<?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 Doreen James on Medium]]></title>
        <description><![CDATA[Stories by Doreen James on Medium]]></description>
        <link>https://medium.com/@jamesdoreen610?source=rss-f1f3c47354a5------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*oC3oYnjLOcdB5_PyQuXOoA.jpeg</url>
            <title>Stories by Doreen James on Medium</title>
            <link>https://medium.com/@jamesdoreen610?source=rss-f1f3c47354a5------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 19:19:46 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@jamesdoreen610/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[IMPORTANCE OF DATA CLEANING FPR ACCURATE DATA ANALYSIS]]></title>
            <link>https://medium.com/@jamesdoreen610/importance-of-data-cleaning-fpr-accurate-data-analysis-c2fb4bc77c6a?source=rss-f1f3c47354a5------2</link>
            <guid isPermaLink="false">https://medium.com/p/c2fb4bc77c6a</guid>
            <category><![CDATA[data-analysis]]></category>
            <dc:creator><![CDATA[Doreen James]]></dc:creator>
            <pubDate>Sat, 30 Nov 2024 21:07:12 GMT</pubDate>
            <atom:updated>2024-11-30T21:07:12.812Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/474/1*xZD7fXLEYYdE27EZ5rK3Vw.jpeg" /></figure><p>In today&#39;s data-driven world, the phrase “garbage in, garbage out” holds true for any analysis. The foundation of accurate data analysis lies in clean, reliable, and well organized data. Data Cleaning, often overlooked, is one of the most crucial steps in data analysis process. It ensures that data sets are free from inconsistencies, inaccuracies, and unnecessary elements that could compromise the validity of insights.</p><p>Excel, a versatile and widely-used tool, provides an array of functions and techniques to make data cleaning easier and more efficient. Through this article, I’ll share ,my learning journey of mastering data-cleaning techniques with Excel, personal insights and reflections on the process.</p><p><strong>LEARNING SUMMARY</strong></p><p>Here are key functions and techniques I explored while diving into data cleaning:</p><ol><li>Handling Empty rows and cells: Used filter to identify blank cells quickly and filled them with appropriate values or removed rows if necessary.</li><li>Replacing values and standardizing formats: Leveraged “Find and Replace” to standardize data formats (correcting inconsistent date formats)</li><li>Trimming unnecessary spaces: The TRIM function was a lifesaver for removing trailing and leading spaces in datasets, which can often cause hidden discrepancies.</li><li>Removing Duplicates: Identified and removed duplicate entries using Excel’s built-in Remove duplicates feature, ensuring the dataset’s uniqueness and accuracy.</li><li>Advanced Techniques: —</li></ol><ul><li>Flash fill: Quickly extracted patterns like first names.</li><li>Conditional Formatting: Highlighted outliers and inconsistencies for manual review.</li></ul><p><strong>PERSONAL INSIGHTS</strong></p><p>Throughout the learning process, I had several “aha” moments:</p><ol><li>The power of automation: Discovering Flash Fill was revolutionary. What once took hours of manual work can now be done in seconds.</li><li>Collaborative Problem Solving: Discussing challenges with my accountability partner helped me refine my approach to specific issues, such as dealing with complex text.</li><li>Small Errors Have Big impacts: It was eye-opening to see how a stray space or duplicate entry could significantly skew results, reinforcing the need for meticulous cleaning.</li></ol><p><strong>REFLECTION</strong></p><p>This journey has significantly enhanced my data cleaning knowledge and skills. I now recognize that data cleaning is not just a preparatory step but a critical phase that dictates the quality of analysis and insights. Moreover, the collaborative aspect of this exercise helped me appreciate the value of peer learning, as discussing issues and sharing solutions provided a deeper understanding.</p><p>Let’s take data cleaning seriously to make data analysis powerful and impactful! <a href="https://medium.com/u/61a3fa6ce26c">DataGirls Tanzania</a></p><p>#DataCleaning #ExcelTips #DataScience #DataGirlsTz</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c2fb4bc77c6a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[DATA GIRLS TANZANIA LINKEDIN PERFORMANCE ANALYSIS]]></title>
            <link>https://medium.com/@jamesdoreen610/data-girls-tanzania-linkedin-performance-analysis-a3f8fa2f8bc3?source=rss-f1f3c47354a5------2</link>
