a given situation is one of those skills that separates someone who&#39;s just writing code from someone who&#39;s writing it well.</p><p>With a structure in hand, the next step is usually to work through it — and Python has a strong set of tools for that too. `len()` tells you how many items you&#39;re dealing with. `range()` generates a sequence of numbers, almost always the backbone of a loop. `sorted()` and `reversed()` let you reorder a sequence either ascending or in reverse, without altering the original. `enumerate()` is one I use constantly; it pairs each item in a list with its position, so you&#39;re never stuck writing a separate counter variable just to track where you are. And `zip()` lets you walk through two or more lists side by side, which is invaluable when you&#39;re matching related pieces of data, like names against scores, or dates against transactions.</p><p>There&#39;s also a more functional style of programming Python supports, one that lets you apply logic across an entire collection without writing an explicit loop. `map()` applies a function to every item in a sequence, while `filter()` keeps only the items that meet a certain condition. `sum()`, `min()`, and `max()` do exactly what they sound like — adding up, or finding the smallest and largest values in a collection. And when you need a yes-or-no answer about an entire group of values, `any()` tells you if at least one item is true, while `all()` tells you if every single one is. Together, these functions make code shorter, often faster, and — once you&#39;re used to reading them — genuinely easier to follow.</p><p>Numbers also need to be presented well, not just calculated correctly. `abs()` strips away the sign of a number to give you its absolute value. `round()` trims a number down to a chosen number of decimal places, which matters more than people expect when you&#39;re reporting figures to a client or stakeholder. And `format()` lets you control exactly how a number appears as text — adding commas, currency symbols, or fixed decimal places so your output looks professional rather than raw.</p><p>Finally, there are two functions that exist purely to help you understand what you&#39;re working with. `type()` tells you the data type of any object, and `id()` returns its unique identity in memory — both useful when you&#39;re debugging or trying to make sense of unfamiliar data.</p><p>None of these thirty functions are flashy. None of them will show up in a viral coding demo. But they are the functions that quietly hold together almost everything you&#39;ll ever write in Python, especially in data work. Built-in functions are optimized, well-documented, and instantly recognizable to any other Python developer reading your code — which means relying on them doesn&#39;t just make your work easier, it makes it easier for others to understand too.</p><p>If you&#39;re early in your data journey, my advice is simple: don&#39;t rush to learn advanced libraries before you&#39;re fluent in these fundamentals. Spend time with this list. Write small scripts that use each one. You&#39;ll find that the more comfortable you become with these thirty functions, the less intimidating everything else in Python starts to feel.</p><p>If you found this useful, follow for more practical, beginner-friendly breakdowns of the tools and concepts that matter in data science and analytics.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7541cceb54d3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Can Data Fix Nigeria’s Broken Fare System? We’re Building a Model to Find Out]]></title>
            <link>https://medium.com/@dappaibialabo1234/can-data-fix-nigerias-broken-fare-system-we-re-building-a-model-to-find-out-aac67176eab5?source=rss-5e729454ef32------2</link>
            <guid isPermaLink="false">https://medium.com/p/aac67176eab5</guid>
            <dc:creator><![CDATA[Dappa ibialabo]]></dc:creator>
            <pubDate>Sat, 20 Jun 2026 11:23:11 GMT</pubDate>
            <atom:updated>2026-06-20T11:23:11.187Z</atom:updated>
            <content:encoded><![CDATA[<p>Can Data Fix Nigeria’s Broken Fare System? We’re Building a Model to Find Out</p><p>If you&#39;ve ever ordered a ride in Lagos, Port Harcourt, or Abuja right after a fuel price hike, you already know the drill. Fares jump overnight, sometimes by margins that feel arbitrary, sometimes triggering driver strikes or passenger backlash. There&#39;s rarely a clear explanation for the numbers — just a sense that *something* changed and prices followed, somehow.</p><p>That &quot;somehow&quot; is the problem.</p><p>Nigeria&#39;s e-hailing industry, like much of its broader transport sector, lacks a transparent, evidence-based system for adjusting fares when fuel prices move. Pricing decisions are often reactive, inconsistent, and disconnected from any model that drivers, riders, or regulators can actually scrutinize. In a market where fuel costs can swing sharply within weeks, that gap isn&#39;t just an inconvenience — it&#39;s a structural weakness.</p><p>This is the problem our team set out to address.</p><p>The Project</p><p>We&#39;re building a data-driven fare prediction model designed to estimate fair transportation charges based on fuel price fluctuations. The goal is simple to state and hard to execute well: create a framework that ties fare adjustments to real, measurable inputs, rather than guesswork or unilateral decisions.