<?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 Fokoye on Medium]]></title>
        <description><![CDATA[Stories by Fokoye on Medium]]></description>
        <link>https://medium.com/@fokoye?source=rss-68ed040cd594------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*zrfd5WQQanKoCicG.jpg</url>
            <title>Stories by Fokoye on Medium</title>
            <link>https://medium.com/@fokoye?source=rss-68ed040cd594------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 19:57:28 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@fokoye/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[The Future of Coding and Data Science in the Age of AI]]></title>
            <link>https://medium.com/@fokoye/the-future-of-coding-and-data-science-in-the-age-of-ai-cb86c6d3b3f3?source=rss-68ed040cd594------2</link>
            <guid isPermaLink="false">https://medium.com/p/cb86c6d3b3f3</guid>
            <dc:creator><![CDATA[Fokoye]]></dc:creator>
            <pubDate>Thu, 20 Mar 2025 04:05:06 GMT</pubDate>
            <atom:updated>2025-03-20T04:05:06.955Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/299/1*_Fxhn8cxiP7d2iXhfberHQ.jpeg" /></figure><p><strong>The Future of Coding and Data Science in the Age of AI</strong></p><p>For most AI end-users, writing code may never be necessary. AI will automate a lot of routine tasks, from data visualization to predictive modeling. But the creators of these tools, platforms, and intelligent agents will always be individuals with a deep understanding of programming, algorithms, and data structures.</p><p>AI itself is heavily dependent on the work of software developers and data scientists. These professionals fine-tune models, create APIs, manage data flows, and build the interfaces and backend systems that power AI tools. Even as AI gets better at generating code or building basic applications, humans are still needed to verify outputs, design systems architecture, and solve complex problems that AI can’t yet handle on its own.</p><p>Many companies eager to implement AI solutions are quickly realizing they lack the proper data infrastructure. AI systems require high-quality, well-organized, and accessible data to perform effectively. This has created a growing demand for data engineers who can build pipelines, manage databases, and prepare data for machine learning models. This implies that AI can’t function without the people behind the scenes.</p><p>While no-code and low-code tools now allow users to automate tasks, build applications, and even create AI workflows without writing traditional code, there is a limit to what these tools can do. Organisations that aim to develop scalable, highly customized, or enterprise-grade AI solutions still need programmers and data engineers. Whether it’s building proprietary AI models, integrating AI with complex systems, or ensuring security and performance at scale, coding remains indispensable. Currently, with advancements in cloud computing, open-source libraries, and generative AI tools, small teams of engineers can now build and deploy sophisticated AI products at a fraction of the cost and time it used to take. This is fueling a wave of AI-driven startups and products emerging from lean, agile teams.</p><p>With the advancement of AI and no-code platforms, <strong>many aspiring data professionals are starting to ask, Do I still need to learn how to code?</strong> It’s a fair question, especially as AI tools become more powerful and user-friendly. The short answer is yes, coding and data skills are still highly valuable, more than ever, depending on your career goals.</p><p>If you’re considering a future in AI, data science, or any related field, learning to code is still a smart investment. Even if you plan to focus on using AI tools rather than building them, understanding how they work under the hood will give you a competitive edge.</p><p>Basic programming and data literacy will allow you to:</p><ul><li>Customize AI tools more effectively</li><li>Understand AI limitations and outputs</li><li>Collaborate with technical teams</li><li>Automate and streamline your workflows</li><li>Make informed decisions based on data and AI insights</li></ul><p>Rather than replacing coders and data professionals, AI is reshaping their roles. In the coming years, coding will evolve to become more collaborative with AI. Tools like AI-assisted development environments and AI pair programmers will help engineers code faster and with fewer errors, but the <strong>Critical thinking, Creativity, and Domain expertise that humans bring will remain irreplaceable.</strong></p><p>At the same time, data science will continue to thrive as organizations strive to extract value from the vast amounts of data they generate. Data engineers, data scientists, and machine learning engineers will be in high demand to build, optimize, and interpret AI systems.</p><p><strong>Conclusion</strong></p><p>AI is democratizing access to technology, but if you want to be part of the teams building the next generation of AI-powered products, coding and data skills are still essential. Whether you aspire to work in AI research, software engineering, or advanced data analytics, investing time in learning programming and data fundamentals will open doors to more impactful and rewarding roles. AI will amplify what you can do but understanding code and data will make sure you can do more, and do it better</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cb86c6d3b3f3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[WHY EVERY MANAGER SHOULD HAVE DATA ANALYTICS SKILLSETS]]></title>
            <link>https://medium.com/@fokoye/why-do-managers-need-data-analytics-e7feadc0171f?source=rss-68ed040cd594------2</link>
            <guid isPermaLink="false">https://medium.com/p/e7feadc0171f</guid>
            <dc:creator><![CDATA[Fokoye]]></dc:creator>
            <pubDate>Fri, 22 Dec 2023 02:00:35 GMT</pubDate>
            <atom:updated>2023-12-22T10:21:08.033Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZvK7Zvzi6FzU7m7o_7xsog.jpeg" /></figure><p><strong>Introduction</strong></p><p>Data analytics plays a pivotal role in empowering managers with the tools and insights necessary for effective decision-making within organizations. In today’s data-driven world, managers are dealing with vast amounts of data from various sources. Data analytics enables managers to process these data, identify the trends and patterns in order to extract actionable insights. By harnessing data analytics skills and techniques such as descriptive analytics, managers can gain a comprehensive understanding of historical trends and performance metrics, providing a solid foundation for informed decision-making. Moreover, predictive analytics equips managers with the ability to anticipate future trends and outcomes, enabling proactive strategies and risk mitigation. This foresight allows managers to navigate uncertainties with confidence, aligning their decisions with the organization’s strategic goals. In addition to enhancing decision-making through trend analysis and predictive modelling, data analytics facilitates a more agile and responsive managerial approach. Managers can leverage interactive dashboards and visualizations to gain immediate insights, facilitating a more dynamic decision-making process. As a result, data analytics not only aids in strategic, long-term planning but also supports day-to-day operational decisions, fostering a culture of adaptability and responsiveness among managerial teams.</p><p><strong>What is Data Analytics?</strong></p><p><strong>Data analytics</strong> can simply be seen as the process of examining data in order to find trends, patterns and draw insights from the data. Data analytics is a way of using technology and tools to collect, organize, and analyze large amounts of data to find patterns, insights, and trends. It involves using mathematical and statistical techniques to make sense of data and turn it into actionable insights.</p><p>Data analytics is like a giant puzzle. Imagine a big pile of puzzle pieces representing information about something you are interested in. Data analytics is the process of taking all those puzzle pieces and putting them together to see what picture they make.</p><p>A use case, a manager at a video game store wants to know what people think about a new video game(Call of Duty). He could collect data by surveying a group of people and asking them to rate the game on a scale from one to ten. Once he has the data, he can use data analytics tools to analyze the responses and find out which features of the game people like and dislike the most. By analyzing the data, he can learn new things and make decisions based on the findings. Therefore, data analytics helps managers put data together to learn something new(insights). The data analytics process is all about studying historical and current data using statistical modeling, algorithms, and data visualization to gain valuable insights, detect trends, and help make informed decisions about your future direction.</p><p>It is important to note that in data analytics, requires you have a very good domain knowledge, basic knowledge in mathematics and statistics (minimum, maximum, average, median, count, probabilities, algebra, matrix, correlation, regression) and also a bit of IT skills (basic programming using Excel, SQL, Python or R)</p><p><strong>Who Needs Data Analytics?</strong></p><p>Data analytics is needed by various individuals, organizations, and industries that deal with large amounts of data and require insights to make informed decisions. Everybody who makes business decisions needs to use data analytics. As a result of development in technology we now have access to data more than ever before. Any manager who formulating strategies and making decisions without considering the data available , could miss the insights on the opportunities or red flags that it communicates. Managers who can benefit from data analytics skills include: Some specific examples of how HR, marketing, sales, finance, and operations can use data analytics:</p><p><strong>Human Resources (HR):</strong> HR departments can use data analytics to identify patterns in employee data related to performance, retention, and engagement. For example, by analyzing employee data, HR can identify factors that contribute to employee turnover and develop retention strategies to address them. They can also use data analytics to identify skills gaps and develop training programs to improve employee performance. The application of data analytics tools and techniques in the domain of HR is referred to as <strong>HR Analytics or Peoples Analytics.</strong></p><p><strong>Marketing:</strong> Marketing teams can use data analytics to identify customer behaviors and preferences, such as purchase patterns, social media engagement, and website traffic. By analyzing this data, they can create targeted campaigns and optimize marketing strategies to improve customer engagement and increase sales. The application of data analytics tools and techniques in the domain of marketing is referred to as <strong>Marketing Analytics.</strong></p><p><strong>Sales:</strong> Sales teams can use data analytics to identify patterns in customer data, such as purchase history and demographics, to identify potential customers and improve sales forecasting. They can also use data analytics to analyze sales performance and identify areas for improvement. The application of data analytics tools and techniques in the domain of sales is referred to as <strong>Sales Analytics.</strong></p><p><strong>Finance:</strong> Finance departments can use data analytics to identify patterns in financial data related to revenue, expenses, and cash flow. For example, they can use data analytics to identify trends in customer payments, optimize pricing strategies, and reduce costs. The application of data analytics tools and techniques in the domain of finance is referred to as <strong>Financial Analytics.</strong></p><p><strong>Operations:</strong> Operations teams can use data analytics to identify patterns in production data related to efficiency, quality, and safety. For example, they can use data analytics to identify bottlenecks in production, optimize inventory management, and improve safety procedures. The application of data analytics tools and techniques in the domain of operations is referred to as <strong>Operations Analytics.</strong></p><p>Overall, analytics can be used in many ways across various departments and industries to improve decision-making, optimize processes, and drive business growth.</p><p><strong>Benefits of Data Analytics for Managers</strong></p><p>Organizations that use data to make decisions are three times more likely to say that these decisions have improved significantly in terms of innovation, growth, and competitive advantage. Organizations in a variety of industries can benefit from data analytics in a number of ways. Some of the main advantages of data analytics are listed below.</p><p><strong>Better decision-making</strong>: Data analytics can help businesses make informed decisions based on real-time insights and data-driven evidence. This can help companies identify new opportunities, avoid potential risks, and make strategic decisions that drive growth and profitability.</p><p><strong>Improved efficiency:</strong> Data analytics can help businesses optimize their operations and cut costs by automating data collection and analysis. Businesses may be able to streamline their procedures and utilize their resources more effectively as a result, increasing productivity and efficiency.</p><p><strong>Enhanced customer experiences:</strong> Data analytics can help businesses better understand their customers’ preferences and behaviors, enabling them to tailor their products, services, and marketing campaigns to better meet their customers’ needs. This can result in higher customer satisfaction and loyalty, and ultimately drive sales growth.</p><p><strong>Competitive advantage:</strong> Businesses can gain a competitive advantage by using data analytics to find new opportunities, anticipate market trends, and make more informed decisions. Businesses can use this to set themselves apart from the competition and become market leaders.</p><p><strong>Improved risk management:</strong> Using data analytics to identify possible hazards and weaknesses in their operations enables businesses to take proactive measures to mitigate risk. This can help businesses limit losses, avert reputational damage, and ensure regulatory compliance.</p><p><strong>Budgeting and forecasting</strong>: Data analytics can be used to determine the budget and investments necessary to achieve a company’s future growth goals by analyzing historical revenue, sales, and cost data alongside those goals.</p><p><strong>Risk management: </strong>By understanding the likelihood of certain business risks occurring and their associated cost, an analyst can make cost-effective recommendations to help mitigate them.</p><p><strong>Marketing and sales</strong>: Data analytics can determine how many leads their efforts need to generate in order to fill the sales pipeline by understanding key metrics like lead to customer conversion rate.</p><p><strong>Product development (or research and development):</strong> Data Analytics can direct future product development, design, and user experience by analyzing how customers reacted to product features in the past.</p><p><strong>How Managers Use Data Analytics</strong></p><p>Data analytics has become increasingly important in the business world, and managers need to have a basic understanding of the principles of data analytics to make informed decisions. In this case study, we will examine how a manager in a retail company can use data analytics to make decisions related to inventory management.