            <guid isPermaLink="false">https://medium.com/p/a3f8fa2f8bc3</guid>
            <category><![CDATA[data-analysis]]></category>
            <dc:creator><![CDATA[Doreen James]]></dc:creator>
            <pubDate>Sat, 30 Nov 2024 19:44:28 GMT</pubDate>
            <atom:updated>2024-11-30T19:44:28.773Z</atom:updated>
            <content:encoded><![CDATA[<p>DATA GIRLS TANZANIA LINKEDIN PERFORMANCE ANALYSIS</p><p>As a data analyst trainee for <a href="https://medium.com/u/61a3fa6ce26c">DataGirls Tanzania</a>, I analyzed the organization’s LinkedIn profile to understand content performance, audience interaction, and engagement trends using excel. This analysis aims to provide actionable insights to enhance engagement, increase visibility, and better communicate the mission of empowering young girls in tech.</p><p>DATA CLEANING</p><p>like every data set, the data girls LinkedIn profile data set also needed to be cleaned to remove unnecessary information. Some of the process i did were;</p><ol><li>Removing blanks</li><li>Removing duplicates</li><li>Removing unnecessary columns</li><li>Removing columns with zero values</li><li>Matching column data types to the values present</li></ol><p>Also, i added new columns like the date column, I extracted days, months from the date column so as i can use them in my analysis and visualization.</p><p>VISUALIZATION AND KEY FINDINGS</p><p>Some of the key findings include;</p><p>A. Content Performance</p><ul><li>Most engaging content type: Video</li><li>Average engagement rate: 15%</li></ul><p>B. Audience Interaction</p><p>. Average Likes per post: 21.2</p><p>. Average comments per post: 1.8</p><p>C. Visual insights</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*13PeUdEjJqysQ7sbCYFhCA.png" /><figcaption>Data Girls Tanzania Linked profile Analysis DaSshboard</figcaption></figure><p>RECOMMENDTAIONS</p><ul><li>Increase video content production</li><li>Focus on interactive content that drives comments</li><li>Maintain a consistent posting schedule</li></ul><p>By focusing on video content and maintaining consistent posting schedule, Data girls Tanzania can significantly enhance its LinkedIn engagement and visibility.</p><p>#Datavisualization #PowerOfData #DataGirlsTanzania</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a3f8fa2f8bc3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[PIZZA SALES ANALYSIS]]></title>
            <link>https://medium.com/@jamesdoreen610/pizza-sales-analysis-fa4a5acb74bf?source=rss-f1f3c47354a5------2</link>
            <guid isPermaLink="false">https://medium.com/p/fa4a5acb74bf</guid>
            <category><![CDATA[data-analysis]]></category>
            <dc:creator><![CDATA[Doreen James]]></dc:creator>
            <pubDate>Thu, 22 Aug 2024 22:27:44 GMT</pubDate>
            <atom:updated>2024-08-22T22:27:44.065Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/474/1*_W8TnJayViGL3zu4I8h8zQ.jpeg" /></figure><p>As a newbie in the data analysis world. I was so overjoyed to work and manipulate an exciting project, which of course was all made possible by <a href="https://medium.com/u/61a3fa6ce26c">DataGirls Tanzania</a>. The project was creating a dashboard focused on pizza sales. It is my first ever Power BI project, it got me to put my brain into use, through implementing different analytics tools and skills and also to work and employ the use of different visualization tools.</p><p>I would like to thank <a href="https://medium.com/u/f9f792defbba">Maven Analytics</a> for this wonderful dataset on Pizza sales. I got a chance to dance with it to produce an amazing dashboard.</p><p><strong>CHALLENGE: Annual Cycle of a Pizza Place🍕</strong></p><p>I had to analyze a year-round dataset of an imaginary pizza place orders. The dataset includes</p><p>1. Order Details: Order Date and the time of the order was placed</p><p>2. Pizza Details: Type, price, size and quantity</p><p><strong>Getting to know the Data and Data Cleaning:</strong></p><p>I obtained the csv file provide by Maven Analytics. Which I loaded it in the power query for further examination, to determine anomalies, inconsistences, duplicates or any detail that will cause it to be bad data. The steps done include</p><p>1. Removing duplicates. It is done to ensure only unique and distinct value are present to avoid data inconsistences</p><p>2. Promoted Headers: During importation of data, some of the table’s headers were wrongly imported, so I had to promote headers for that specific table.