<?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 Rose Kawila on Medium]]></title>
        <description><![CDATA[Stories by Rose Kawila on Medium]]></description>
        <link>https://medium.com/@rosekawila?source=rss-39a870992daa------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*zYFyywDi-fRswHiMT4fcmg.jpeg</url>
            <title>Stories by Rose Kawila on Medium</title>
            <link>https://medium.com/@rosekawila?source=rss-39a870992daa------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 02:18:24 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@rosekawila/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[Are We Keeping Up With Technology?]]></title>
            <link>https://rosekawila.medium.com/are-we-keeping-up-with-technology-0aebcaed4433?source=rss-39a870992daa------2</link>
            <guid isPermaLink="false">https://medium.com/p/0aebcaed4433</guid>
            <category><![CDATA[education]]></category>
            <category><![CDATA[information-technology]]></category>
            <category><![CDATA[knowledge]]></category>
            <category><![CDATA[africa]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Rose Kawila]]></dc:creator>
            <pubDate>Mon, 21 Sep 2026 07:49:53 GMT</pubDate>
            <atom:updated>2026-09-21T08:05:07.346Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*OT80GG2n9Go40ODr" /><figcaption>Photo by <a href="https://unsplash.com/@hngstrm?utm_source=medium&amp;utm_medium=referral">H&amp;CO</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>Recently, we were sending messages to our house manager on WhatsApp and realised she wasn’t responding to any of them.</p><p>At first, we thought she was ignoring us.</p><p>Then we realised the problem.</p><p>She didn’t know how to send messages on WhatsApp.</p><p>She had only recently acquired a smartphone.</p><p>That small experience got me thinking about how quickly technology is changing and how easy it is for people to be left behind.</p><p>With all the recent developments in AI, my parents are still trying to understand what AI actually is. And I don’t mean Gen Alpha parents who grew up surrounded by technology. I mean those of us whose parents didn’t grow up with smartphones, social media, or AI assistants.</p><p>And honestly, if I am struggling to keep up, what about people who don’t work in technology?</p><h3>Technology is moving faster than we can keep up</h3><p>Every day, someone seems to be building a new technology.</p><p>And sometimes, by the time you discover a tool, there is already another one replacing it.</p><p>I recently started an LLM course and discovered a platform called DeepInfra. It has thousands of AI models, many of which I had never heard of before. There are models for text, speech, music and many other applications.</p><p>Then, while taking a Microsoft course on Microsoft Fabric, I started discovering just how much can now be done in data engineering.</p><p>More recently, during a workshop at The University of WitWatersRand in Johannesburg, South Africa, I was introduced to Microsoft Foundry.</p><p>You can give an LLM a system prompt, provide it with context such as company documents, and build an application that responds based on that information. You can even get the code needed to integrate the model into your backend.</p><p>And then there are tools that can take an idea described in a prompt and help you build and publish an application.</p><p>These are just a few examples.</p><p>There are probably hundreds of other tools and platforms being developed that I haven’t even discovered yet.</p><p>And that is what makes me stop and think.</p><p><strong>I work in technology, and even I am struggling to keep up with what is being built. What about someone who doesn’t work in technology?</strong></p><p>What about a small business owner who has never heard of an LLM?</p><p>What about a startup founder who doesn’t know that there are already tools that could help them automate part of their operations?</p><p>What about someone who is still learning how to use WhatsApp?</p><h3>So where is the gap?</h3><p>This is the question I keep coming back to.</p><p>We have access to more technology than ever before. Large companies like Microsoft, Google and others are building and releasing new tools at an incredible pace.</p><p>But are we actually using them?</p><p>Especially in Africa.</p><p>There are startups trying to solve very real problems while working with limited funding and small teams. At the same time, there are technologies that can potentially reduce the time and cost of building certain products and services.</p><p>Yet many people only discover these technologies much later.</p><p>Sometimes we hear about a tool when it is already being replaced or deprecated.</p><p>So, is the problem access?</p><p>Is it awareness?</p><p>Is it the cost of learning?</p><p>Is it the lack of people translating these technologies into practical solutions for African businesses?</p><p>Or are we simply not plugged into the channels where these developments are happening?</p><p>I don’t think there is one simple answer.</p><h3>Maybe the problem isn’t the technology</h3><p>Going back to our house manager, the smartphone was not the problem.</p><p>The technology existed.</p><p>WhatsApp existed.</p><p>The ability to communicate existed.</p><p>The missing piece was knowing how to use it.</p><p>Perhaps AI is facing a similar challenge.</p><p>The technology is developing incredibly quickly, but the ability to understand, access and apply that technology is not developing at the same pace.</p><p>And this isn’t only an African problem.</p><p>But in countries where businesses already have fewer resources, the gap can have a bigger impact.</p><p>A startup shouldn’t necessarily need a huge team just to build capabilities that can now be supported by existing technologies.</p><p>But knowing that those technologies exist is the first step.</p><p>Knowing how to use them is the second.</p><p>And knowing <strong>which problem to solve with them</strong> might be the most important step of all.</p><h3>So, do we build for Africa?</h3><p>I keep thinking about a question I heard during the AFRETEC SEED Workshop:</p><p><strong>Do we build for Africa?</strong></p><p>I think the question goes beyond creating AI models or applications specifically for African markets.