</p><p><strong>Case Study</strong></p><p><em>Sarah is the manager of a retail store that sells clothing and accessories. She is responsible for managing the inventory levels of the store and ensuring that the store has the right amount of inventory to meet customer demand. However, she is finding it difficult to make decisions about inventory management based on her intuition and experience alone. She has heard about data analytics and wants to explore how it can help her make better decisions. Sarah took a course on Data Analytics. She started by analyzing the historical sales data of the store to identify trends in customer demand. She looked at the sales data for the last two years and used a data visualization tool to create charts to help her visualize the data. This analysis helped her to identify the products that were selling well and the products that were not selling as well. Next, Sarah used a forecasting model to predict the future demand for each product. She used the historical sales data to train the model and used it to predict the demand for the next three months. This helped her to identify which products were likely to sell well in the future and which products were not.</em></p><p><em>The analysis revealed that certain products were selling well during specific seasons or events. During the winter clothing and accessories sold well during the winter months, while summer clothing and accessories sold well during the summer months. Sarah was able to use this insights to adjust the inventory levels of the store to meet customer demand. The forecasting model helped Sarah to predict the future demand for each product. This allowed her to order the right amount of inventory to meet customer demand without overstocking or understocking the store.</em></p><p><em>By using data analytics, Sarah was able to make informed decisions about inventory management. She was able to identify trends in customer demand, predict future demand, and adjust inventory levels accordingly. This helped her to ensure that the store had the right amount of inventory to meet customer demand, while also minimizing the costs associated with overstocking or understocking the store. As a manager, having a basic understanding of data analytics can be extremely valuable in making informed decisions that can lead to improved performance and profitability.</em></p><p><strong>Types of Data Analytics</strong></p><p>There are generally four types of data analytics: Descriptive Analytics, Diagnostic Analytics, Predictive Analytics, and Prescriptive Analytics</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/602/1*7ZJYu1wKpfAyL98cA-8mMA.png" /></figure><p><em>source: Gartner Analytics Ascendancy Model (GAAM, 2019).</em></p><p><strong>Descriptive Analytics:</strong> Descriptive analytics is the simplest type of analytics, allowing you to pull trends from raw data and describe what happened or is currently happening. It involves analyzing historical data to understand what has happened in the past, answering the question <strong>“What happened?”</strong></p><p><em>The sales manager could use descriptive analytics to analyze historical sales data and gain insights into sales trends and patterns. For example, they could look at sales by product, by region, by salesperson, and by month or quarter. By doing so, they can identify which products are selling the most, which salespeople are performing the best, and which regions have the highest sales growth.</em></p><p><strong>Diagnostic Analytics:</strong> The next logical question, <strong>“Why did this happen?”</strong> is answered by diagnostic analytics. This type of analytics goes a step further by comparing concurrent trends or movements, finding correlations between variables, and, when possible, establishing causal linkages. Analytical diagnostics are helpful for identifying the underlying causes of organizational problems.</p><p><em>Once the sales manager has identified patterns in the data using descriptive analytics, they can use diagnostic analytics to understand the root causes of those patterns. For example, if they notice that sales are declining in a particular region, they can investigate why. They might find that a competitor has entered the market, or that there’s been a change in consumer preferences. By identifying the cause of the decline, they can take appropriate action to address the issue.</em></p><p><strong>Predictive Analytics: </strong>Predictive analytics tries to answer the question of <strong>what might happen in the future?</strong> You can accurately estimate what the future may hold for your firm by examining historical data along with current industry trends. Forecasting, statistical models and machine learning algorithms are used in predictive analytics to assess historical data and forecast future events. Finding patterns and trends aids in making predictions about the future. Among other things, predictive analytics is used to forecast market trends, sales, and demand. Making forecasts about the future might assist your company in developing plans based on probable outcomes. Using prescriptive analytics, we can determine what to do next.</p><p><em>Using predictive analytics, the sales manager can forecast future sales trends based on historical data and other factors. For example, they could use predictive analytics to forecast sales for the next quarter or year. This would enable them to make better-informed decisions about inventory, staffing, and marketing. They could also use predictive analytics to identify which customers are most likely to make a purchase, so that they can focus their efforts on those customers.</em></p><p><strong>Prescriptive Analytics:</strong> Prescriptive analytics is the most advanced form of data analytics. It involves using machine learning algorithms and artificial intelligence to analyze data and provide recommendations for actions to take. Prescriptive analytics proposes actionable insights after accounting for all potential aspects in a scenario.</p><p><em>The sales manager could use prescriptive analytics to make data-driven recommendations for how to improve sales performance. For example, they could use prescriptive analytics to identify which products are most likely to sell well in a particular region, or which sales strategies are most effective with different types of customers. By making these recommendations, they can help their sales team make better-informed decisions and ultimately drive sales growth.</em></p><p><strong>Data Analytics Tools for Managers</strong></p><p>Data analytics tools are essential for managers in today’s data-driven business environment. These tools enable managers to make informed decisions by providing insights into customer behavior, market trends, and operational efficiency. Some of the data analytics tools that managers need to be familiar with:</p><p>• <strong>Microsoft Excel</strong>: The first tool you will need to learn is Microsoft Excel. Excel is the number one data analytics tool. Excel can be used to analyze data, from simple data entry to complex data modeling. Excel provides tools to sort, filter, and manipulate data.</p><p>Excel is a valuable data analytics tool that allows managers to analyze <strong>past</strong> data by creating reports and doing analytics, as well as analyze the <strong>future</strong> by building models and making forecasts/predictions. Managers can use Excel to analyze sales data, customer feedback, and financial statements.</p><p>• <strong>SQL (MySQL/Microsoft SQL Server)</strong>: One of the task in data analytics is data extraction. Most times in the organization, the data isn’t stored in excel due to the limitation of the volume of data excel can handle. Managers need to have some basic knowledge of SQL to communicate with the database.</p><p>• <strong>Power BI</strong> — This is a business intelligence tool that supports multiple data sources, helps in asking questions and getting immediate insights.</p><p>• <strong>Tableau</strong> — This helps in creating several kinds of visualizations for presenting insights and trends in a better way.</p><p>• <strong>Python</strong> — This object-oriented open-source programming language is used for manipulating, visualizing, and modelling data.</p><p>• <strong>R</strong> — An open-source programming language used in numerical and statistical analysis.</p><p><em>To begin your journey in data analytics, it is recommended to start with Excel first, followed by SQL(MySQL), Power BI, and then maybe Python if you really want to build your programming skills.</em></p><p><strong>CRISP-DM FRAMEWORK</strong></p><p>The CRISP-DM framework provides a structured approach to data analytics, by helping organizations to maximize the value of their data and make informed decisions based on data insights.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/464/1*TnXDn037M3ARU_eCI7-RNQ.png" /></figure><p><em>Figure 1: Phases of Cross-Industry Standard Process for Data Mining</em></p><p><strong>CRISP-DM stands for Cross-Industry Standard Process for Data Mining</strong>, and it is a widely used framework for data mining and analytics. It consists of six phases:</p><p><strong>Business Understanding</strong>: In this phase, the focus is on understanding the business problem or opportunity that needs to be addressed. The goal is to identify the objectives and requirements of the project and define the scope of the project.</p><p><strong>Data Understanding:</strong> In this phase, the focus is on gathering and exploring the data that is needed to solve the business problem or address the opportunity identified in the first phase. The goal is to assess the quality of the data and identify any gaps or limitations that need to be addressed.</p><p><strong>Data Preparation:</strong> In this phase, the focus is on cleaning, transforming, and preparing the data for analysis. The goal is to create a high-quality dataset that can be used for analysis.</p><p><strong>Modeling:</strong> In this phase, the focus is on building and testing the models that will be used to analyze the data. The goal is to identify the best model that meets the project’s objectives and requirements.</p><p><strong>Evaluation:</strong> In this phase, the focus is on evaluating the model’s performance and assessing its effectiveness in addressing the business problem or opportunity identified in the first phase. The goal is to ensure that the model meets the project’s objectives and requirements.</p><p><strong>Deployment:</strong> In this phase, the focus is on deploying the model and integrating it into the organization’s systems and processes. The goal is to ensure that the model is being used effectively and that it continues to meet the project’s objectives and requirements.</p><p><strong>Data Analytics vs Reporting</strong></p><p>Both data analytics and reporting deal with data, they do so in different ways. The simplest way to understand it is that reporting shows what is happening while data analytics explains why it is happening.</p><p>Reporting entails gathering data to gain a better understanding of the performance of a company’s various functions. Analytics is the ability to interpret data at a deeper level to make better decisions. For example, a company can use reporting to evaluate progress on various marketing campaigns, then use analytics to better prioritize them.</p><p>Reporting generates facts about the performance of different functions within a company. Examples of these functions can include sales, marketing, production, and so on. Outputs usually follow a structured reporting template and timeline (e.g. daily, weekly, monthly, quarterly or annual reports). Analytics takes it a step further and provides insights, detects patterns, and generates recommendations</p><p>Therefore, the major difference is that reporting is about extracting, organizing, and consolidating the data from the various sources. Analytics, on the other hand, is concerned with examining, questioning, comparing, and identifying trends from the data.</p><p><strong>Data Analytics System</strong></p><p>A data analytics system is a robust tool that allows its users to interpret data to help them make data-driven decisions. The process should be automated, pull data from several sources within your company, and provide a rounded picture of the overall health of the company. It bridges the gap between siloed data in the various departments within your organisation, it also helps simplify the data reporting and analysis process to save employees time.</p><p><strong>Data Analytics System vs Reporting System</strong></p><p>A data analytics system is not the same as an operating or reporting system. A reporting system is also called an Operating System (OS).Many software vendors sell operational systems (OS) which have some analytics features. But these analytics features are not analytics systems , they are just reporting on siloed data collected by that specific operating system which tells you what is happening. On the other hand, analytics systems combine data from multiple sources. The data from various operating systems and other data sources is the key part of this system, the relationships between these data sources give better insights on why things happen, also giving richer insights to aid the decision making process. The richer and more diverse the data collected about a transaction, the more contextual the information the more powerful your analytic system will be.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/408/1*MSbnb5J1AabaMJI5N6ILNw.png" /></figure><p>Traditionally, organizations data has been guarded by the IT department, serving as the gatekeeper to ensure the security and integrity of company data. This approach is logical and imperative for safeguarding sensitive data. However, there is a growing recognition among companies that fostering alignment across various departments is not only crucial for data security but is also a key driver for overall business growth. In this evolving paradigm, companies understand that a cohesive approach to data management is vital for achieving synergy across different business units. The traditional notion of compartmentalizing data within specific departments is giving way to a more integrated and interconnected model where data becomes a shared resource. This approach facilitates a holistic understanding of the company’s operations, thereby fostering an environment conducive to sustainable business growth.</p><p>The ownership of sales reporting can vary depending on the organizational structure, priorities, and the specific needs of the business. In many cases, a collaborative approach involving both the Head of Analytics and the Head of Sales and Marketing is ideal. It can improve customer experience and provide a 360-degree view of the customer journey, Increase decision-making speed, optimize each and every person’s role at the company and also help each person and team understand their role in the eyes of the customer. Ultimately, the key is to create a symbiotic relationship between the analytical capabilities of the Head of Analytics and the practical business acumen of the Head of Sales and Marketing, ensuring that sales reporting serves the organization’s overarching goals.</p><p>The IT departments should be able to own the OS where data is generated. Analytics systems, on the other hand, should be owned by the users, the people who collect that data day in and day out and own the process of evaluating that data. Think about departments like HR, finance and accounting, sales, product development, and so on as the C-suite executives requires the bigger picture.