</p><p>3. Checking for validity of the data if there are no errors or empty cells in the tables.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*56nkj2Jhn6H6pLENLX25fQ.png" /><figcaption>Data Cleaning</figcaption></figure><p><strong>🔍Data Modeling</strong></p><p>This next step helps to answer some of the questions on this specific data set. In data modelling this is where we’ll see relationships of the tables plus the creation a calculated columns which are; “DayOfWeek”, “Day”, “MonthName”📅 and “Hour” and Measures. In Power BI, it has Leveraged us the power of Power Query and DAX functions to transform and model the raw data into actionable insights.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*LIGaUSL3LepRQLhyOdVw5w.png" /><figcaption>Calculated Columns of DayOfWeek, Day, MonthName and Hour</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*IOlcey-0AIazPZWIpRtbwg.png" /><figcaption>Total Sales Measure</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*oeuiOMjou_D8laqKg6EW8Q.png" /><figcaption>Bestselling Pizza Measure</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*RbbZtMlhWrPLL2gNA2Zogw.png" /><figcaption>Sales by Category measure</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*V_ZCm4Qne1fW-mhmPp3NOw.png" /><figcaption>Sales trend Measure</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*NN_QNBDSf2xFgakJ3-pOFg.png" /><figcaption>Sales by Hour measure</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*hK6778p1Tb6gJNHXE8Btsg.png" /><figcaption>Tables Relationship Pane</figcaption></figure><p><strong>📊DATA ANALYSIS</strong></p><p>Data modelling helped me come up with some questions which are beneficial during analyzing and preparing visuals for my data. Some of the questions are</p><p>1. What is the total sales overtime and how the sales trend can be determined overtime</p><p>2. What is the best-selling pizza and what pizza types are driving the most sales</p><p>3. What is the peak sales period on hourly basis.</p><p>By formulating these questions, I was able to perform analysis of the dataset using the calculated measures and columns from data modelling by creating charts like bar, line and Donut charts, which resulted to an interactive dashboard that summarizes the dataset.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mmQFCa_SRv6BrDuKaLnzMw.png" /><figcaption>Pizza Sales Dashboard</figcaption></figure><p><strong>💡INSIGHTS UNCOVERED</strong></p><p>1. Total Sales Overview: the total sales amount in the year 2015 of that pizza place was 817.86k, with a total of 21k orders. This indicates a strong demand for pizzas. The average order of 1.02 suggests that most customers are purchasing a single pizza per order, which could indicate individual or small group purchase.</p><p>2. The sales distribution across different categories show a balanced demand. However, the “classic” and “Supreme” categories seem to be the most popular. This suggests that traditional pizza options remain the top choice for customers.</p><p>3. Best Selling Pizza: The “Barbecue Chicken Pizza”, “Classic Deluxe Pizza”, and “Hawaiian Pizza” are among the top sellers.</p><p>4. Sales by Hour: Also, we see that most orders are placed around 6–9PM, this indicates that dinner time is the busiest period.</p><p><strong>📜RECOMMENDATIONS</strong></p><p>1. Regularly review the menu so as to be able to promote or revamp low performing pizza types. Also, highlighting seasonal and specialty pizzas can help to attract customers looking for something new.</p><p>2. Staff training and staff increase, like hiring temporal staff to operate during the peak hours, so as that services can be offered with problems or delays.</p><p>3. Feedback collection can help the business know how their performing also be easy to know which kind of population they are dealing with.</p><p><strong>✔CONCLUSION</strong></p><p>Leveraging the use of Power BI, I was able to analyze key metrics for the pizza place. From sales trend to understanding customer preferences, data driven insights guided strategic decisions.</p><p>😄I welcome everyone to share their thoughts and ideas on these insights, and what can be improved, changed or eliminated. Let’s connect and learn from each other on LinkedIn through:</p><p><a href="https://www.linkedin.com/posts/doreen-james-231ba71b1_datagirlstanzania-dataanalysis-mavenanalytics-activity-7232507392700088320-qNzS?utm_source=share&amp;utm_medium=member_desktop">https://www.linkedin.com/posts/doreen-james-231ba71b1_datagirlstanzania-dataanalysis-mavenanalytics-activity-7232507392700088320-qNzS?utm_source=share&amp;utm_medium=member_desktop</a></p><p>TAHNK YOU</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fa4a5acb74bf" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[DATA CLEANING IN POWER BI.]]></title>