</p><p>It is also about making sure that African businesses, developers, students and communities are not simply consumers of technology developed elsewhere.</p><p>We need to understand what is available.</p><p>We need to experiment with it.</p><p>We need to ask how these technologies can be applied to our own problems.</p><p>And where existing technology doesn’t solve the problem, then perhaps that is where we build.</p><p>Because technology is not slowing down for us.</p><p>New models will continue to appear. New platforms will continue to be released. New tools will continue to make things that once required entire teams possible with a much smaller amount of time and resources.</p><p>The question is whether we will keep discovering these technologies years later, or whether we will find ways to stay closer to the frontier.</p><p>Maybe the question isn’t whether Africa is ready for AI.</p><p><strong>Maybe the question is whether we are keeping up with what is already available.</strong></p><p>Are we building for Africa, or are we still waiting for technology to reach us?</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=0aebcaed4433" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Beyond the Dashboard: Using Statistics to Analyze Malaria Trends in Kenya]]></title>
            <link>https://rosekawila.medium.com/beyond-the-dashboard-using-statistics-to-analyze-malaria-trends-in-kenya-63d307ad347f?source=rss-39a870992daa------2</link>
            <guid isPermaLink="false">https://medium.com/p/63d307ad347f</guid>
            <category><![CDATA[trends]]></category>
            <category><![CDATA[health]]></category>
            <category><![CDATA[malaria]]></category>
            <category><![CDATA[data]]></category>
            <category><![CDATA[statistics]]></category>
            <dc:creator><![CDATA[Rose Kawila]]></dc:creator>
            <pubDate>Thu, 05 Mar 2026 06:57:56 GMT</pubDate>
            <atom:updated>2026-03-05T06:57:56.822Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*CTvFS8QYrUZorS7h" /><figcaption>Photo by <a href="https://unsplash.com/@wolfgang_hasselmann?utm_source=medium&amp;utm_medium=referral">Wolfgang Hasselmann</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>We talk a lot in data science about using data to generate insights that inform strategic decisions. But there is a difference between simply analyzing data and producing insights that actually mean something. I once read that the real test of good analysis is whether you can answer the question: <strong>“What’s in it for us?”</strong> In other words, can the insight lead to action?</p><p>Most analysts rely heavily on dashboards and visualization tools like Power BI or Tableau. Those tools are incredibly useful for identifying trends, comparisons, and patterns at a glance. But sometimes statistics can reveal layers of meaning that a chart alone might not fully explain.</p><p>This project was an opportunity to explore that idea. I analyzed a simple dataset from the <strong>World Health Organization’s Global Health Observatory</strong> containing malaria indicators for Kenya between <strong>2000 and 2024</strong>. The dataset focused on two key indicators:</p><ul><li><strong>Estimated Malaria Mortality Rate</strong> (per 100,000 population)</li><li><strong>Estimated Malaria Incidence Rate</strong> (per 1,000 population at risk)</li></ul><p>The main question was straightforward: <strong>Have malaria mortality and incidence rates in Kenya actually declined over time in a statistically significant way?</strong></p><p>A few secondary questions also emerged:</p><ul><li>Was the decline consistent across the entire period?</li><li>Did the early malaria interventions of the 2000s make a measurable difference?</li><li>How strong is the relationship between time and each indicator?</li></ul><p>To answer these questions, the analysis used a mix of statistical approaches:</p><ul><li><strong>OLS Linear Regression</strong></li><li><strong>Mann–Kendall Trend Test</strong></li><li><strong>Mann–Whitney U Test</strong></li><li><strong>Spearman Rank Correlation</strong></li></ul><p>Combining both <strong>parametric and non-parametric methods</strong> helped ensure the conclusions were robust.</p><h3>First Look at the Data</h3><p>Before running any statistical tests, the first step was simply exploring the distribution of the data. Both indicators showed <strong>right-skewed distributions</strong>. Most of the values clustered toward the lower end of the scale, while a smaller number of high values appeared in the early 2000s.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VA43FukHz2ucarfkB6_U7A.png" /></figure><p>This immediately hinted at a story. Malaria rates were <strong>much higher at the beginning of the study period</strong>, and had dropped substantially by the later years. But visual impressions alone are not enough. The next step was to confirm the statistical properties of the data.</p><h3>Step 1: Testing Normality : Shapiro-Wilk Test</h3><p>Many statistical methods assume that the underlying data follows a <strong>normal distribution</strong>. If this assumption does not hold, some tests become unreliable. To check this, the <strong>Shapiro–Wilk test</strong> was used. This test is particularly appropriate for small samples like this one (25 observations per indicator). The results were clear:</p><pre>Shapiro-Wilk Test: Mortality Rate<br>W statistic : 0.7121<br>p-value: 0.000011<br>Sample size : 25<br><br>Shapiro-Wilk Test: Incidence Rate<br>W statistic : 0.6770<br>p-value     : 0.000004<br>Sample size : 25</pre><p>Both p-values are far below the standard significance level of <strong>0.05</strong>, meaning the null hypothesis of normality must be rejected. In simple terms: <strong>the data is not normally distributed.</strong></p><p>That finding aligned with the skewed histograms observed earlier. Because of this, the analysis relied heavily on <strong>non-parametric methods</strong>, which do not require normally distributed data.</p><h3>Step 2: Is There Actually a Downward Trend?