</p><p>Executives need to have access to the analytic systems in order to see what is happening throughout the entire organisation. When data analytics are done this way, it’s an all-around win for your organization. IT no longer has to waste time downloading and sending out data that’s requested by the individual units. Siloed departments don’t have to keep asking the same questions again and again. Also company executives can have a birds-eye view of the entire business in real-time — ultimately leading to better business decisions. Therefore, the data analytics systems should be accessed by everyone in the organization.</p><p><strong>What Should a Data Analytics System Be Able to Tell Managers?</strong></p><p>Data analytics system should be able to give you as a manager some insights about your business. This information will be different across your departments, for example:</p><p><strong>Your sales analytics system should be able to tell you:</strong></p><p>· <em>What customers are buying,</em></p><p>· <em>Sales by Demographics: gender, age, race, etc.</em></p><p>· <em>When and where they are buying that product and on what device.</em></p><p>· <em>Detailed comparisons over time and targets by multiple categorical subdivisions.</em></p><p><strong>Your HR analytics system should be able to tell you:</strong></p><p>· <em>Who is performing &amp; who might not be in the coming months.</em></p><p>· <em>The overall level of engagement &amp; retention.</em></p><p>· <em>About time tracking and workforce productivity.</em></p><p>· <em>The effectiveness of training, etc.</em></p><p><strong>Your accounting analytics system should be able to tell you:</strong></p><p>· <em>Predictive analytics, like financial projections, based on sales data.</em></p><p>· <em>Where you might be wasting money.</em></p><p>· <em>Core financial ratios and their trend over time.</em></p><p>· <em>Cost-benefit analysis.</em></p><p><strong>Your overall organisational dashboard should be able to:</strong></p><p>· <em>Show visually what is working within and outside of the company.</em></p><p>· <em>Show the C-suite the highs and lows by multiple metrics.</em></p><p>· <em>Answer specific and complex business questions.</em></p><p><strong>Challenges and Limitations of Data Analytics</strong></p><p>While data analytics offers many benefits, there are also several challenges and limitations that organizations should be aware of. Some of the main challenges and limitations of data analytics:</p><p><strong>Data quality:</strong> The quality of the data used for analytics can be a challenge, as it may be incomplete, inaccurate, or inconsistent. The data quality can affect the accuracy and reliability of the insights and decisions based on the data.</p><p><strong>Data privacy and security:</strong> These concerns can be a significant challenge for data analytics, especially in industries that handle sensitive data. Organizations need to ensure that they comply with privacy regulations and protect data from unauthorized access and cyber threats.</p><p><strong>Resource requirements:</strong> Implementing data analytics can require significant resources, including specialized talent, technology, and infrastructure. Organizations need to ensure that they have the resources and budget necessary to support a successful implementation.</p><p><strong>Complexity:</strong> Data analytics can be a complex process, involving multiple stakeholders, technologies, and data sources. Organizations need to manage this complexity effectively to ensure that the insights and decisions are based on accurate and meaningful data.</p><p><strong>Bias and interpretation:</strong> Organizations must be aware of these potential biases and ensure their analytics efforts are transparent and objective.</p><p><strong>Legal and ethical considerations:</strong> The use of data analytics can raise legal and ethical considerations, such as data privacy, intellectual property, and discrimination. Organizations need to ensure that they comply with relevant laws and ethical standards when collecting, using, and sharing data.</p><p><strong>Conclusion</strong></p><p>The ability to distill meaningful insights from data empowers managers to make informed decisions, navigate complexities, and align strategies with organizational objectives. From descriptive analytics to prescriptive analytics, data analytics provides a multifaceted lens through which managers can understand and anticipate trends, fostering a proactive and strategic decision-making culture. Managers play a pivotal role in fostering a data-driven culture within an organization, and their influence is instrumental in driving this cultural shift from the top down. By embracing data analytics, managers should cultivate a culture of adaptability and responsiveness among managerial teams, ensuring they are well-equipped to thrive in an environment where data-driven insights are integral to the success of the organisation.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e7feadc0171f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A COMPREHENSIVE GUIDE TO DATA CLEANING FOR DATA ANALYSTS AND DATA SCIENTISTS]]></title>
            <link>https://medium.com/@fokoye/a-comprehensive-guide-to-data-cleaning-for-data-analysts-and-data-scientists-d4844c32fd41?source=rss-68ed040cd594------2</link>
            <guid isPermaLink="false">https://medium.com/p/d4844c32fd41</guid>
            <category><![CDATA[data-cleaning]]></category>
            <category><![CDATA[data-wrangling]]></category>
            <category><![CDATA[data]]></category>
            <dc:creator><![CDATA[Fokoye]]></dc:creator>
            <pubDate>Mon, 20 Nov 2023 15:04:40 GMT</pubDate>
            <atom:updated>2024-01-11T09:20:50.824Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/612/1*mi1u_dsGb6OpNFrwQk0w1A.jpeg" /></figure><p><strong>Introduction</strong></p><p>Data has always been an asset, but it has experienced exponential growth in recent years. The main reason behind this is the technology that makes it easier than ever to collect, transfer, and store data. However, the raw data is usually not ready-to-use and requires lots of cleaning and transformations. Unfortunately , we find that 80–90% of the time we spend in cleaning and transforming data, which is the most important steps that lay the groundwork for extracting valuable insights. Dealing with messy or incomplete datasets, handling outliers, and structuring data for analysis are essential components of the data analyst’s /data scientist responsibility. This process is not only about correcting errors but also about shaping the data to align with the specific analytical goals and objectives. Another similar concept is <strong>Data wrangling </strong>which is an iterative and often creative process that requires a deep understanding of both the data and the objectives of the analysis. Successful data wrangling lays the foundation for robust and meaningful insights, enabling more effective decision-making and accurate model training. Data cleaning is an indispensable step in the data analysis and machine learning pipeline. By addressing outliers and missing data effectively, analysts and data scientists can ensure that their models and insights are built on a solid foundation of clean and reliable data. Adopting a systematic approach to data cleaning enhances the quality of analyses, improves the performance of machine learning models, and ultimately contributes to more accurate and trustworthy results.</p><p><strong>Why Data Cleaning Matters</strong></p><p>Data cleaning, also known as data preprocessing, is a critical step in the data analysis and data science process. Its importance lies in several key aspects that directly impact the quality, accuracy, and reliability of the analyses and models developed by data analysts and data scientists. Data cleaning is a foundational step in the data analysis and data science lifecycle. It ensures that the data used for analysis is of high quality, leading to more accurate, reliable, and actionable insights. Some reasons why data cleaning is important:</p><p><strong>Ensures Data Accuracy:</strong> Data may contain errors, inconsistencies, or inaccuracies due to various reasons, such as human error during data entry or system issues. Cleaning the data helps identify and rectify these issues, ensuring the accuracy of the information being analyzed.</p><p><strong>Improves Model Performance:</strong> For data scientists working on machine learning models, the quality of input data significantly affects the performance of the models. Clean and well-structured data can lead to more accurate and reliable predictions or classifications.</p><p><strong>Facilitates Effective Analysis:</strong> Clean data simplifies the analysis process. Analysts can spend more time extracting meaningful insights rather than dealing with outliers, missing values, or inconsistent formatting.</p><p><strong>Enhances Data Consistency:</strong> Incomplete or inconsistent data can lead to confusion and misinterpretation. Data cleaning helps ensure that data is consistent across all fields, making it easier to analyze and draw valid conclusions.</p><p><strong>Reduces Bias and Errors:</strong> Biases in the data, whether due to sampling issues or systematic errors, can affect the results of analyses and models. Cleaning the data helps identify and mitigate biases, reducing the risk of drawing incorrect conclusions.</p><p><strong>Mitigates the Impact of Missing Data:</strong> Missing data is a common issue, and how it’s handled can impact the validity of analyses. Data cleaning involves strategies for handling missing data, such as imputation or exclusion, to minimize its impact on the results.</p><p><strong>Ensures Compatibility Across Datasets:</strong> When combining data from multiple sources, cleaning is necessary to ensure that data is compatible and can be effectively integrated. This is crucial for creating comprehensive analyses or building models that incorporate diverse datasets.</p><p><strong>Increases Stakeholder Trust:</strong> Decision-makers and stakeholders rely on data-driven insights to make informed choices. Cleaned and well-prepared data increases the trustworthiness of the results, fostering confidence in the analyses and recommendations provided by data analysts and scientists.</p><p><strong>Saves Time and Resources:</strong> Cleaning the data upfront can save time in the long run. Data Analysts and Data scientists can avoid repeatedly addressing data issues during the analysis process, streamlining their workflow and making more efficient use of resources.</p><p><strong>What is Missing Data?</strong></p><p>Missing data refers to the absence of values in a dataset for certain observations or variables. Missing data can occur for various reasons, including data entry errors, system issues, or intentional non-responses. Dealing with missing data is a crucial step in the data cleaning process, as it can impact the quality and reliability of analyses or models. Missing data can be categorized as either <strong>systematic</strong> or <strong>random</strong>, and understanding the nature of missing data is crucial in deciding on the appropriate strategies for handling the missing data.</p><p><strong>Systematic Missing Data: </strong>Systematic missing data occurs in a structured manner. There is a clear pattern or reason behind why certain values are missing.<strong> </strong>Example<strong> </strong>If a survey includes a question that respondents find uncomfortable, leading to a consistent refusal to answer, the missing values would be systematic.</p><p><strong>Random Missing Data</strong>: Random missing data occurs in an unstructured manner. There is no clear pattern or reason behind why certain values are missing. For example, in a survey, if respondents occasionally skip questions without a consistent pattern, the missing values would be considered random.</p><p>Your first step in handling missing data is to go back to the source/department that generated the data to ask for the updated/complete data. If this will not be possible, the following strategies for handling missing data maybe used:</p><p>· <strong>Identify Missing Data:</strong> Before addressing missing data, it’s essential to identify where and how much data is missing. Use descriptive statistics or visualization tools to understand the extent of the issue.</p><p>· <strong>Remove Columns with High Missing Values:</strong> If a particular variable has a large proportion of missing values and isn’t critical for analysis, you may consider removing the entire column.</p><p>· <strong>Impute with Mean, Median, or Mode:</strong> For numerical variables, you can replace missing values with the mean or median. Use the mean if there are no outliers in the data, if there are outliers in the data then use the median. For qualitative/categorical variables use mode(Highest Frequency) of that variable.</p><p>· <strong>Impute with Predictive Modeling:</strong> Use advanced techniques, such as regression models, to predict missing values based on other variables. This is particularly useful when missing data is related to the values of other variables in the dataset.</p><p>· <strong>Create a Missing Indicator:</strong> Instead of imputing missing values, you can create a binary indicator variable that flags whether a value was missing or not. This way, the fact that data was missing is retained as information.</p><p>· <strong>Use Fill Down/Fill Up:</strong> If the missing data is systematic, you can use fill down/fill up, but be sure it fills with the right data.</p><p>· <strong>Consideration of Business Policy/Business Rules/Domain Knowledge</strong>: Sometimes, domain knowledge or business policy/rules can guide the imputation strategy. For example, in financial datasets, a missing value for income might be imputed differently than in a healthcare dataset.</p><p>· <strong>Remove Rows with Missing Values:</strong> One straightforward approach is to remove entire rows that contain missing values. This is suitable when the missing values are limited and removing those observations doesn’t significantly impact the analysis. This should be your last option.</p><p><strong>What is an Outlier?</strong></p><p>An outlier is an observation(data point) that lies outside the normal range of your data. These anomalies can arise due to errors in data collection, measurement noise, or rare events. That is, outliers could arise either by nature or by error. Also, outliers can be identified as unusually high or low values can distort your statistical analyses and machine learning models, which makes it essential to identify and handle them appropriately.</p><p><strong>How to Find Outliers</strong></p><p><strong>Descriptive Statistics:</strong> Basic descriptive statistics such as the Minimum and Maximum values will show unusually high or low values which can distort your analysis. Your domain knowledge will guide you to know if the minimum or maximum value is an outlier or not.</p><p><strong>Visual Inspection:</strong> Plotting the data using box plots, scatter plots, or histograms can provide a visual indication of potential outliers.</p><p><strong>IQR (Interquartile Range):</strong> Calculate the interquartile range (IQR) and identify observations outside the range [Q1–1.5 * IQR, Q3 + 1.5 * IQR] as potential outliers.</p><p><strong>Z-Score:</strong> Compute the z-score for each observation. A z-score measures how many standard deviations an observation is from the mean. Observations with high z-scores (typically beyond ±3) may be considered outliers.