            <link>https://medium.com/@jamesdoreen610/data-cleaning-in-power-bi-cc5c236e832c?source=rss-f1f3c47354a5------2</link>
            <guid isPermaLink="false">https://medium.com/p/cc5c236e832c</guid>
            <dc:creator><![CDATA[Doreen James]]></dc:creator>
            <pubDate>Wed, 21 Aug 2024 11:35:07 GMT</pubDate>
            <atom:updated>2024-08-21T11:35:07.178Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JwL-SsHkWFhwNAuYjPMpxQ.jpeg" /><figcaption>Data Cleaning</figcaption></figure><p>The concept Data cleaning is an essential part of data analysis. It involves identifying and correcting errors, inconsistencies, and inaccuracies in the dataset to ensure that the data is accurate, complete, and ready for analysis.</p><p>In this guide, we will document the data cleaning process performed on a dataset in Power BI using Power Query, focusing on the issues identified, the methods used to address them, and the resulting changes. I will use the “<a href="https://docs.google.com/spreadsheets/d/1Gq-acPg4sJJ-TueU7rLYeNUuR7HZ-M8Q/edit?usp=sharing&amp;ouid=117178077226169901442&amp;rtpof=true&amp;sd=true">US_Regional_Sales_Data</a>”. It is an excel file consisting of six sheets which are; Customers Sheet, Products Sheet, Regions Sheet, Sales Sheet, Sales Team Sheet and Store Locations Sheet.</p><p><strong>Loading Data in Power BI:</strong></p><p>First thing first, when performing data cleaning with Power BI you have to load the data into the power bi desktop. This can be done by opening Power BI Desktop, go to “<strong>Home tab</strong>” then click the<strong> <em>Get data</em></strong> icon to open a drop down list, you’ll choose the excel workbook that you will be working with from the list.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*KS6Ul-fQlRxmllKFLUBllg.png" /><figcaption>Gettin Data into the Work book</figcaption></figure><p>After selecting the excel work book. Make sure to choose the tables and do a thoroughly preview of those tables, after that click on the transform button to load your data in power BI.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/826/1*rOOq21f8WYZVNau5iKW4MA.png" /><figcaption>Loading data into the power query editor.</figcaption></figure><p>The process of Transforming data can be done when loading the dataset that you want work with by clicking the “Transform Data” button as seen above or it can be done, by first clicking the load button, then after the data is loaded. You can transform it in the power query by going to the “Transform” tab.</p><p>In my case I clicked the “Transform Data” button so as it can reduce some of the errors, I’ll find in the data set, that helps me do my work quicker since I have to check for some errors that the editor couldn’t identify. So, my data set will appear as follows.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*UWqa1U6netTmWDj6tOtmWQ.png" /></figure><p>On the bottom we can see words written “column profiling based on top 1000 rows”. We have to change that to “column profiling based on entire data set”. This will perform data cleaning on the entire dataset since by default power query is set to perform cleaning based on the top 1000 rows.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/544/1*C41JThjIEzxED1QH1X55zg.png" /><figcaption>Column Profiling</figcaption></figure><p><strong>BELOW ARE SOME ACTIONS I TOOK DURING THE DATA CLEANING PROCESS:</strong></p><p><strong>I)</strong> <strong>REMOVING DUPLICATES</strong>: The first step I took, was removing any kind of duplicated data so as to narrow down my data set and remain with unique data. You can remove duplicates by going to home tab, the choose “remove rows” once clicked you’ll get a drop-down menu and in that you see “remove duplicates”. Once clicked the power query will remove the duplicates and you’ll see the changes to the right side of your power query in the “Applied Steps” section.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*enNhqHlXh32CFVXNZCcH1w.png" /><figcaption>AFTER REMOVING DUPLICATES</figcaption></figure><p><strong>II) REMOVING ERRORS:</strong> The Sales table has an error on the “order date” column, to remove errors I hovered to the Order Date column with error and click the remove errors dialog that appeared.