</h3><p>The core question of the project was whether malaria rates in Kenya show a <strong>statistically significant downward trend over time</strong>. To answer this, two complementary approaches were used:</p><ul><li><strong>OLS Linear Regression</strong> to estimate the slope of the trend</li><li><strong>Mann-Kendall Trend Test</strong>, a non-parametric test widely used in environmental and health time-series analysis</li></ul><p>Using both methods helps validate the results even when distributional assumptions are violated.</p><h3>OLS Regression Results</h3><pre>OLS Linear Regression - Mortality Rate<br>Slope (β₁): -0.7565 units per year<br>Intercept: 1548.0138<br>R²: 0.6683<br>p-value : 0.000000 (one-tailed)<br>95% CI slope : [-0.9864, -0.5266]<br><br>OLS Linear Regression - Incidence Rate<br>Slope (β₁): -6.0810 units per year<br>Intercept: 12347.5378<br>R²: 0.6145<br>p-value : 0.000002 (one-tailed)<br>95% CI slope : [-8.1585, -4.0035]</pre><p>Both slopes are negative and highly significant. This means that, on average:</p><ul><li>The <strong>malaria mortality rate decreased by about 0.76 deaths per 100,000 people each year</strong></li><li>The <strong>incidence rate decreased by roughly 6 cases per 1,000 people at risk each year</strong></li></ul><p>Time alone explains roughly <strong>67% of the variation in mortality</strong> and <strong>61% of the variation in incidence</strong>, which is surprisingly strong for a single variable.</p><h3>Mann-Kendall Trend Test</h3><pre>Mann-Kendall Trend Test<br>  Trend: decreasing<br>  Tau (τ): -0.8867<br>  p-value : 0.000000<br>  Sen&#39;s Slope  : -0.3683 units per year<br>  S statistic  : -266.0</pre><p>The Mann-Kendall test confirmed the same conclusion: <strong>both indicators show statistically significant downward trends. </strong>The Tau statistic is particularly informative. For mortality, a Tau of <strong>−0.8867</strong> means that in nearly <strong>89% of all year-to-year comparisons</strong>, the later year had a lower mortality rate than the earlier year. That level of consistency is extremely rare in real-world public health data.</p><h3>Step 3: Did Early Malaria Interventions Make a Difference?</h3><p>When the trend lines were plotted, an interesting pattern emerged. The decline did not occur evenly across the entire 25-year period. Instead, the data appeared to show <strong>two distinct phases</strong>:</p><ol><li><strong>A steep decline between 2000 and 2008</strong></li><li><strong>A plateau from 2009 to 2024</strong></li></ol><p>This raised another question: Was the difference between these two periods statistically significant? To test this, the <strong>Mann-Whitney U Test</strong> was used. This test compares two independent groups without assuming normality.</p><pre>Mann-Whitney U Test - Mortality Rate<br>U statistic  : 144.0000<br>p-value : 0.000026 (one-tailed)<br><br>Mann-Whitney U Test - Incidence Rate<br>U statistic  : 135.0000<br>p-value : 0.000201 (one-tailed)</pre><h3>Descriptive Comparison</h3><pre>Early period (2000–2008) : n=9,  mean=32.82,  median=32.51<br>  Late period  (2009–2024) : n=16,  mean=22.14,  median=21.86<br><br>Early period (2000–2008) : n=9,  mean=170.52,  median=184.42<br>  Late period  (2009–2024) : n=16,  mean=80.00,  median=81.03</pre><p>Both indicators dropped dramatically between the two periods.</p><h3>Mann-Whitney Results</h3><p>The mortality result is particularly striking. A <strong>U statistic of 144 out of 144</strong> means that every single mortality value in the early period was higher than every value in the later period. That represents <strong>perfect separation</strong> between the two groups. The incidence results were almost as strong.</p><p>This confirms that the early intervention period had <strong>significantly higher malaria rates than the later years</strong>, providing strong statistical evidence of real progress.</p><h3>Step 4: Measuring the Strength of the Relationship</h3><p>The final step was measuring how strongly each indicator is associated with time. For this, <strong>Spearman’s Rank Correlation</strong> was used.</p><pre>Spearman Correlation - Mortality Rate<br>ρ (rho) : -0.9654<br>p-value: 0.000000 (one-tailed)<br>n : 25<br><br>Spearman Correlation - Incidence Rate<br>ρ (rho) : -0.7562<br>p-value: 0.000006 (one-tailed)<br>n : 25</pre><p>A correlation of <strong>−0.9654</strong> is extremely strong. It means that simply knowing the year provides a very accurate indication of the mortality rate. The incidence correlation is still strong but noticeably weaker. That difference turns out to be meaningful.</p><h3>The Bigger Picture</h3><p>Putting all of these results together reveals an important story. Kenya’s malaria control efforts during the early 2000s , including <strong>insecticide-treated bed nets, artemisinin-based therapies, and the WHO Roll Back Malaria initiative, </strong>had a measurable and statistically significant impact.</p><p>The decline in malaria indicators during this period is not just visible in charts. It holds up across multiple statistical tests. However, the data also highlights a challenge. After around <strong>2008</strong>, both indicators entered a plateau where progress slowed significantly. The difference between the mortality correlation (<strong>ρ = −0.9654</strong>) and the incidence correlation (<strong>ρ = −0.7562</strong>) suggests something interesting: Kenya has become <strong>much better at treating malaria once people get it</strong>, but <strong>preventing transmission has proven harder to sustain</strong>. That is the type of insight statistics can uncover beyond what a simple trend line might suggest. The story is not just that malaria rates went down. It is <strong>how consistently they declined, when the major shift occurred, and where progress appears to have slowed</strong>.</p><h3>Final Thoughts</h3><p>This project was a reminder that even a relatively small dataset can produce meaningful insights when approached carefully. Using a combination of exploratory analysis and statistical testing makes it possible to move from simply describing trends to <strong>understanding the dynamics behind them</strong>. The analysis also highlights an important point about data science: sometimes the most valuable insights come from asking deeper questions about patterns that initially seem obvious.</p><p>If you’d like to explore the analysis in more detail, including the full notebook, feel free to leave a comment below.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=63d307ad347f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Africa and the Global AI Race: Are We Ready?]]></title>