</p><p><strong>How to Deal with Outliers</strong></p><p>It is important to note that the decision to deal with outliers depends on the context of the analysis and the objective of the analysis. In some cases, outliers may be genuinely representative of the phenomenon being studied. Also consider how the outliers affect the results. If the outlier significantly distort interpretations, it may be necessary to address them. Instead of removing outliers immediately, investigate the reason for the outliers.</p><p>The following strategies can be used in handling outliers:</p><p><strong>Transform Data:</strong> Apply transformations like log transformation to reduce the impact of extreme values. This can help normalize the distribution.</p><p><strong>Winsorizing:</strong> Winsorizing involves replacing extreme values with values closer to the mean/median or within a specified range. This approach reduces the impact of outliers without removing them entirely.</p><p><strong>Imputation:</strong> For missing data caused by outliers, impute values based on statistical measures or predictive modeling.</p><p><strong>Separate Analysis:</strong> Perform separate analyses with and without outliers to understand their impact on results.</p><p><strong>Resampling Techniques:</strong> Use resampling techniques like bootstrapping to assess the stability of statistical estimates in the presence of outliers.</p><p><strong>Create Indicator Variables:</strong> Rather than removing outliers, create indicator variables to flag observations as potentially problematic.</p><p><strong>Remove Outliers:</strong> The most straightforward approach is to remove identified outliers from the dataset. This should however be your last option, and it should be done cautiously, as it may lead to data loss.</p><p><strong>Data Transformation and Data Cleaning Techniques</strong></p><p>It is important you understand the concept of Data Transformation/Feature Engineering. Transforming the data involves converting it into a more suitable format. This can include standardizing units of measurement, normalizing variables, or applying mathematical transformations to better fit the assumptions of statistical models. While Feature engineering involves creating new features or variables from existing ones to enhance the predictive power of the dataset. This can include extracting relevant data from date columns, creating interaction terms, or combining variables to capture meaningful patterns. For example, creating Gross Margin from Profit and Sales data or creating Cost from Sales and Profit data.</p><p>Below is a comprehensive list of data cleaning and preprocessing tasks that are commonly performed by data analysts and data scientists. These tasks are crucial for ensuring that the data used in analyses and modeling is accurate, consistent, and suitable for the intended purpose.</p><p>· Remove Empty Rows</p><p>· Remove Empty Columns</p><p>· Make Header as the First Row</p><p>· Unpivot Date to a single column</p><p>· Unpivot Categories/Variables to a single column</p><p>· Remove Obstructions Around the Data</p><p>· Remove Total or Sub-total in the data sheet</p><p>· Ensure Every Table Has a Unique Identifier (Primary Key) if not create a Primary Key for the table</p><p>· Check and Correct Spelling Errors</p><p>· Trim or Clean or Remove Unnecessary white Spaces or special characters</p><p>· Fill Up/Fill Down if missing data is systematic, be sure it fills correctly</p><p>· Drop NA Values if absolutely necessary</p><p>· Check for Duplicates</p><p>· Remove Duplicates</p><p>· Replace NA with Average/Median</p><p>· Replace NA with Text(Most Frequent category)</p><p>· Convert to Date Format(Y-M-D format)</p><p>· Transpose</p><p>· Unpivot Table</p><p>· Pivot Table where necessary</p><p>· Merging Tables/ Join tables when they comes from multiple sources</p><p>· Concatenation</p><p>· Split Columns (using delimiters “ “,@,#”:;)</p><p>· Remove unwanted Quotation Marks “” ‘ ‘</p><p>· Convert Numbers to DateTime(to_datetime()</p><p>· Convert Strings to DateTime(to_datetime()</p><p>· Extract Year, Month, Quarter, Day of Week from your fields/columns</p><p>· Extract Time — Hour</p><p>· Date Difference in Year, Month, or Days</p><p>· Make Ordinal/Nominal Data using One-Hot Encoding/Label Encoding/Dummy Variable(1 and 0)</p><p><strong>Tools for Data Cleaning</strong></p><p>Various tools and programming languages provide functionalities for effective data cleaning. Some commonly used tools for data cleaning used by Data Analysts/Data Scientists:</p><p><strong>Excel:</strong> While Excel is primarily a spreadsheet tool, it offers various functionalities for data cleaning, such as sorting, filtering, and using formulas to clean and transform data.</p><p><strong>Power Query</strong> : Power Query simplifies and accelerates the data preparation process, empowering users to clean, shape, and combine data from diverse sources with ease. Its integration with Power BI and Excel makes it a versatile tool for individuals and organizations engaged in data analysis and reporting. Power Query supports connecting to a wide range of data sources, including databases (SQL Server, Oracle), Excel files, text files, web pages, and cloud-based sources (Azure, SharePoint, etc.). Users can perform a series of transformations on the imported data to shape it according to their analytical needs. Transformations include filtering, sorting, grouping, pivoting, unpivoting, and merging tables. Power Query includes features for cleaning and handling errors in data, such as removing duplicate values, handling missing data, and dealing with errors during the data loading process. Power Query uses a formula language called “M” to define custom expressions and transformations. Users can write custom M expressions to create calculated columns, custom functions, and advanced transformations.</p><p><em>Please take note that Power Query handles CSV files differently compared to Excel files. Excel files, empty cells are filled with null, whereas in a CSV file, they remain blank. Therefore, you must manually replace the blank cells with null in the CSV file before cleaning or filling down.</em></p><p><strong>Python:</strong> <em>Pandas</em> is powerful data manipulation library in Python. It offers functions for data cleaning, handling missing values, filtering, and transforming data easily. Often used in conjunction with Pandas, is <em>NumPy</em> which provides numerical operations and array manipulation, facilitating data cleaning tasks.</p><p><strong>R:</strong> The <em>dplyr</em> is a popular package in R for data manipulation. It provides functions like filter(), mutate(), and summarize() for cleaning and transforming data. Also, <em>tidyr</em><strong>,</strong> works well with <em>dplyr</em> for reshaping and cleaning data by handling missing values and converting between wide and long formats.</p><p><strong>SQL: </strong>SQL databases and queries can be used to clean and preprocess data directly in the database. SQL provides commands like SELECT, UPDATE, and DELETE for data manipulation.</p><p><strong>Tableau Prep</strong> : Tableau Prep is a data preparation tool designed to help users clean, shape, and combine data for analysis in Tableau or other data visualization and analytics platforms. It is part of the Tableau ecosystem and is specifically crafted to streamline the data preparation process, making it more accessible to users without extensive coding or scripting skills. Users can easily clean and reshape their data using a variety of built-in operations. Tasks such as removing duplicates, filtering data, handling missing values, and pivoting columns can be performed visually. Tableau Prep allows users to combine multiple data sources through join and union operations, facilitating the integration of diverse datasets.</p><p><strong>Conclusion</strong></p><p>As data becomes increasingly integral to decision-making and innovation, the knowledge gained from this article serves as a valuable resource for data professionals navigating the complexities of real-world data. Data cleaning is not only about refining datasets but also about enhancing the overall effectiveness of data-driven projects, ultimately leading to more robust insights and informed decision-making. Embracing the principles outlined in this article empowers practitioners to address common data quality challenges, ensuring that their analytical endeavors are built on a foundation of trustworthy and relevant information.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d4844c32fd41" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[UNDERSTANDING YOUR DATA: BEGINNER’S GUIDE TO DATA ANALYTICS/DATA SCIENCE]]></title>
            <link>https://medium.com/@fokoye/understanding-your-data-beginners-guide-to-data-analytics-data-science-1826d4d82e2f?source=rss-68ed040cd594------2</link>
            <guid isPermaLink="false">https://medium.com/p/1826d4d82e2f</guid>
            <dc:creator><![CDATA[Fokoye]]></dc:creator>
            <pubDate>Mon, 20 Nov 2023 13:49:04 GMT</pubDate>
            <atom:updated>2023-11-22T23:24:45.439Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/626/1*0o38vEcF0U6JmOAO2v-q4g.jpeg" /></figure><p><strong>Introduction</strong></p><p>In most organisations, when a business encounters challenges or seeks strategic insights, management raises key questions that demand precise answers. The journey as a data analyst begins with identifying the questions posed by management, which serve as a guide for your analytical process. In order to provide answers to the questions posed by management, you will need the data. Where is the data stored? The data is stored in the organisation’s database. The next phase entails exploring the organization’s database, underscoring the importance of your proficiency as a data analyst in the language used to communicate with the database known as SQL. Your responsibility in this role is to extract the relevant data from databases by employing SQL queries. Following the extraction process, the subsequent steps involve the essential tasks of data cleaning and transformation.</p><p>In the realm of data analysis and machine learning, the phrase “garbage in, garbage out” underscores the importance of clean, accurate, and reliable data. Data cleaning is an essential step in the data preprocessing pipeline, aiming to enhance the quality of datasets by addressing issues such as outliers and missing data. Data is rarely clean or ready for analysis. As the data, is often characterized by a lot of issues such as ,misspellings, white spaces, missing values, and outliers, requires careful cleaning. The heart of your role as the data analyst/data scientists lies in the analysis phase, where statistical methods, machine learning algorithms, and exploratory data analysis techniques come into play using tools such as Excel, Python or R. Through this analytical lens, you uncover trends , patterns and extract actionable insights from the data. Your aim is to not only answer the posed questions what has happened but also to provide answers to why it has happened.</p><p>After you are done with your analysis, the last stage involves presenting these insights to the management in form of reports and dashboards. Visualization tools such as PowerBI and Tableau, which enables you to build reports, and dashboards become your instruments of communication, translating complex analyses into a narrative that empowers decision-makers with clarity. More importantly, is your ability to communicate your insight in form stories. It is important as data analyst/data scientists, we need to have data storytelling skills, to communicate with the management team who are usually non-technical people. Through the lens of data, you become a strategic partner, guiding the management towards evidence-based decisions that have the power to shape the trajectory of the business.</p><p><strong>Variables</strong></p><p>When addressing the questions/queries posed by management, an important aspect of the analytical process involves identifying the variables embedded within these questions. For instance, if management is asking <strong>“Which regions generate the most sales?”</strong> The management is specifically interested in understanding the sales performance across different regions. The first variable that can be identified is “Region” while the second variable is “sales,” which serves as the quantitative measure reflecting the revenue or volume of transactions occurring in each region.</p><p>In the context of statistics a variable is a characteristic or attribute that can vary among individuals or objects in a study. In simple terms, a variable is something that can vary or be different. Variables are the building blocks of data, and they represent the different factors that data analysts/data scientists, measure, observe, or manipulate in order to understand and analyze a problem.</p><p>Variables can be broadly classified into two main types: <strong>Quantitative variables</strong> and <strong>Qualitative variables</strong>.</p><p><strong>Quantitative Variable:</strong> A quantitative variable, also known as a numerical variable, represents measurable quantities and is characterized by numerical values. This type of variable we can perform arithmetic operations like addition, subtraction, multiplication, and division, and it will make sense. Examples of quantitative variables include age, height, income, profit, sales, cost, temperature, and the number of items sold.</p><p><strong>Qualitative Variable: </strong>A qualitative variable, also known as a categorical variable, represents categories or labels that we can group in buckets. This type of variable describes qualities or characteristics and is often non-numeric. Examples of qualitative variables include gender, color, type of car, or the brand of a product. We can also have some numeric variables that can be classified as qualitative variables such as phone numbers, account numbers and so on. Qualitative variables does not have numerical values with inherent meaning. You cannot perform arithmetic operations on them, the only kind of analysis you can do is to count or find the proportion. For example , A phone number has numbers does not make it a quantitative variable, it is typically considered qualitative data. The phone number or account number, is a sequence of digits and symbols that serve as a unique identifier for a telephone line or customer, making it a label or category. We cannot add up all the customers phone or find the average account number of the customers, as it will not make any sense.</p><p>The distinction between quantitative and qualitative variables is crucial in data analysis, as it guides the choice of appropriate statistical methods and tools that will be used in the analysis. Quantitative variables lend themselves to statistical measures such as means, and arithmetic operations, while qualitative variables are often analyzed using methods like counts, frequency distributions and percentages.</p><p><strong>Measurement</strong></p><p>The act of measurement transforms variables from theoretical constructs into actionable entities. It empowers individuals to track progress, identify trends, and make informed adjustments based on quantifiable data. In the absence of measurement, managing variables becomes akin to navigating in the dark, lacking the necessary insights to steer towards success. Variables exist in the realm of ideas and concepts until they are measured. The saying “what you cannot measure, you cannot manage” encapsulates the essence of the importance of measurement in effective management and decision-making. Without measurement, variables remain abstract and elusive. It is through measurement that we assign numerical values, dimensions, or scales to these variables, allowing us to capture their characteristics and assess their impact. When variables are measurable, they become manageable, providing a foundation for analysis, comparison, and informed decision-making. For instance, in a business context, key performance indicators (KPIs) are often measured to gauge the success of strategies and operations.