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/889/1*c_cV-8mKbduDT0tWsdNtdQ.png" /><figcaption>BEFORE REMOVING AN ERROR</figcaption></figure><p>After removing the errors, it will appear as follows. The changes made can be noticed on the Applied steps section on the right-hand side of the power query.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*rMAC7TsMWN-aBfqmZTnR3A.png" /><figcaption>AFTER REMOVING AN ERROR</figcaption></figure><p><strong>III) REMOVING EMPTY: </strong>On the same table Sales, on the “Order date” column the was an empty cell, so when you hover your cursor on the green line, you’ll see a shade of grey on that column, when you place the cursor there a dialog box remove empty will appear. If you click it, the cell contain null will be removed.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*dS-03-NVxCa0nBs-JHcSag.png" /><figcaption>BEFORE REMOVING EMPTY</figcaption></figure><p>After the changes. We will notice the gray color under the header “Order Date” is gone, that means the empty cell is no longer present.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*CzfVSYWP4RArgbNQzIKQSQ.png" /><figcaption>AFTER REMOVING EMPTY</figcaption></figure><p><strong>IV) RENAMING COLUMNS:</strong> Some of the columns in various tables contained underscores in their headers, so the best practice is to remove the underscores and replace them with space.</p><p>We can see the “SalesTeamID” column on the Sales Table column has an underscore at the begin. I replaced that with space by selecting and right clicking that column and choose the<strong> “Rename”</strong> option on the drop down menu of actions to be done on the column. And the outcome will be seen on the second picture. And I did the same step with other columns with the underscores in their Headers.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*hMcu1P9iTeMVBaPMzk6FNg.png" /><figcaption>BEFORE RENAMING THE SALES TEAM ID COLUMN</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*UKNR9yjOkcCKO26tBMtMlA.png" /><figcaption>AFTER RENAMING THE SALES TEAM ID COLUMN</figcaption></figure><p><strong>V) REPLACING VALUES:</strong> On “Sales” table in the column “Sales Channel”. I replaced On line with Online, In__Store with In-Store, Dis_tributor with Distributor and Whole#_sale with Wholesale, in order to have uniform names of the channel since they all meant the same thing. Select the column then right click, on the drop-down menu select <strong>“Replace Values”</strong> and another dialog will open and type the value to find and the value you want to replace with then click OK to perform the changes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/811/1*RBo7KyiODMdJxGaEyukkRw.png" /><figcaption>Values to replaced</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*uDHXjwY8LgV01WAHy5toKQ.png" /><figcaption>Replacing misspelled values with appropriate values</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ayTGPUK4pcax8SkcEdwbkQ.png" /><figcaption>Correct values</figcaption></figure><p><strong>VI) CLOSE AND APPLY:</strong> After performing data cleaning, you need to apply the changes you have modified in the data set, by selecting the close and apply icon on the <strong>“home tab”</strong> this will close the Power Query Editor and loads the applied changes to Power BI desktop.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/806/1*2-kSTC7CUy08HeUFHu3LbA.png" /><figcaption>CLOSE AND APPLY CHANGES</figcaption></figure><p>Other useful tools that can enhance you data cleaning process in Power BI include, hovering your cursor on the green line or using the “column quality” tab on the <strong>View tab</strong> can help you know the validity, error or empty status of your data set in respective columns. So, this feature in Power query has simplified the cleaning process in Power BI.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*08MEswrieGqm1bXjeAcj9g.png" /><figcaption>Valid, Error and Empty Status of the Dataset</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*zYuND6yCHry_3arPu2KPWw.png" /><figcaption>Column Quality</figcaption></figure><p>This guide provides a comprehensive overview of the data cleaning steps performed in Power BI, complete with screenshots to illustrate the before-and-after states of the data. By following these steps, you can ensure that your datasets are well-prepared for insightful and meaningful analysis in Power BI.</p><p>Happy cleaning everyone, I cannot wait for your useful input.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cc5c236e832c" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>