            <link>https://rosekawila.medium.com/africa-and-the-global-ai-race-are-we-ready-6f1d1bdc5d08?source=rss-39a870992daa------2</link>
            <guid isPermaLink="false">https://medium.com/p/6f1d1bdc5d08</guid>
            <category><![CDATA[global]]></category>
            <category><![CDATA[africa]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[policy]]></category>
            <category><![CDATA[information-technology]]></category>
            <dc:creator><![CDATA[Rose Kawila]]></dc:creator>
            <pubDate>Tue, 17 Feb 2026 11:38:55 GMT</pubDate>
            <atom:updated>2026-02-17T11:38:55.127Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*spNTBRk0oRbYKoop" /><figcaption>Photo by <a href="https://unsplash.com/@randytarampi?utm_source=medium&amp;utm_medium=referral">Randy Tarampi</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>The recent U.S. AI policy direction really caught my attention. The focus on positioning America for global dominance in Artificial Intelligence shows just how strategic AI has become. It is no longer just about innovation , it is about economic power, national security, and global influence. For example, the recent U.S. AI Action Plan introduces export controls on advanced AI chips, particularly restricting sales to China, because the computing power used to train AI models is now considered a strategic asset.</p><p>Reading this made me pause and ask: where are we as Africa in this global AI race?</p><p>When we compare ourselves to global leaders, it is clear that we still have a long way to go, especially in foundational areas.</p><p>For instance, infrastructure remains a major challenge. AI development, particularly training large models, requires stable electricity and high-performance data centers. Yet, according to the World Bank, nearly half of Sub-Saharan Africa still lacks access to reliable electricity. Even in urban areas, power interruptions increase operational costs for businesses and research institutions. Without stable energy, large-scale AI innovation becomes difficult.</p><p>Despite lagging in large-scale AI development, AI adoption across Africa is already widespread. Recent reports show strong usage of tools like ChatGPT, with countries such as Kenya ranking among the highest users on the continent. AI is being applied in fintech, agriculture, healthcare, logistics, environmental monitoring, and other industries. Companies are even using generative AI to create billboard images, reducing the cost of hiring models and photographers. This demonstrates that adoption is accelerating, even if advanced AI development is not progressing at the same pace.</p><p>However, as adoption accelerates, I find myself asking whether adequate policies are in place to govern this growth. We have seen cases where AI has been used both positively and negatively. For instance, AI-generated images have been deployed in political campaigns for propaganda. Given Africa’s digital divide, such misuse can exacerbate social divisions and impact communities widely. From a policy perspective, progress remains uneven.</p><p>Encouragingly, progress is being made. The African Union has developed a Continental AI Strategy aimed at ensuring that AI development on the continent is ethical, inclusive, and aligned with African priorities. Countries like Kenya are in the process of developing national AI strategies. However, many countries still lack AI-specific policies. Some have data protection laws such as Kenya’s Data Protection Act of 2019 which provide a strong foundation. But others, do not yet have comprehensive data protection frameworks in place.</p><p>This highlights a clear gap: AI adoption is moving faster than policy development.</p><p>Personally, I believe our short-term focus should be on strengthening the fundamentals, including digital infrastructure, governance frameworks, electricity access, data protection, and digital literacy. Competing in advanced chip manufacturing or training frontier AI models will require significant capital, computing infrastructure, and coordinated strategy. That will take time. However, building strong systems for responsible AI use and developing context-specific solutions to African challenges is something we can actively work on now.</p><p>The global AI race is accelerating. The question for Africa is not whether AI will affect us. It already is. The real question is whether we will intentionally shape its development or simply consume what others build.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6f1d1bdc5d08" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What “The Making of a Manager” by Julie Zhuo Taught Me About Leadership and Growth]]></title>
            <link>https://rosekawila.medium.com/what-the-making-of-a-manager-by-julie-zhuo-taught-me-about-leadership-and-growth-3312f47c1894?source=rss-39a870992daa------2</link>
            <guid isPermaLink="false">https://medium.com/p/3312f47c1894</guid>
            <category><![CDATA[reading]]></category>
            <category><![CDATA[education]]></category>
            <category><![CDATA[books]]></category>
            <category><![CDATA[book-review]]></category>
            <category><![CDATA[professional-development]]></category>
            <dc:creator><![CDATA[Rose Kawila]]></dc:creator>
            <pubDate>Sat, 25 Oct 2025 08:16:27 GMT</pubDate>
            <atom:updated>2025-10-25T08:16:27.427Z</atom:updated>