</p><p>If it works the way we intend, the model won&#39;t just be an academic exercise. It&#39;s meant to function as a practical reference point — something e-hailing companies, transport regulators, and policymakers can actually use to make fare decisions that are explainable and defensible, instead of opaque.</p><p>Why This Matters Beyond the Numbers</p><p>It&#39;s easy to treat a project like this as a purely technical exercise — pull some data, run a regression, present a chart. But the stakes here are human. Drivers need fares that reflect their actual operating costs. Riders need pricing they can trust isn&#39;t being adjusted on a whim. Regulators need a credible basis for oversight instead of negotiating fare caps in the dark.</p><p>A transparent fare model has the potential to reduce friction across all three groups — not by making everyone happy, but by replacing ambiguity with a shared, data-backed reference point.</p><p>Where We’re Headed</p><p>This is still early-stage work, and we&#39;re approaching it the way any serious analytics project should be approached: starting with the data, testing assumptions, and resisting the urge to overpromise before the model has actually been validated. The road from &quot;interesting correlation&quot; to &quot;tool people can rely on&quot; is long, and we intend to walk it properly.</p><p>We&#39;ll be sharing updates as the project develops — the data sources we draw on, the modeling choices we make, and eventually, what the results actually show.</p><p>This is, at its core, a belief we&#39;re putting into practice: that analytics shouldn&#39;t just serve dashboards and quarterly reports. It can — and should — be pointed at problems that affect how people get to work, get paid, and get home every day.</p><p>---</p><p>Team: Esther and others</p><p>Follow along for updates as the model takes shape.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=aac67176eab5" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What Should a Bolt Ride Really Cost When Fuel Increases by ₦100?]]></title>
            <link>https://medium.com/@dappaibialabo1234/what-should-a-bolt-ride-really-cost-when-fuel-increases-by-100-5263126fdb5a?source=rss-5e729454ef32------2</link>
            <guid isPermaLink="false">https://medium.com/p/5263126fdb5a</guid>
            <dc:creator><![CDATA[Dappa ibialabo]]></dc:creator>
            <pubDate>Fri, 19 Jun 2026 14:41:52 GMT</pubDate>
            <atom:updated>2026-06-19T14:41:52.761Z</atom:updated>
            <content:encoded><![CDATA[<p>What Should a Bolt Ride Really Cost When Fuel Increases by ₦100?</p><p>It sounds like a simple question. But the more you think about it, the more you realise nobody has actually answered it properly.</p><p>Every time fuel prices go up in Nigeria, the same cycle plays out. Drivers feel shortchanged by app-generated fares. Riders feel the negotiated prices are unfair. Arguments happen at pickup points. Trust between drivers and passengers erodes a little more. And everyone goes home frustrated, with no real resolution.</p><p>The problem is not that people are unreasonable. The problem is that there is no clear, data-backed formula that tells any party — drivers, riders, regulators, or e-hailing companies — exactly how transport fares should adjust when fuel prices increase. Decisions are made on gut feeling, desperation, or whoever argues loudest.</p><p>My team and I think data can change that.</p><p>---</p><p>## The Project</p><p>We have started a comprehensive analytics project focused on transport pricing in Port Harcourt. Specifically, we are investigating the relationship between:</p><p>- Fuel (PMS) prices<br>- Distance travelled<br>- App-generated fares (Bolt, etc.)<br>- Driver-negotiated fares<br>- General transportation costs in the city</p><p>Our goal is to develop a model that can estimate a **fair transport charge per kilometre for every ₦100 increase in PMS price** — something transparent, explainable, and actually useful in the real world.</p><p>---</p><p>## Why This Matters</p><p>This is not a classroom exercise. Port Harcourt is one of Nigeria&#39;s busiest commercial cities, and the tension between fuel costs and transport fares affects hundreds of thousands of daily commuters. If a reliable pricing framework existed, it could serve:</p><p>**Drivers** — who need to know their earnings are fair relative to their running costs.</p><p>**Riders** — who deserve pricing they can understand and predict.</p><p>**E-hailing companies** — who need dynamic fare models grounded in local economic realities.</p><p>**Transport ministries and regulators** — who currently have little data to inform policy decisions.</p><p>**Researchers** — building on the broader question of urban mobility pricing in developing economies.</p><p>---</p><p>## Where We Are Now</p><p>Data collection has begun. We are gathering real fare and fuel price data from across Port Harcourt, and the analysis phase is coming soon.</p><p>I am working on this project alongside my team members Esther Nwodo and others — a group that believes the most meaningful data work is the kind that solves problems people actually live with every day.</p><p>When the results are ready, we will share them openly.</p><p>---</p><p>## A Question Worth Sitting With</p><p>When fuel increases by ₦100, how much should your transport fare increase per kilometre?</p><p>We are gathering the data to find out. Follow along.</p><p>---</p><p>*This project is part of ongoing work at Havvox Tech Hub. #DataAnalytics #DataDriven #TechForImpact #BoltNigeria #Havvox*</p><p>---</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5263126fdb5a" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>