</p><p><strong>Types of Measurement Scales</strong></p><p>The four fundamental types of measurement scales : <em>Nominal, Ordinal, Interval, and Ratio scale</em></p><p><strong>Nominal Scale</strong>: At the most basic level, the nominal scale classifies variables into distinct categories or groups with no inherent order or ranking. Nominal variables are qualitative in nature and often involve categorical distinctions. Common examples include gender, ethnicity, color, or type of car. When assigning numerical values to categorical variables like gender, it’s essential to recognize that these numerical values are just labels and do not imply any inherent order, superiority, or magnitude.</p><p>The use of numerical labels for categorical variables is a common practice, but it’s crucial to interpret them correctly and communicate their meaning to ensure accurate analysis and understanding of the data. When you want to use the data for the purpose of analysis, we cannot use the text as they are, “Male” or “Female”, in running the analysis, you will need to assign values. For example, you may assign 1 to Male and 2 to Female, this is what is referred to as label encoding. This is a common practice in coding categorical variables for analysis or machine learning. However, the choice of 1 or 2 (or any other numerical values) is arbitrary, and these values are used purely as identifiers rather than indicating any quantitative relationship between the categories.</p><p><strong>Ordinal Scale: </strong>The ordinal scale introduces a sense of order or ranking among categories, though the intervals between them are not uniform or meaningful. Variables measured on an ordinal scale convey relative positioning but lack precise measurement of the distance between ranks. Examples include education levels, income level, customer preference, customer satisfaction ratings, or socioeconomic status. When collecting data for ordinal variables, the focus is on preserving the order while acknowledging the absence of equal intervals.</p><p>A typical use case is the Likert scale. The Likert scale is a widely used psychometric tool that measures attitudes, opinions, and perceptions. It provides respondents with a range of choices, typically in the form of a statement, and asks them to express the extent to which they agree or disagree with that statement. The Likert scale is named after its creator, psychologist Rensis Likert. It’s a valuable tool in various fields, including business, education, and social sciences.</p><p>Here’s how a basic Likert scale is structured:</p><p>1 Strongly Disagree, 2 Disagree, 3 Neutral, 4 Agree, 5 Strongly Agree</p><p>Another example, an organization wants to assess and categorize the performance levels of its employees. The company may use an ordinal scale to rank employees based on their performance, with categories such as: Exceptional Performance, High Performance, Satisfactory Performance, Needs Improvement, Unsatisfactory Performance. You can assign ranks to them 5 for Exceptional Performance , 4 for High Performance, 3 for Satisfactory Performance, 2 for Needs Improvement, and 1 for Unsatisfactory Performance. In this case, 5 is greater than 4, 3,2,1 which shows there is an order.</p><p>A company may use an ordinal scale to measure customer satisfaction. The scale could include categories such as: Very Satisfied 5, Satisfied 4, Neutral 3, Dissatisfied 2, Very Dissatisfied 1. Customers would provide feedback, and their responses would be categorized on this ordinal scale. This helps the business understand the overall satisfaction level of its customer base.</p><p>When assessing the quality of products, a business might use an ordinal scale to rate them:</p><p>4 <em>Excellent , 3 Good, 2 Average, 1 Poor.</em> This scale allows the company to categorize the perceived quality of its products, helping in decision-making related to product improvements or adjustments.</p><p>When collecting feedback on specific features of a product, an ordinal scale could be used.</p><p><em>5 Excellent, 4 Good, 3 Fair, 2 Poor, 1 Very Poor.</em></p><p>Both nominal and ordinal scales are valuable in capturing and analyzing qualitative variables. They provide a structured way to organize and understand categorical variables, allowing data analysts, and data scientists to draw insights from non-numeric variables.</p><p><strong>Interval Scale:</strong> The interval scale introduces equal intervals between values, but it lacks a true zero point. In the context of measurement and scales, a “true zero point” refers to a point on the scale that represents a complete absence or lack of the quantity being measured. A true zero point indicates that the measured attribute is entirely absent. A common example is the Celsius temperature scale. While the intervals between degrees are equal (the difference between 20°C and 30°C is the same as the difference between 30°C and 40°C), the zero point (0°C) is arbitrary. It doesn’t mean the absence of temperature; it’s just a point chosen as the freezing point of water.</p><p><strong>Ratio Scale:</strong> The ratio scale possesses all the properties of the interval scale but includes a true zero point, indicating the absence of the measured attribute. Examples of ratio variables include height, weight, income, or age.</p><p>Let’s say you have 4 cars, and your other friend has 2. You can say, “I have twice as many cars as my friend.” That’s because 4 is twice as much as 2. The ratio scale allows you to understand not just how many you have but also how much more or less it is compared to others. A ratio scale gives you a starting point (zero) and lets you compare amounts in a mathematical way. It’s perfect for when you want to talk about how much of something you have and how it compares to what others have. In the context of a variable like sales, if the value is 0, it typically means there were no sales during the specified period. In this case, 0 represents the absence or lack of the measured attribute, which is sales in this scenario.</p><p><strong>What is Data?</strong></p><p>We have seen how we move from the conceptualization of variables to the process of measurement. After measurement of the variables, the next step is to collect the data. The progression from variables to measurement to data represents the foundation of any analytical task, emphasizing the importance of clearly defining and quantifying the elements under investigation to derive meaningful conclusions.</p><p>Data simply means the raw facts and figures. When the data has been processed it becomes information. For example let us use the LEGO which most children love playing with to make an illustration. Each LEGO brick is like a tiny piece of data. If you want to build something amazing, like a spaceship or a castle, you need to organize and understand all those LEGO pieces. This is a bit like what data is for grown-ups! Data are the facts that we collect and organize to understand things better. It can be numbers, texts, pictures, video or even sounds. Just like how your LEGO bricks come together to create something cool, data helps us make sense of things in our organisation.</p><p>Also, imagine your home is completely dark, and you need to walk from your bedroom to the living room. Since you’re familiar with your home and have lived there for a while, you can navigate through the darkness using your memory. You know the general layout, and you can feel your way around. However, in the process, you might occasionally bump into furniture or kick something accidentally because you can’t see where you’re going. Now, let’s think of this scenario in terms of an organization and data. In this analogy, your home represents the organization, and the journey from the bedroom to the living room symbolizes making decisions or taking actions within the organization.</p><p>Without data, it’s like moving through the organization in the dark. People rely on their experience and knowledge, trying to make the best decisions based on what they remember or feel. This can work to some extent, but there’s a chance of stumbling upon challenges or making mistakes along the way. Now, consider turning on a flashlight. Suddenly, you can see everything clearly. The path is illuminated, and you can navigate without any hiccups. The flashlight acts as a guiding light, providing clarity and preventing unnecessary obstacles.</p><p>In the organizational context, data serves as that guiding light. It brings clarity to decision-making processes. With data, organizations can see patterns, trends, and potential opportunities and challenges. It helps in making informed decisions, just like how the flashlight enables you to move confidently without tripping over things. So, data is like the light for an organization. It brings visibility, reduces risks/ uncertainties, and allows for more effective navigation. Instead of relying solely on past experiences and memories, organizations can use data to illuminate the path forward, making their journey smoother and more efficient.</p><p><strong>Understanding Data</strong></p><p>We can start by examining data from three distinct perspectives — its source, nature, and timing — we gain a comprehensive understanding that enriches our ability to derive meaningful insights.</p><ol><li><strong>The Source of Data:</strong> Understanding the source of data is fundamental to assessing its reliability and relevance. Data can be broadly categorized into two types: primary and secondary.</li></ol><p><strong>Primary Data:</strong> These are data you generate yourself. These are data that can be internally generated in the organization either through surveys, experiments, and direct observations are common methods of obtaining primary data.</p><p><strong>Secondary Data:</strong> In contrast, secondary data are data generated externally (other people). This could include data from government reports, academic studies, or organizational databases.</p><p><strong>2. Understanding the Nature of Data:</strong> Data can be further classified based on its nature, encompassing both quantitative (numeric) and qualitative (categorical) dimensions.</p><p><strong>Quantitative Data:</strong> Quantitative data are numerical in nature. quantitative data represents measurable quantities and can be subjected to mathematical operations. Quantitative data can indeed be classified into two main types: <em>discrete and continuous.</em></p><p><strong>Discrete Data</strong>: Discrete data are often counted in whole numbers (integers). For example, the number of students in a class, the count of cars in a parking lot, or the number of books on a shelf are all examples of discrete data.</p><p><strong>Continuous Data:</strong> Continuous data are data that can be expressed in decimal or fractional form. Height, weight, temperature, and time are examples of continuous data. For instance, a person’s height can be 175.5 cm, and it could be 175.6 cm or any value in between.</p><p><strong>Qualitative Data: </strong>Qualitative data, also known as categorical data, are generally non-numeric and represents categories or labels. Examples of qualitative data include gender (male or female), color (red, blue, green), or product categories (electronics, clothing). However, there are numeric data like Phone numbers and account numbers that can be considered as nominal data within the realm of statistics, which is a category of qualitative data.</p><p><strong>3. The Timing Perspective:</strong> The temporal dimension of data provides insights into how observations are organized either at a point in time or over time.</p><p><strong>Cross-Sectional Data: </strong>Cross-sectional data captures a snapshot of a population or phenomenon at a specific point in time. This type of data allows for comparisons and analyses across different groups but does not provide insights into changes over time.</p><p><strong>Time Series Data:</strong> Time series data, in contrast, involves observations collected over successive and evenly spaced time intervals. This format facilitates the exploration of trends, patterns, and seasonal variations. Time series data is crucial for forecasting and understanding the evolution of phenomena over time. The frequency of your time series data can be: Annual, Quarterly, Monthly, Weekly, Daily…</p><p><strong>How should your table be structured?</strong></p><p>A guide on how you might structure your table:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/780/1*7tBaAmOKk7kIblAwjjauNQ.png" /></figure><p>· <strong>No Empty Rows: </strong>There should be no empty rows, ensure that every row in your table contains relevant data.</p><p>· <strong>No Empty Columns:</strong> Similarly, there should be no empty columns, ensure that every column in your table has meaningful data. Avoid leaving any columns blank.</p><p>· <strong>One Row of Header:</strong> Have a single row at the top of your table that serves as the header. This row should contain clear and concise labels for each column.</p><p>· <strong>No Total/Sub Total:</strong> Avoid including total or subtotal in your table. The total/subtotal will be done in your analysis.</p><p>· <strong>Dates in a Single Column:</strong> All dates should be in a single column.</p><p>· <strong>Category/Variables in a Single Column:</strong> All categories or variables should be in a single column. This helps in maintaining the integrity of your data and makes it easier to filter or sort.</p><p>· <strong>No Obstructions Around the Data:</strong> Avoid unnecessary obstructions (such as additional text, images, or formatting) around the data. Keep the table clean and focused on the information it presents.</p><p>· <strong>Every Table Must Have a Primary Key:</strong> Every table should have a unique identifier. This unique identifier is called a Primary Key which ensure data integrity.</p><p>Adhering to these rules is crucial for creating a well-structured and organized table ready for analysis. If your table does not meet these criteria, you may need to clean and transform the data to ensure it aligns with these guidelines. The guidelines above will guide you on how to clean and transform your data.</p><p><strong>Understanding Each Field/Column in your Data</strong></p><p>The key to unlocking the full potential of your data lies in understanding the intricacies of each field/column and employing a meticulous approach to data cleaning. How you clean the data is depends on how well you understand the data. In the process of examining each column, identifying missing data, outliers, unique values, and explore the transformative potential of creating new data.</p><p>The first step in effective data cleaning is to ask yourself with each column of the dataset. This involves more than a cursory glance; it requires a deep dive into the characteristics and nature of the data. Here’s a step-by-step guide to scrutinizing each column:</p><ol><li><strong>Missing Data:</strong> Identify any missing data within the column. Missing values can significantly impact analyses and models, and understanding their distribution helps in devising appropriate strategies for imputation or handling.</li><li><strong>Outliers:</strong> Unearth outliers that might skew your analysis. Outliers can distort statistical measures and machine learning models, so it’s crucial to decide whether to adjust, remove, or transform these extreme values.