            <content:encoded><![CDATA[<p>When I first picked up Julie Zhuo’s <em>The Making of a Manager</em>, I wasn’t sure what to expect. The title alone made me hesitate , after all, I’ve never really seen myself as a “manager.” I’ve always thought of management as something corporate and rigid, while leadership, to me, has always felt more natural and human.</p><p>But Zhuo’s honesty in this book drew me in. She became a manager at just 25, and she admits she was nervous, unprepared, and plagued by imposter syndrome. That vulnerability resonated with me deeply. Many of us find ourselves in leadership positions, whether at school, at work, or in our communities , without formal training or preparation. We just… figure it out as we go.</p><p>I remember one of our student leaders once confessing how nervous they were to take on their role. Reading Zhuo’s story made me wish I’d discovered this book earlier . I would have shared it with them in a heartbeat.</p><h3>Redefining What It Means to Be a “Manager”</h3><p>Before reading the book, I thought a manager and a leader were basically the same thing. In my mind, managers were just leaders with extra responsibilities and a bigger workload. But Zhuo changed that perspective for me.</p><p>A <em>manager</em>, she writes, isn’t automatically a leader. A title doesn’t make you one. A great manager leads by focusing on the <em>team’s success</em>, not their own. They believe that goals are achieved <em>together</em>, not individually.</p><p>That insight hit home. I’ve always found myself stepping up to lead group projects or initiatives, even when I didn’t want to. If I’m not active, progress tends to stall. Reading Zhuo helped me understand that leadership isn’t about control or authority, it’s about alignment, purpose, and trust.</p><h3>Lessons from My Own Experience</h3><p>A few years ago, I managed a small sales team. My boss constantly reminded me that hitting sales targets was the ultimate measure of success. Back then, I thought teamwork alone was enough. As long as the team was working hard and things were running smoothly, I assumed we were doing fine.</p><p>Zhuo’s perspective helped me see things differently. A manager’s success isn’t about being busy, it’s about <em>results</em>. If the team isn’t achieving its goals, the manager has missed the mark. That realization was humbling.</p><p>She also emphasizes three timeless principles of success: <strong>purpose, process, and people</strong>. That trio now forms the lens through which I view any leadership challenge.</p><h3>The Power of Inspiration over Instruction</h3><p>One of Zhuo’s biggest takeaways is that great managers inspire rather than instruct. People don’t respond well to being told what to do; they respond to <em>why</em> it matters.</p><p>I’ve always shied away from management roles because they seemed overwhelming. But the book helped me realize that managers don’t have to lead massive teams. You can be a manager of just a few people or even a project and still apply these principles.</p><p>Zhuo also introduced me to the idea of “critique meetings,” where her team would present ideas and receive candid feedback. That struck me as brilliant. In school, we often hold our ideas close, fearing they’ll be stolen or dismissed. But feedback, both positive and negative , is a gift. It sharpens ideas, strengthens collaboration, and builds trust.</p><h3>Embracing Imperfection</h3><p>One of my favorite insights from Zhuo is that managers don’t have to know everything. Their job isn’t to be the smartest person in the room. It’s to create a room where <em>everyone</em> can be smart.</p><p>That really resonated with me. I’ve often felt pressure to be perfect, to have all the answers. But Zhuo reminds us that it’s okay to say “I don’t know.” Growth happens when we learn alongside our teams.</p><p>Building trust also stood out to me. Trust doesn’t happen overnight. It’s built through time, consistency, and honesty. I’ve noticed that people respond better when I’m open about my struggles or uncertainties. Authenticity builds connection.</p><h3>Feedback, Culture, and Hiring Right</h3><p>Zhuo dedicates a lot of attention to feedback, and for good reason. Feedback, when delivered with care, is one of the most powerful tools a manager has. I still remember how motivated I felt after receiving positive feedback from a former boss. It reminded me that recognition fuels growth.</p><p>But she also warns against waiting too long to give feedback. No one should walk into a quarterly review surprised. Continuous, task-specific feedback helps people improve in real time and keeps everyone aligned.</p><p>Another section that struck me was her discussion on hiring. It’s tempting to bring in friends or familiar faces, but the best hires are those who have the <em>skills</em> and <em>fit</em> the culture. I’ve seen firsthand how hiring the wrong person, even someone well-meaning , can slow down a team.</p><h3>Managing People, Not Tasks</h3><p>It’s easy to micromanage, especially when you’re an expert in your field. I’ve fallen into that trap before, doing tasks myself instead of delegating because I thought it would be faster. But Zhuo helped me see that a manager’s job is to <em>build systems and culture</em>, not to do everything themselves. Delegation isn’t laziness, it’s leadership.</p><p>Conflicts, of course, are inevitable. When they arise, Zhuo encourages managers to approach them with empathy and directness. Difficult conversations, like handling underperformance or layoffs , are part of the job, but they must be handled with compassion.</p><h3>The Art of a Great Meeting</h3><p>Zhuo also taught me what makes meetings worthwhile. The best meetings, she says, are those where everyone feels their time was well spent, where people leave with more clarity, motivation, and focus than when they arrived.</p><p>She recommends setting clear agendas, defining next steps, and regularly evaluating whether recurring meetings are still useful. I especially liked her advice about one-on-one meetings: they should be about <em>the team member</em>, not the manager. These moments are opportunities to coach, unblock, and connect.</p><h3>My Biggest Takeaway</h3><p>At the end of the day, a manager’s success equals the team’s success. The goal isn’t to look good, it’s to help others thrive. Zhuo calls this the <em>multiplier effect</em>: if I can sell 1,000 lemonades, and I teach someone else to do the same, together we sell 2,000. True leadership is about multiplying impact.</p><p>Ultimately, the biggest lesson I took from <em>The Making of a Manager</em> is this:</p><blockquote><em>A great manager is not just a leader — they are a builder of leaders.</em></blockquote><blockquote>And that, I believe, is the kind of leadership the world needs more of.</blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3312f47c1894" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why is Checking the Distribution of Your Data Important?]]></title>
            <link>https://rosekawila.medium.com/why-is-checking-the-distribution-of-your-data-important-5570d759c3cb?source=rss-39a870992daa------2</link>