</li><li><strong>Unique Values:</strong> Examine the unique values within each column. Understanding the range and diversity of values provides insights into the nature of the data and aids in making informed decisions during the cleaning process.</li><li><strong>Calculations and New Data:</strong> For each column, ask yourself what calculations may be needed. If a column contains date-related information, consider extracting additional features such as year, month, quarter, day of the week, hour, etc. For categorical data, explore the possibility of creating binary indicators or encoding them for machine learning models.</li></ol><ul><li>Date Column: Extract year, month, quarter, day of the week.</li><li>Time Column: Extract the hour.</li><li>Date of Birth: Calculate Age or determine the generation (e.g., Gen X, Gen Z, Millennials).</li></ul><p>Extracting age from a Date of Birth column, for instance, provides valuable demographic information. Additionally, categorizing individuals into generational groups can give insights into behavior and preferences. If there is price and quantity, you need that to compute the revenue, same thing if you see cost and revenue, you will need to compute the profit.</p><p><strong>Conclusion</strong></p><p>In recent years, the value of data has surged, experiencing an exponential growth in volume, and it has evolved into a critical asset for every organization. The primary driver behind this surge is technology, which has made collecting, transferring, and storing data easier than ever before. Always, have at the back of your mind that every data you collect will be analyzed sometime in the future, the data should be put in a structure ready for analysis. In essence, understanding your data empowers individuals and organizations to harness the full potential of information. It’s not just about having data; it’s about extracting meaningful insights, making informed choices, and driving positive outcomes in various domains. This leads you to building a data culture which involves instilling a mindset where data is central to decision-making, learning, and innovation. Organizations fostering a strong data culture are well-equipped to navigate the challenges of a data-driven world and fully leverage on their data assets. The nature of the data at your disposal plays a pivotal role in shaping the type of analysis you can undertake.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1826d4d82e2f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Python in Excel: A Game-Changer for Accountants and Financial Analysts]]></title>
            <link>https://medium.com/@fokoye/python-in-excel-a-game-changer-for-accountants-and-financial-analysts-e41d3d9a7b9a?source=rss-68ed040cd594------2</link>
            <guid isPermaLink="false">https://medium.com/p/e41d3d9a7b9a</guid>
            <dc:creator><![CDATA[Fokoye]]></dc:creator>
            <pubDate>Wed, 23 Aug 2023 03:26:12 GMT</pubDate>
            <atom:updated>2023-08-23T03:26:12.202Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*GHNglYFNToYlJ-Nfo8gyGQ.png" /></figure><p><strong>Introduction</strong></p><p>Anaconda and Microsoft have unveiled an innovation that is set to reshape the landscape of data analysis and visualization. The convergence of Python and Excel brings together two powerful tools, each with its own unique strengths, to create a synergy that empowers both Excel users and Python practitioners in unprecedented ways. This milestone development bridges the gap between spreadsheet-based data manipulation and the versatile capabilities of Python programming, offering a wealth of benefits, particularly for accountants and financial analysts.</p><p><strong>Empowering Excel Users</strong></p><p>For accountants and finance professionals accustomed to Excel’s interface, the introduction of Python integration brings a wave of transformation. Imagine being able to harness Python’s extensive libraries within the familiar environment of Excel. This means that tasks that were once considered the domain of data scientists and programmers are now within reach for Excel users. With Python in Excel, accountants can perform intricate data analysis using popular libraries like pandas and statsmodels. Complex financial data can be manipulated, cleansed, and transformed effortlessly, enabling a deeper understanding of financial trends and patterns. Tasks such as variance analysis, trend forecasting, and budget optimization can be executed more efficiently and with a higher degree of accuracy.</p><p><strong>Integration of Python in Excel</strong></p><p>Microsoft’s move to seamlessly integrate Python code within Excel’s grid is a testament to the ever-evolving nature of technology. It bridges the gap between spreadsheets and programming, democratizing access to Python’s capabilities and enabling professionals to perform advanced data analysis, manipulation, and visualization tasks without the need for extensive technical know-how. This innovation has the potential to revolutionize how we approach data-driven tasks, making Excel an even more powerful tool in the modern professional’s toolkit.</p><p>Traditionally in the Accounting and Finance world, Excel has been the go-to software for financial modeling, data cleaning, basic data and financial analysis. Its user-friendly interface and familiar grid layout have made it indispensable for professionals across various industries, especially in accounting and finance. However, as the complexity of data analysis increased, Excel’s limitations became apparent. This is where Python steps in as a versatile and robust programming language, capable of handling intricate statistical computations, data manipulation, and advanced visualizations. The recent collaboration between Anaconda, a prominent player in the Python data science ecosystem, and Microsoft, the company behind Excel, has led to a groundbreaking development. Python can now be seamlessly integrated into Excel, offering users the best of both worlds. This integration eliminates the need to switch between different tools for different tasks, streamlining the workflow and enhancing productivity. Now you can write Python code directly in Microsoft Excel with no Python installation required.</p><p><strong>Elevating Python Enthusiasts</strong></p><p>For professionals well-versed in Python programming, this integration offers a new dimension to their workflow. Python practitioners can now leverage the widespread accessibility of Excel to share their insights and analyses seamlessly. By embedding Python scripts directly within Excel, they can create interactive and dynamic spreadsheets that showcase data analysis, predictions, and visualizations. This opens up avenues for effective communication with colleagues, stakeholders, and clients who might not be well-versed in programming but are comfortable with Excel. Moreover, Python’s robust libraries for data visualization, such as Matplotlib and Seaborn, can be employed to create visually appealing charts, graphs, and dashboards. These visualizations, when combined with Excel’s presentation capabilities, enable the creation of comprehensive reports that are both informative and engaging.</p><p><strong>Why Accountants Should Learn Python</strong></p><p>In the ever-evolving landscape of finance and accounting, professionals are continually seeking new ways to enhance their skills and stay ahead of the curve. One skill that has gained significant traction in recent years is programming, and in particular, learning Python. While accounting traditionally involves number crunching, financial analysis, and reporting, the integration of technology and data-driven decision-making is reshaping the field. Some reasons why accountants should consider learning Python:</p><p><strong>Adaptation to Industry Trends:</strong> The finance and accounting industry is undergoing a digital transformation. Technologies like blockchain, artificial intelligence, and big data are reshaping how financial transactions are recorded, analyzed, and reported. Learning Python equips accountants with a foundational programming language that can be used to understand and adapt to these technological shifts.</p><p><strong>Automation and Efficiency:</strong> Python is renowned for its simplicity and versatility, making it an ideal language for automating repetitive and time-consuming tasks. Accountants often deal with large volumes of data, from data entry to reconciliations. By learning Python, accountants can create scripts to automate these tasks, freeing up valuable time for more strategic and analytical work. This leads to increased efficiency and reduced chances of human error.</p><p><strong>Data Analytics and Visualization:</strong> Accountants are no longer limited to working solely with financial statements. Modern accounting involves analyzing vast amounts of data to identify trends, patterns, and insights. Python’s data manipulation libraries, such as pandas, allow accountants to clean, transform, and analyze data with ease. Additionally, libraries like Matplotlib and Seaborn enable the creation of compelling visualizations that can aid in decision-making and communication.</p><p><strong>Building Reports and Dashboards:</strong> Python offers tools to generate dynamic and interactive reports, dashboards, and presentations. With libraries like Jupyter Notebook and Plotly, accountants can create reports that update in real-time, providing stakeholders with the most up-to-date information. Interactive dashboards can offer a comprehensive overview of financial performance, highlighting key metrics and facilitating strategic planning.</p><p><strong>Predictive Analytics:</strong> Python’s machine learning libraries, such as scikit-learn, enable accountants to delve into predictive analytics. By building models that analyze historical financial data, accountants can make informed predictions about future trends, aiding in budgeting, forecasting, and risk assessment.</p><p><strong>Career Advancement:</strong> In today’s competitive job market, possessing programming skills like Python can set accountants apart. Employers value professionals who can leverage technology to solve complex problems and drive innovation. Accountants with Python skills are well-positioned to take on roles that involve data analysis, financial modeling, and process improvement.</p><p><strong>Conclusion</strong></p><p>In summary, while the prospect of learning a programming language might seem daunting at first, the benefits that Python offers to accountants are substantial. From automation to data analysis, Python equips accountants with tools to work smarter, uncover insights, and contribute to strategic decision-making. As the finance and accounting field continues to embrace technology, learning Python can be a valuable investment in your professional growth and long-term success. The unveiling of Python in Excel signifies a significant stride towards enhancing the capabilities of both Excel users and Python practitioners. Accountants and data analysts are positioned to benefit immensely from this collaboration, as they can now seamlessly merge the power of Python’s data manipulation and visualization tools with the convenience and familiarity of Excel. The integration not only streamlines workflows but also fosters collaboration and knowledge sharing across diverse teams. As this innovation takes root, it promises to reshape the way professionals approach data analysis, offering a harmonious blend of simplicity and sophistication.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e41d3d9a7b9a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Turning Industry Experience into Research Gold: Tips for Early Career Academic Researchers…]]></title>
            <link>https://medium.com/@fokoye/turning-industry-experience-into-research-gold-tips-for-early-career-academic-researchers-6816bc40d007?source=rss-68ed040cd594------2</link>
            <guid isPermaLink="false">https://medium.com/p/6816bc40d007</guid>
            <category><![CDATA[research]]></category>
            <category><![CDATA[practical-research]]></category>
            <category><![CDATA[industry-research]]></category>
            <dc:creator><![CDATA[Fokoye]]></dc:creator>
            <pubDate>Mon, 24 Jul 2023 22:29:43 GMT</pubDate>
            <atom:updated>2023-07-24T22:29:43.082Z</atom:updated>
            <content:encoded><![CDATA[<h3><strong>Turning Industry Experience into Research Gold: Tips for Early Career Academic Researchers transitioning from Industry to Academia</strong></h3><p><strong>Anita’s Story</strong></p><p>Uche Anita had always been fascinated by numbers and finance since her early academic years. After completing her undergraduate degree in Banking and Finance, she embarked on a successful career in the banking industry. With time, she developed a keen interest in understanding the theoretical underpinnings of finance and accounting. However, her busy schedule as a banking professional left her little room to explore her academic curiosities. During a chance meeting with her old friend, Emeka, who was pursuing a Ph.D. in Accounting, Uche shared her longing for more in-depth research and academic pursuits. Emeka, aware of Uche’s dedication to her banking career, encouraged her to consider taking up a Ph.D. herself, emphasizing the importance of practitioner researchers in bridging the gap between industry and academia. Emeka believed that professionals like Uche could bring a unique perspective to academic research, making it more relevant and applicable to real-world challenges.</p><p>Intrigued by the idea of blending her banking expertise with academic research, Uche made the bold decision to enroll in a Ph.D. program in Accounting while continuing her job on a part-time basis. This allowed her to maintain her industry connection and access to valuable data while immersing herself in the world of academia.</p><p>Uche Anita was passionate about addressing the practical issues faced by the banking industry. During her time as a banking professional, she observed the inefficiencies in risk management practices, which often led to suboptimal decision-making and potential financial losses for the banks. Uche believed that optimizing risk management strategies could significantly benefit the banking sector, and this became the focus of her research interest. As Uche explored the possibility of conducting research on optimizing risk management, she encountered the first major challenge: the topic’s non-measurability. The concept of “optimization” was broad and subjective, making it difficult to define precise variables and measurable outcomes. Uche realized that for her research to be academically rigorous and practically relevant, she needed a more defined and quantifiable research question. As Uche delved deeper into her Ph.D. research, she realized that her friend Emeka was right. Most academic studies in accounting and finance were focused on theoretical constructs and lacked the practical insights that industry professionals could offer. Uche noticed that this disconnect often led to research outcomes with limited application in real-world banking scenarios. Determined to address the gap she had identified, Uche structured her Ph.D.</p><p>Another hurdle Uche faced was the limited existing literature on the specific aspect of risk management she wanted to explore. While there were some studies on risk management practices in the banking industry, the information was often fragmented and not directly applicable to her envisioned research direction. The lack of a solid foundation in the literature made it challenging for Uche to build a strong theoretical framework for her study.