            <guid isPermaLink="false">https://medium.com/p/5570d759c3cb</guid>
            <category><![CDATA[education]]></category>
            <category><![CDATA[data-analysis]]></category>
            <category><![CDATA[data-visualization]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[information-technology]]></category>
            <dc:creator><![CDATA[Rose Kawila]]></dc:creator>
            <pubDate>Sun, 23 Feb 2025 19:04:59 GMT</pubDate>
            <atom:updated>2025-02-23T19:04:59.267Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*WjboaJBjpcfJ9Hl4" /><figcaption>Photo by <a href="https://unsplash.com/@markusspiske?utm_source=medium&amp;utm_medium=referral">Markus Spiske</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>In my journey as a data scientist, I’ve learned that one crucial step in data analysis that should never be overlooked is checking the distribution of your data. It’s easy to skip this step, but doing so can significantly impact the accuracy and reliability of your analysis. But why is it so important?</p><h4>1. Analyzing Data Spread</h4><p>Examining the distribution helps analysts understand how values are spread across the dataset. Key insights include:</p><ul><li>Which values appear most frequently?</li><li>Is the data skewed? If so, in which direction?</li><li>Are there missing values or outliers?</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IqX5qay-Uaae3O7Ueg1onw.png" /></figure><p>Identifying outliers early allows you to apply appropriate handling techniques, ensuring they do not distort your analysis.</p><h4>2. Choosing the Right Analysis or Model</h4><p>The distribution of your data determines which statistical or machine-learning techniques are appropriate. For example:</p><ul><li>If the relationship between variables is nonlinear (e.g., energy consumption vs. temperature), a linear regression model would not be suitable.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/940/1*sgsxLkXxtyKhCRXYaEFw-Q.png" /></figure><ul><li>In classification problems, checking distribution helps detect class imbalances in the target variable, which could impact model performance.</li><li>Some statistical tests, like the t-test and ANOVA, assume normality in the data. Checking the distribution ensures you apply the correct test.</li></ul><p>Ignoring data distribution can lead to incorrect assumptions and poor model selection.</p><h4>3. Feature Engineering and Data Transformation</h4><p>Data distribution plays a crucial role in feature engineering. For example:</p><ul><li>If data is heavily skewed, applying transformations like logarithmic scaling can normalize it, improving model performance.</li><li>Understanding the distribution can help uncover hidden patterns, such as seasonality in time series data.</li></ul><p>By analyzing distributions, you gain better control over preprocessing, leading to more effective and accurate predictions.</p><p>Checking the distribution of your data is fundamental. It provides insights into data structure, guides model selection, improves feature engineering, and ensures the application of appropriate statistical tests. By prioritizing this step, you can avoid common pitfalls and enhance the accuracy and reliability of your analysis.</p><p>Feel free to share your thoughts in the comments! 🚀</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5570d759c3cb" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The importance of Learning: From Classroom to Real Life.]]></title>
            <link>https://rosekawila.medium.com/the-importance-of-learning-from-classroom-to-real-life-349f28c16bde?source=rss-39a870992daa------2</link>
            <guid isPermaLink="false">https://medium.com/p/349f28c16bde</guid>
            <category><![CDATA[education]]></category>
            <category><![CDATA[life]]></category>
            <category><![CDATA[innovation]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[purpose]]></category>
            <dc:creator><![CDATA[Rose Kawila]]></dc:creator>
            <pubDate>Wed, 25 Sep 2024 09:47:41 GMT</pubDate>
            <atom:updated>2024-09-25T09:48:39.712Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*u4cwe5O_2JsQVpnS" /><figcaption>Photo by <a href="https://unsplash.com/@markfb?utm_source=medium&amp;utm_medium=referral">Mark Fletcher-Brown</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>Let’s take a trip down memory lane to our school days, when everyone told us that education was the key to success. I remember in high school, my friends and I often joked about the math formulas we were learning. We wondered, “When will we ever use this?” We laughed about chemistry labs and the red and green litmus paper, thinking they wouldn’t matter in real life.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*af89NxzEEv_CsZv7" /><figcaption>Photo by <a href="https://unsplash.com/@vedranafilipovic?utm_source=medium&amp;utm_medium=referral">Vedrana Filipović</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>Even in university, the jokes continued. We questioned the many math theorems we were studying. This made me wonder: Why do we learn all this?</p><p>What’s interesting is that everything we study is created by people. We learn how computer systems work, but why is that important? Why do we need to know about processors or how reading and writing function? How do we apply this knowledge in everyday life?</p><p>I believe we learn these details so we can use them to improve what already exists and invent new things. This is why we have AI today . It simplifies our lives today because countless minds dedicated themselves to understanding complex systems.</p><p>As we move forward, let’s keep learning, improving, and finding new solutions. Education is not just about knowledge; it’s about making a difference in the world.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=349f28c16bde" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Exploring Extrajudicial Killings in Kenya Through Tableau.]]></title>
            <link>https://rosekawila.medium.com/exploring-extrajudicial-killings-in-kenya-through-tableau-0befeee691d3?source=rss-39a870992daa------2</link>
            <guid isPermaLink="false">https://medium.com/p/0befeee691d3</guid>
            <category><![CDATA[kenya]]></category>
            <category><![CDATA[tableau]]></category>
            <category><![CDATA[extrajudicial-killings]]></category>
            <category><![CDATA[data-analysis]]></category>
            <dc:creator><![CDATA[Rose Kawila]]></dc:creator>
            <pubDate>Wed, 10 Apr 2024 15:38:32 GMT</pubDate>
            <atom:updated>2024-04-10T15:43:20.751Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/612/1*X2WsnSig8VrPmiT37HnP7Q.jpeg" /><figcaption>photo from gettyimages</figcaption></figure><p>In recent years, Kenya has been witness to a disturbing trend of demonstrations that escalate into chaos, resulting in tragic loss of life. Among the casualties are innocent victims struck by stray bullets fired by law enforcement, while others have fallen victim to mysterious disappearances, their bodies later discovered in rivers. The recent demise of bloggers underscores the gravity of the situation, shedding light on the disheartening reality of extrajudicial killings.</p><p>Though I have been fortunate to not experience a direct encounter with extrajudicial killings, the pain of loss resonates deeply with me, especially when it affects individuals at a young age. In my latest project, I delved into the realm of data analysis using Tableau to examine the prevalence of extrajudicial killings in Kenya. Leveraging data sourced from the Missing Voices website, my analysis reveals that these incidents are occurring nationwide, with the most common causes of death being “shot,” “torture,” and “missing.”</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6A2oiUt23yfyEK6Lr_kckQ.png" /></figure><p>Join me in uncovering the insights gathered from this project by exploring the comprehensive analysis available on my GitHub repository.</p><p><a href="https://github.com/RoseKyalo/My-Tableau-Project">GitHub - RoseKyalo/My-Tableau-Project</a></p><p>Together, let’s shine a light on this pressing issue and strive for justice and accountability. Don’t hesitate to share your thoughts and insights in the comments section below.</p><p>#Extr<a href="https://github.com/RoseKyalo/My-Tableau-Project">a</a>judicialKillings #DataAnalysis #TableauProject #JusticeForAll #Kenya</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=0befeee691d3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Transforming Neurological Disorder Detection: A Mother’s Vision for Early Intervention]]></title>
            <link>https://rosekawila.medium.com/transforming-neurological-disorder-detection-a-mothers-vision-for-early-intervention-ede1a8f971b8?source=rss-39a870992daa------2</link>
            <guid isPermaLink="false">https://medium.com/p/ede1a8f971b8</guid>
            <dc:creator><![CDATA[Rose Kawila]]></dc:creator>
            <pubDate>Mon, 26 Feb 2024 03:28:18 GMT</pubDate>
            <atom:updated>2024-02-26T03:28:18.513Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*nBZyFAs4UrBaxtDc" /><figcaption>Photo by <a href="https://unsplash.com/@bellefoto?utm_source=medium&amp;utm_medium=referral">Alex Pasarelu</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>Motherhood has been an incredible journey filled with joy and countless precious moments with my son. Witnessing his milestones, from his first ‘mama’ to his daily discoveries, fills me with immense happiness. However, I’m also aware that not all parents have such smooth experiences. Many babies are born with neurological disorders, which can be challenging to detect and treat, especially in regions with limited healthcare resources.</p><p>In Kenya, the shortage of mental health professionals and pediatric neurologists poses a significant barrier to early diagnosis and intervention. With 75 percent of Kenyans relying on out-of-pocket payments for healthcare, the financial burden of treating neurological disorders can be overwhelming for families.</p><p>Drawing from personal experiences and a deep understanding of these challenges, I’ve developed a concept paper aimed at transforming the landscape of neurological disorder detection and treatment. At the heart of this concept is a baby-friendly tool designed to assess both brain and limb activities at birth. This innovative system will enable early detection of neurological disorders, alerting practitioners and parents via a mobile app for prompt intervention and treatment.</p><p>My goal is to seamlessly integrate this tool into existing healthcare systems, ensuring accessibility and affordability for all. I invite you to explore my concept paper on GitHub(<a href="https://github.com/RoseKyalo/Concept-Paper">https://github.com/RoseKyalo/Concept-Paper</a>) and share your feedback. Together, let’s pave the way for a brighter future where every child has the opportunity to thrive.</p><p>#NeurologicalDisorders #EarlyDetection #HealthcareInnovation #MotherhoodJourney</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ede1a8f971b8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[NABII-DIMBA: Revolutionizing Football Betting with Machine Learning]]></title>
            <link>https://rosekawila.medium.com/nabii-dimba-revolutionizing-football-betting-with-machine-learning-1fc74cf473f8?source=rss-39a870992daa------2</link>
            <guid isPermaLink="false">https://medium.com/p/1fc74cf473f8</guid>
            <category><![CDATA[football]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[betting-tips]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Rose Kawila]]></dc:creator>
            <pubDate>Wed, 14 Feb 2024 02:46:17 GMT</pubDate>