</p><p>Despite these challenges, Uche was determined to find a viable research topic that would fill the gap between industry and academia. To overcome the non-measurability issue, she decided to narrow down her focus to a specific aspect of risk management: “The Impact of Bank Characteristics on Risk Disclosures in the Nigerian Banks.” This new research question allowed her to explore the effects of bank characteristics on risk disclosures, making the topic more measurable and objective.</p><p>To address the lack of existing literature, Uche conducted an extensive literature review that spanned various disciplines, including finance, accounting, and banking regulatory framework. She also collaborated with academic mentors and industry experts, seeking guidance on potential sources and relevant studies. Through perseverance and dedication, Uche identified emerging trends in risk management and gathered enough literature to construct a solid theoretical framework for her research.</p><p>Upon completing her Ph.D., Uche’s research earned widespread recognition for its innovative approach and relevance to the banking industry. Her findings shed light on best practices that significantly influenced decision-making processes within her bank. The practical solutions she proposed were successfully implemented, leading to improved risk disclosures in the Nigerian banking industry. Inspired by her own transformation, Uche became a vocal advocate for encouraging more banking professionals to pursue Ph.D. research in accounting and finance. She actively engaged with industry forums and academic circles, promoting the idea that practitioner researchers could bridge the gap between theory and practice and drive more meaningful advancements in the banking industry.</p><p><strong>Introduction</strong></p><p>As we can see from Uche Anita’s journey from being a dedicated banking professional to becoming a academic researcher exemplifies the power of bridging the gap between industry and academia. Her experience highlights the crucial role that professionals like her can play in contributing practical and relevant insights to academic research. Uche’s story serves as an inspiration to industry practitioners, encouraging them to consider the path of practitioner researchers, thereby fostering a mutually beneficial relationship between the industry and the academic community. When starting your journey from industry to academia, there are several questions that you will be faced with:</p><p>· What topic do I choose for my PhD or DBA dissertation?</p><p>· What problem am I trying to solve and can I implement or deploy the solution in my organization?</p><p>· Do I have access to the relevant data and do I know how to measure the variables?</p><p>· What will be my contribution to knowledge?</p><p><strong>Selecting a Dissertation Topic: </strong>One of the primary challenges is deciding on a suitable topic for the dissertation. It requires careful consideration to identify a research area that aligns with one’s expertise and interests while also addressing relevant issues in the industry. Having an idea of a problem in the industry is a great starting point, but it’s vital to conduct a thorough review of existing literature to avoid redundancy and understand the current state of research. By reviewing previous studies, you can identify what issues have already been addressed and what gaps still remain, which helps refine your research question and ensures your contribution is novel and valuable. Sometimes, what may sound brilliant as a topic might not be practically researchable. It’s essential to assess the feasibility of conducting research on the chosen topic, considering factors such as data availability, accessibility, and ethical considerations. Additionally, some research questions may require complex methodologies or techniques to measure the variables accurately. Thus, carefully thinking about how to measure the variables is crucial to ensuring the validity and reliability of your research findings.</p><p>The process of selecting a dissertation topic requires a balance between having a meaningful research problem, ensuring its originality, and making it feasible and measurable. Engaging in a thorough literature review and thoughtful consideration of research design will guide you in choosing a topic that aligns with both academic standards and practical application, setting the foundation for a successful and impactful dissertation.</p><p><strong>Defining the Research Problem : </strong>Clarifying the problem to be solved is essential for a successful research endeavor. Additionally, practitioners often contemplate whether the solutions proposed through research can be practically implemented or deployed within their organizations. Defining the research problem is a critical aspect of any successful research undertaking. One of the primary considerations is identifying the research gap that your study aims to fill. Clarifying the problem to be solved is essential to ensure that your research has a clear purpose and direction. Additionally, practitioners often contemplate the practical implementation or deployment of the proposed solutions derived from research within their organizations. This practical aspect ensures that the study’s findings are actionable and can make a tangible impact in real-world scenarios. Research gaps can take various forms, including:</p><p><strong>Resolving Conflicting Findings:</strong> Addressing conflicting or contradictory results from previous research to arrive at more definitive conclusions.</p><p><strong>New Measurement Approaches:</strong> Introducing innovative ways to measure variables, improving accuracy, and providing a fresh perspective on the topic.</p><p><strong>New Methodology:</strong> Incorporating novel research methodologies that offer a unique and comprehensive understanding of the subject matter.</p><p><strong>Extending Sample Period:</strong> Extending the study’s time frame to provide a longitudinal view, enabling a deeper analysis of trends and changes over time.</p><p><strong>Comparative Analysis:</strong> Conducting comparative studies to analyze differences or similarities between groups or conditions, such as pre- and post-implementation of a particular intervention.</p><p><strong>Mediation or Moderation Analysis:</strong> Incorporating mediation or moderation variables to explain and understand certain relationships between variables.</p><p>By identifying and addressing such research gaps, practitioner researchers can make valuable contributions to the existing body of knowledge and offer practical solutions that benefit their organizations and industries. A well-defined research problem, combined with careful consideration of practical applicability, ensures that the study’s outcomes are not only academically sound but also relevant and impactful in real-world settings.</p><p><strong>Relevant Data and Measurement Techniques: </strong>Look before you leap. Before diving into your research , it is crucial to carefully plan and assess the data required for the research and the methods to measure variables accurately. Acquiring access to pertinent data can be a daunting task, and researchers must ensure they have the necessary information to answer their research questions. Having access to the necessary data can indeed significantly impact the speed and efficiency of completing your study. When researchers have their data readily available, they can expedite the analysis process and move forward with interpreting the results and drawing conclusions. Additionally, understanding how to measure variables accurately is crucial for maintaining the validity of the research findings. Here are some key points to keep in mind:</p><p><strong>Data Requirements: </strong>Clearly outline the data needed to address your research questions. Consider the type of data (quantitative, qualitative, or mixed), the sources from which the data will be obtained, and the timeframe for data collection.</p><p><strong>Data Accessibility:</strong> Assess the feasibility of accessing the necessary data. Depending on the nature of the study, data collection can be a challenging and time-consuming process. Ensure that the data sources are accessible and that you have the necessary permissions or agreements in place to use the data.</p><p><strong>Data Quality: </strong>Verify the quality and reliability of the data you plan to use. Ensuring the data’s accuracy and completeness is vital for drawing valid conclusions and making well-founded recommendations.</p><p><strong>Measurement Techniques:</strong> Choose appropriate measurement techniques that align with your research objectives. The accuracy and validity of your findings depend on using reliable and valid measurement tools and methods.</p><p><strong>Sampling:</strong> If your study involves sampling, carefully select a representative sample to ensure the generalizability of your findings. Justify your sampling approach and consider potential biases.</p><p><strong>Ethical Considerations:</strong> Address any ethical considerations related to data collection and use. Obtain necessary approvals, protect participants’ confidentiality, and ensure data handling complies with ethical guidelines.</p><p><strong>Data Analysis:</strong> Plan your data analysis approach based on the nature of your data and research questions. Select appropriate statistical or qualitative analysis methods that align with your study’s objectives. Understanding data analysis techniques allows researchers to plan their data collection effectively and select the appropriate types of data needed to address their research questions.</p><p>By thoroughly considering relevant data and measurement techniques before starting the research, you can avoid potential pitfalls and increase the likelihood of producing meaningful and robust research findings. Thoughtful planning and preparation in this regard are essential for maintaining the integrity and validity of your research and ensuring its overall success.</p><p><strong>Contribution to Knowledge: </strong>As practitioners it is vital to comprehend the potential contributions of their research to existing knowledge. Identifying the unique insights and value that the research can add to the field is essential for academic success and practical application. Understanding the potential contribution to knowledge. As a practitioner researcher, it is vital to consider how your study can add value and make a meaningful impact in various areas:</p><p><strong>Existing Literature: </strong>Consider how your research will contribute to the existing body of knowledge in your field. Will your findings support, challenge, or extend previous research? Identifying gaps in the literature and offering new insights can enrich academic discussions and shape future research.</p><p><strong>Policy Development: </strong>Depending on the nature of your research, your findings may have implications for policy development. By providing evidence-based recommendations, your study can influence decision-makers and contribute to the formulation of effective policies in relevant industries or sectors.</p><p><strong>Industry Relevance</strong>: As a practitioner researcher, your work has the potential to directly impact your industry. Identify how your research outcomes can be applied to real-world challenges and bring positive changes to organizational practices or strategies.</p><p><strong>Practical Application:</strong> Consider how the solutions proposed through your research can be deployed in your organization. Your study’s practical implications can lead to process improvements, cost savings, or enhanced decision-making within your workplace.</p><p>By understanding the potential contributions of your research, you can ensure that your study is both academically valuable and practically applicable. Recognizing the unique insights and value your research brings to the field enhances its significance and fosters a greater appreciation for the role of practitioner researchers in advancing knowledge and driving positive change.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6816bc40d007" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Effective Strategies for Data Analytics Projects: Tips and Best Practices for Beginners]]></title>
            <link>https://medium.com/@fokoye/effective-strategies-for-data-analytics-projects-tips-and-best-practices-for-beginners-7172ffbd54c1?source=rss-68ed040cd594------2</link>
            <guid isPermaLink="false">https://medium.com/p/7172ffbd54c1</guid>
            <dc:creator><![CDATA[Fokoye]]></dc:creator>
            <pubDate>Sun, 02 Jul 2023 00:12:45 GMT</pubDate>
            <atom:updated>2023-07-05T08:03:23.520Z</atom:updated>
            <content:encoded><![CDATA[<h3>Emily’s Story</h3><p>Emily had recently been assigned a challenging data analytics project, which happened to be her first hackathon. Excitement and nervousness filled her heart as she realized the immense opportunity to derive valuable insights and make a meaningful impact. However, as the deadline drew closer, Emily started to feel overwhelmed. The pressure to deliver meaningful results within a limited timeframe began to weigh heavily on her shoulders. She found herself staring at a blank screen, unsure of where to start or how to approach the project. In her moment of uncertainty, Emily decided to seek guidance from more experienced colleagues. They emphasized the importance of adopting a systematic approach and following best practices to navigate the challenges of a data analytics project successfully.</p><p>Armed with their advice, Emily began by taking a step back and understanding the project requirements thoroughly. She realized the need to clarify any uncertainties and gain a clear understanding of the objectives, scope, and expectations. Next, Emily focused on defining the domain and identifying the key questions or management needs that the analysis aimed to address. She asked herself the crucial “Who, What, When, Where, and How” questions, aligning her efforts with the desired outcomes of the project.</p><p>Understanding that data quality played a pivotal role, Emily devoted time to explore and understand the data thoroughly. She studied the data dictionary, identifying missing data, duplicates, and any white spaces that might impact the analysis. Recognizing the significance of clean and transformed data, she meticulously carried out the necessary data cleaning and preprocessing steps.</p><p>As Emily delved into the analysis, she faced the crucial decision of selecting the right tools. Considering the complexity of the project, she evaluated options such as SQL, Excel, Python, R, and data visualization tools like Power BI and Tableau. By choosing the tool that aligned with her project requirements and expertise, Emily could enhance efficiency and facilitate better data exploration and analysis.</p><p>With her data organized and her tools in place, Emily embarked on building the structure of her dashboard or report. She outlined the key insights she had observed from the data analysis and utilized visualizations and charts to convey complex information in an understandable format. When recommending strategies to management, she focused on solutions that could increase revenue, reduce costs, or drive business growth. Her recommendations were clear, actionable, and backed by the patterns and trends discovered in the data.</p><p>As the hackathon came to a close, Emily realized that her systematic approach and adherence to best practices had paid off. Despite the initial overwhelm, she had successfully tackled the project and extracted valuable insights. Her hard work and dedication had not only impressed her colleagues but also provided a solid foundation for driving positive outcomes for the organization.