            <atom:updated>2024-02-14T05:36:32.185Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*YAByVn2JTtU8JyC8" /><figcaption>Photo by <a href="https://unsplash.com/@skucinic9?utm_source=medium&amp;utm_medium=referral">Sven Kucinic</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>In the vibrant Kenyan market, football betting stands as one of the most enthusiastic activities among sports enthusiasts. From my days in campus, I vividly recall seeing young men huddled over tiny pieces of paper, pen tucked behind their ears, meticulously jotting down their game predictions. However, these predictions were often based solely on guesswork and instincts.</p><p>Enter NABII-DIMBA, a user-friendly system developed by a dedicated team of six individuals from Moringa School. This innovative platform revolutionizes the world of football betting by harnessing the power of machine learning algorithms. Unlike traditional methods reliant on team statistics or mere guesswork, NABII-DIMBA employs sophisticated machine learning techniques to deliver accurate match predictions.</p><p>The system allows users to select a specific date, upon which all games scheduled for that day are displayed along with predictions for various match outcomes. These predictions include the likelihood of either the home or away team winning, the total goal count for each team, and even predictions on fouls, such as the likelihood of a yellow or red card.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3rVniWmMkymVCFH0gZRTBg.jpeg" /><figcaption>A snapshot of Nabii-Dimba</figcaption></figure><p>So, what sets NABII-DIMBA apart from existing prediction platforms? While other sites may rely solely on team statistics, NABII-DIMBA goes a step further by integrating machine learning algorithms into its predictive analysis. This integration significantly enhances the accuracy of predictions, providing users with invaluable insights into upcoming matches.</p><p>To ensure accessibility and ease of use, NABII-DIMBA has been deployed on Streamlit, a user-friendly platform accessible to all. Users are encouraged to test the system and provide feedback for further improvements. After all, with NABII-DIMBA at your disposal, there’s no need to second-guess your bets — our machine learning-powered predictions have got you covered!</p><p>So, don’t hesitate to dive into the world of football betting with confidence. With NABII-DIMBA leading the way, your next winning bet is just a prediction away!</p><p>Discover NABII-DIMBA on Streamlit: 🚀<a href="https://eplprecisionpredictor.streamlit.app/">https://eplprecisionpredictor.streamlit.app/</a></p><p>Dive into the exciting world of EPL Precision Predictor(Nabii-Dimba) and unleash your betting potential. Explore the entire project on GitHub <a href="https://github.com/waynekipngeno/group_7_capstone">https://github.com/waynekipngeno/group_7_capstone</a> or <a href="https://github.com/RoseKyalo/capstone_project">https://github.com/RoseKyalo/capstone_project</a> for a deeper insight. Don’t hesitate to share your valuable feedback with us at <a href="/service/mailto:rosekyalo94@gmail.com">rosekyalo94@gmail.com</a>. Happy betting and may the odds be ever in your favor! 🎉</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1fc74cf473f8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Unraveling Exciting Features in My Data Science Coursework.]]></title>
            <link>https://rosekawila.medium.com/unraveling-exciting-features-in-my-data-science-coursework-8ba0d522b22a?source=rss-39a870992daa------2</link>
            <guid isPermaLink="false">https://medium.com/p/8ba0d522b22a</guid>
            <category><![CDATA[data-analytics]]></category>
            <category><![CDATA[researchers]]></category>
            <category><![CDATA[academic]]></category>
            <category><![CDATA[business]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Rose Kawila]]></dc:creator>
            <pubDate>Fri, 01 Dec 2023 14:05:02 GMT</pubDate>
            <atom:updated>2023-12-01T14:14:47.816Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*BUBiCDEbvPXAG65c" /><figcaption>Photo by <a href="https://unsplash.com/@aaronburden?utm_source=medium&amp;utm_medium=referral">Aaron Burden</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>Welcome back to the world of writing! 📝 My Data Science course has been an absolute whirlwind, and I’m thrilled to say I’m nearing the finish line! Let me take you through some of the amazing features that have made this journey so exhilarating:</p><h4>1. Folium Map</h4><p>The Folium package has been a game-changer! It’s empowered me to craft interactive maps and integrate them with API requests. In one project (find it on my GitHub <a href="https://github.com/RoseKyalo/PHASE-2-PROJECT">here</a>), we visualized house prices distribution, clustering them to identify areas with the highest prices.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GhPGGwdCd2LpY1t_wNr8Dg.png" /></figure><h4>2. Seaborn</h4><p>Seaborn, another gem, has enabled me to create captivating plots. Take a peek at a simple visual I crafted using Seaborn!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/851/1*ihMe4nIx2LA7ExXQ_c2JPg.png" /></figure><h4>3. API request</h4><p>One line of code — that’s all it takes to extract website content through API requests. The thrill came from scraping data and exploring its potential.</p><pre>import os<br>import googleapiclient.discovery<br>import pandas as pd<br><br># Set your API key<br>api_key = &quot;api key&quot; <br><br># Create a YouTube API client<br>youtube = googleapiclient.discovery.build(&quot;youtube&quot;, &quot;v3&quot;, developerKey=api_key)<br><br># Specify the channel username<br>channel_username = &quot;name of the channel&quot; <br><br><br># Get the channel&#39;s uploads playlist ID<br>channel_request = youtube.channels().list(<br>    part=&quot;contentDetails&quot;,<br>    forUsername=channel_username<br>)<br><br>channel_response = channel_request.execute()</pre><p><strong>4. Building Models</strong></p><p>The magic of classifying images! We delved into creating models that could identify hidden patterns within a collection of images, even for people.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/262/1*RzxsOeo6sRtQqDZcOfFTZA.jpeg" /><figcaption>from pythonprogramming.net</figcaption></figure><p><strong>5. Word count plot</strong></p><p>Ever wondered about the most common words used in YouTube comments or any text corpus? With this feature, I’ve been able to build plots showcasing the most frequently used words.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/286/1*9aqETf02xTxIaBkuR9v3aw.jpeg" /><figcaption>from bookdown.org</figcaption></figure><p>As I finish my coursework, I’m eager to use these skills in the real world! The prospect of putting these tools into practice in practical settings is exhilarating! Happy reading and stay tuned for more adventures as I journey into applying these skills beyond the classroom.</p><p>🌟 #DataScienceJourney #PracticalApplications #ExcitedForWhatsNext</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8ba0d522b22a" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>