</p><p>From that day forward, Emily continued to apply the lessons she had learned. With each new data analytics project that came her way, she approached it with confidence and a strategic mindset. She knew that by adopting a systematic approach, following best practices, and seeking guidance when needed, she could overcome any challenges and unlock the full potential of data analytics.</p><p><strong>Introduction</strong></p><p>Data analytics projects can be exciting opportunities to derive valuable insights and make informed decisions. However, as we can see with the Case of Emily,the pressure to deliver meaningful results within a limited timeframe can be overwhelming. To navigate these challenges successfully, it is crucial to adopt a systematic approach and follow best practices. This article presents a set of tips to help you effectively deal with a data analytics project , enabling you to extract valuable insights and drive positive outcomes.</p><ol><li><strong>Embrace the Pressure: </strong>One of the first steps in tackling a data analytics project is to recognize the pressure associated with the task. Rather than being overwhelmed, view it as an opportunity to excel. Take the time to thoroughly understand the objectives, scope, and requirements of the project. Do not assume anything and clarify any uncertainties with the project stakeholders. A clear understanding of the task will help you stay focused and prioritize your efforts effectively and ease the pressure.</li></ol><p><strong>2. Define the Domain and Identify Key Questions: </strong>Before diving into the analysis, it is important to establish the domain or industry context of the project. Familiarize yourself with the domain-specific terminology, business processes, and challenges.</p><p>Identifying the key questions or management needs is a crucial step in any data analytics project. By asking pertinent questions related to the “Who, What, When, and Where,” you can ensure that your analysis focuses on extracting relevant insights and providing actionable recommendations. Here are examples of questions that can guide your analysis:</p><p><strong>Who:</strong> Who are our target customers? Who are the key stakeholders in this project? Who are the high-value or at-risk customers?</p><p><strong>What:</strong> What are the main factors influencing sales/revenue/profit? What are the primary drivers of customer churn? What are the key trends or patterns in the data? What are the most common customer complaints or issues?</p><p><strong>When:</strong> When do we see seasonal fluctuations in sales? When do customers tend to make repeat purchases? When do we observe spikes in customer complaints or support requests?</p><p><strong>Where:</strong> Where are our customers geographically located? Where are the areas with the highest customer acquisition costs? Where are the bottlenecks or inefficiencies in our processes?</p><p>By addressing these “Who, What, When, and Where” questions, you can narrow down the focus of your analysis, identify the most critical areas to investigate, and generate actionable insights and recommendations. Remember to align these questions with the specific objectives and needs of the project, ensuring that your analysis provides the necessary guidance to drive positive outcomes.</p><p><strong>3. Understand the Data: </strong>Data quality plays a pivotal role in the success of any data analytics project. Start by exploring and understanding the data thoroughly. Study the data dictionary, which provides information about the variables and their meanings. Identify missing data, duplicates, and any white spaces that may affect the analysis. Cleaning and transforming the data are essential steps to ensure accurate and reliable results.</p><p><strong>4. Select the Right Tools: </strong>Determining the appropriate tool for your data analytics project depends on several factors, including the complexity of the project, your specific requirements, and your familiarity with the tools. Here’s a breakdown of when to use each tool.</p><p><strong>Excel:</strong> Excel is a versatile tool that is widely accessible and easy to use. It is suitable for smaller datasets and relatively simple analyses. Excel’s strengths lie in data organization, basic calculations, and creating charts and graphs for visualization. It is commonly used for exploratory data analysis, basic statistical analysis, and creating reports with limited complexity.</p><p><strong>SQL (Structured Query Language):</strong> SQL is ideal for working with structured databases and performing data retrieval, manipulation, and aggregation tasks. It is particularly useful when dealing with large datasets and performing complex queries involving multiple tables. SQL is commonly used for data extraction, transformation, and loading (ETL) processes, as well as for data cleaning and preprocessing.</p><p><strong>Data Visualization Tools (Power BI/ Tableau):</strong> Data visualization tools like Power BI and Tableau are specialized platforms that focus on creating visually appealing and interactive dashboards and reports. These tools provide a user-friendly interface for creating dynamic visualizations, enabling users to explore data and communicate insights effectively. They are commonly used for data storytelling, interactive visualizations, and creating executive-level reports.</p><p><strong>R:</strong> R is a programming language specifically designed for statistical analysis and data visualization. It has a vast collection of packages and libraries dedicated to data analytics and provides a wide range of statistical and graphical techniques. R is commonly used for statistical modeling, data visualization, data exploration, and advanced statistical analysis.</p><p><strong>Python:</strong> Python is a powerful programming language with extensive libraries and frameworks for data analysis and machine learning. It provides flexibility and scalability, making it suitable for a wide range of data analytics tasks. Python is commonly used for data cleaning, preprocessing, statistical analysis, advanced visualization, machine learning, and building complex analytical models.</p><p>When selecting the appropriate tool, consider the complexity of your project, the size of your dataset, the type of analysis you need to perform, the level of interactivity and visualization required, and your own proficiency with the tools. It’s often beneficial to have a combination of tools in your toolkit, allowing you to leverage the strengths of each tool for different aspects of your data analytics project.</p><p><strong>5. Build a Structured Dashboard or Report: </strong>Use visualizations and charts to convey complex information in an understandable format. When recommending strategies to management, focus on solutions that can increase revenue, reduce costs, or drive business growth. Provide clear and actionable recommendations based on the patterns and trends discovered in the data. To present your findings effectively and demonstrate how they can increase revenue, profit, or reduce costs, it is essential to create a structured dashboard or report. Here is a step-by-step approach:</p><p><strong>Outline Key Insights: </strong>Begin by summarizing the key insights derived from your data analysis. Highlight the most important findings that are relevant to the management’s objectives and concerns. These insights should be supported by data-driven evidence and should provide valuable information for decision-making.</p><p><strong>Utilize the Right Charts:</strong> Visualizations and charts play a crucial role in conveying complex information in a clear and understandable format. Select appropriate visualizations, such as line graphs, bar charts, Donut charts, or Treemaps, to represent the patterns and trends discovered in the data. Ensure that the visualizations are well-designed, visually appealing, and effectively communicate the key insights.</p><p><strong>Focus on Revenue, Profit, and Cost Reduction: </strong>When recommending strategies to management, emphasize solutions that have the potential to increase revenue, improve profitability, or reduce costs. For example, if the analysis reveals a specific customer segment with high potential for upselling or cross-selling, suggest targeted marketing campaigns to capture additional revenue. If cost inefficiencies are identified in certain business processes, propose optimization strategies to streamline operations and reduce expenses.</p><p><strong>Provide Clear and Actionable Recommendations:</strong> Based on the patterns and trends discovered in the data analysis, provide clear and actionable recommendations for management to implement. Ensure that your recommendations are specific, feasible, and supported by the data insights. Clearly articulate the expected impact on revenue, profit, or cost reduction and provide a roadmap for implementation, including any necessary resources or support required.</p><p><strong>Tailor the Presentation to the Audience:</strong> Consider the audience when creating your structured dashboard or report. Customize the level of technical detail and use language that is easily understandable by the management team. Focus on the business implications and present the information in a logical and organized manner, making it easy for the audience to grasp the key points and understand the potential impact on the organization’s financial goals. Effective data storytelling skills are crucial for engaging stakeholders and conveying insights. When tailoring presentations to the audience, it is also important to understand their roles and needs. Customize technical detail to strike a balance between relevance and overwhelming information. Organize information logically, using visual aids to enhance comprehension. Incorporate storytelling techniques to make the presentation engaging and memorable. Address questions and concerns to foster understanding. By following these steps and presenting the key insights, along with clear strategy recommendations focused on revenue generation, profitability, and cost reduction, you can effectively demonstrate the value of your data analysis and provide actionable strategies for the management team to implement.</p><p><strong>Conclusion</strong></p><p>Dealing with a data analytics project can be challenging, but by following these tips, you can navigate the process more effectively. Remember to embrace the pressure, understand the task, define the domain, ask the right questions, and thoroughly analyze the data. By selecting the appropriate tools and building structured dashboards or reports, you can provide valuable insights and make recommendations that drive positive outcomes for your organization. With a systematic approach and attention to detail, you can successfully tackle any data analytics project.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7172ffbd54c1" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Manipulative Power of Marketing: Rejecting Consumerism and Embracing Minimalism for Financial…]]></title>
            <link>https://medium.com/@fokoye/the-manipulative-power-of-marketing-rejecting-consumerism-and-embracing-minimalism-for-financial-1afdae64f75b?source=rss-68ed040cd594------2</link>
            <guid isPermaLink="false">https://medium.com/p/1afdae64f75b</guid>
            <dc:creator><![CDATA[Fokoye]]></dc:creator>
            <pubDate>Wed, 03 May 2023 00:42:40 GMT</pubDate>
            <atom:updated>2023-05-03T00:42:40.581Z</atom:updated>
            <content:encoded><![CDATA[<h3><strong>The Manipulative Power of Marketing: Rejecting Consumerism and Embracing Minimalism for Financial Freedom</strong></h3><p>Marketing is an essential component of modern society, driving the global economy and shaping our lives. Marketing is about manipulating the mind to create a need. From billboards and TV ads to social media and personalized push notifications, we are exposed to thousands of marketing messages every day. The power of marketing to influence our behavior and create needs where there are none is both fascinating and concerning. Using technology to track our location and push targeted ads to us is just one way that companies can personalize their marketing messages and increase the chances of making a sale. The pressure to keep up with the latest trends and own the latest gadgets can be overwhelming, leading to people buying things they don’t really need and may not be able to afford. One-click buying and same-day delivery have made it easier than ever to accumulate more stuff, which can lead to hoarding and an unhealthy attachment to material possessions. Deficit advertising is another technique used in marketing, which preys on our insecurities and makes us feel like we are not enough. The message is clear: “Buy our product, and you’ll be happier, more successful, and more attractive”. This constant pressure to maintain a certain image can be exhausting and expensive.</p><p>But what if we took a step back and asked ourselves, “How will my life be better with less?” By embracing minimalism, we can focus on the things that truly matter in life, such as our health, relationships, and finances. We can learn to distinguish between our wants and needs and own things that add value and have a purpose or function. Minimalism is not about deprivation or living with nothing. It’s about owning less and being intentional about what we choose to keep in our lives. It’s about finding true happiness in the things that really matter and defining success on our own terms.</p><p>By taking a single step towards minimalism, we can begin to make progress in that direction. Commencing with decluttering our living space is a fantastic initial move towards a more intentional and minimalist existence. As we discard items that no longer serve us and donate them to those in need, we make room for the things that truly hold importance, and also positively impact the lives of those around us.</p><p>The manipulative power of marketing is real, and it’s essential to be aware of its influence on our lives. By embracing minimalism, we can focus on the things that truly matter and define success on our own terms. It’s about finding a balance between owning things and being owned by them, and ultimately leading a happier, more fulfilling life.</p><p>The financial implications of falling prey to the manipulative power of marketing can be significant. We may find ourselves spending money we don’t have on things we don’t need, all in the pursuit of a fleeting feeling of satisfaction or to keep up with the latest trends. This can lead to debt, financial stress, and even a sense of financial insecurity. On the other hand, embracing minimalism can have a positive effect on our personal finance. By learning to distinguish between our wants and needs, we can reduce unnecessary spending and prioritize our financial goals. We can focus on paying off debt, building an emergency fund, and investing in our future. Through minimalism, we can also find satisfaction in the things that truly add value to our lives, rather than constantly seeking out the next big purchase. This can lead to a sense of financial freedom and security, as we learn to live within our means and focus on the things that truly matter.</p><p>Ultimately, embracing minimalism can have a profound effect on our personal finance, as we learn to prioritize our financial goals, reduce unnecessary spending, and find satisfaction in the simple joys of life. It’s a powerful way to take control of our lives and define success on our own terms.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1afdae64f75b" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>