Lumivero’s Post

i pulled an all-nighter making figures for my research paper. so instead of doing that again, i pulled ANOTHER all-nighter... and built Eliee. Eliee is a no-code, data-visualization tool for researchers who hate matplotlib/R. all you need to do is: 1. upload your data 2. prompt Eliee ["boxplot comparing the main datapoints"] 3. highlight areas you want fixed, and instantly change them 4. export for papers! Eliee is currently in private beta; you can try it at eliee.sh! [#PhD]

Have you used TEXTJOIN function in excel? If you're tired of concatenating cells one by one, this is the game-changer you've been waiting for. It simplifies joining text strings, even with a delimiter! Seriously, it's a huge time-saver. ---------------------------------------------------- Learn Excel in 5 minutes a week 👉 https://lnkd.in/ggwajHJ5 ----------------------------------------------------

Have you used TEXTJOIN function in excel? If you're tired of concatenating cells one by one, this is the game-changer you've been waiting for. It simplifies joining text strings, even with a delimiter! Seriously, it's a huge time-saver. ---------------------------------------------------- Learn Excel in 5 minutes a week 👉 https://lnkd.in/ggwajHJ5 ----------------------------------------------------

🚨 If you use Excel daily, these errors are costing you time. How many times have you seen:
#DIV/0!
#VALUE!
#REF!
#NAME?
#N/A
#NUM! and paused your entire workflow to figure out what went wrong? I created this Excel Error Cheat Sheet to make troubleshooting 10x faster. Instead of guessing, here’s the quick logic: 🔹 #DIV/0! → You’re dividing by zero or a blank cell
🔹 #VALUE! → Wrong data type (text mixed with numbers)
🔹 #NAME? → Misspelled function or undefined name
🔹 #REF! → Deleted or invalid cell reference
🔹 #N/A → Lookup value not found
🔹 #NUM! → Invalid numeric input
🔹 ######## → Column too narrow 💡 Pro Tip: Wrap risky formulas with IFERROR() or IFNA() to keep your reports clean and professional. Whether you're:
• Preparing dashboards
• Cleaning data
• Building financial models
• Practicing for interviews
• Competing in Excel challenges Knowing these errors instantly gives you an edge. Save this post. You’ll need it. 🔔 What Excel error confuses you the most? Let’s solve it together in the comments. #Excel #dataanalysis

I’ve just published a Substack post looking at how we analyse compositional proportion data (where outcomes sum to 1), and why Dirichlet regression is often a better choice than modelling each proportion separately. The post walks through: - why treating proportions as independent outcomes can be misleading - how Dirichlet models respect the trade-offs inherent in compositional data - what you gain analytically when interest allocation, time use, or attention is split across categories I re-examine an existing paper by Liam Satchell as a concrete example, showing how a Dirichlet approach could have offered additional insight into relative changes across components (i.e. because the data structure naturally calls for a joint model). If you work with proportions, shares, or allocations (eye-tracking, time budgets, behavioural coding, survey percentages), this approach is often worth considering. P. S. There is also a little tutorial on Ordered Beta regression using the #ordbetareg package in R Link to the post in the comments 👇 #Dirichlet #brms #proportions #composite #mixedeffects #rstats #eyetracking #substack

A well-designed questionnaire can make the difference between useful insights and misleading data. Here are a few simple rules to avoid common survey design mistakes: 🔹 Keep questions clear and concise. Complex wording leads to confusion and confused respondents give unreliable answers. 🔹 One question, one idea. Double-barreled questions distort results and are hard to interpret. 🔹 Think about the order. The sequence of questions can influence how people respond. 🔹 Test before launching. Even small checks can prevent big data issues later. At YOM, our questionnaire programming focuses on clarity, logic, and data quality, so every survey is built to deliver insights you can trust. Because good research starts with good questions. 📩 Get in touch with us today to discuss your research needs. Your Opinion Matters and we’re here to make sure it’s heard. 

𝐓𝐡𝐢𝐬 𝐒𝐐𝐋 𝐪𝐮𝐞𝐬𝐭𝐢𝐨𝐧 𝐥𝐨𝐨𝐤𝐬 𝐞𝐚𝐬𝐲… 𝐮𝐧𝐭𝐢𝐥 𝐨𝐧𝐞 𝐥𝐢𝐧𝐞 𝐜𝐡𝐚𝐧𝐠𝐞𝐬 𝐞𝐯𝐞𝐫𝐲𝐭𝐡𝐢𝐧𝐠. 𝐌𝐨𝐬𝐭 𝐩𝐞𝐨𝐩𝐥𝐞 𝐜𝐚𝐧’𝐭 𝐬𝐨𝐥𝐯𝐞 𝐭𝐡𝐢𝐬 𝐒𝐐𝐋 𝐪𝐮𝐞𝐬𝐭𝐢𝐨𝐧 𝐮𝐧𝐝𝐞𝐫 𝐩𝐫𝐞𝐬𝐬𝐮𝐫����. “Find cumulative spend per user.” -> 𝘌𝘢𝘴𝘺. “Now reset it after 2+ days of inactivity.” -> 𝘕𝘰𝘵 𝘴𝘰 𝘦𝘢𝘴𝘺 😏 Most people stop at window functions. But the real challenge is: 👉 detecting gaps 👉 creating dynamic groups 👉 then applying cumulative sum 💡 Hint: Use LAG() + DATEDIFF() + running group id 🎯 Takeaway: SQL isn’t hard. Real-world logic is. 👉 Would you solve this in an interview? Follow Mohit K for more SQL interview question #SQL #DataEngineering #InterviewPrep #DataAnalytics #LearnInPublic #Analytics

🔍 For Data Analysts, Researchers & Students, or MEAL professional Seeking a Free SPSS Alternative, I have found PSPP to be a game-changer. PSPP is a GNU open-source and free statistical software designed as a replacement for SPSS, and it delivers most of the functionality many of us rely on—without the licensing cost. 🔹 Why PSPP Stands Out ✅ Completely free & open source ✅ Opens SPSS (.sav), CSV, .ods and other files directly ✅ Familiar SPSS-like interface ✅ Supports both menu-based and syntax-based analysis ✅ Ideal for both Academic research, business data & Survey data analysis 🔹 What You Can Do with PSPP 📊 Descriptive statistics (frequencies, means, SDs) 📊 Crosstabs & Chi-square tests 📊 T-tests & ANOVA 📊 Correlation analysis 📊 Linear & logistic regression 📊 Factor Analysis, etc Are you ready to try this out? 🔗 Installation & documentation here: https://lnkd.in/dbcqyPh7 🔗Basic analysis in PSPP: https://lnkd.in/dP8s5uwG #PSPP #SPSSAlternative #DataAnalysis #MEL #MonitoringAndEvaluation #OpenSource #Statistics #ResearchMethods #NGOData #SurveyAnalysis

I can tell instantly if a plot was generated with Excel or R. Unfortunately, reviewers can tell too and they could reject your paper for that. Many researchers spend thousands on reagents, Spend sleepless nights in the lab, Generate beautiful, complex data. Only to transfer the data to Excel, click "generate bar plot" and then copy and paste it into word. The result of this simple exercise is often a "Pixelated Bar Chart". A pixelated plot is a blurry, non-interactive, low-quality plot. Look at the comparison below. These two plots show the exact same data. Left: Excel-style plot (pixelated): Messy gridlines. Hard to read. It looks like a rough draft. Right: R plot(publication standard): 300 DPI resolution. Clean annotations. It looks like a published paper. If you plan to do impactful research in 2026, you need to understand this difference. You need to understand that high-quality data visualisation is a measure of scientific rigour as much as that literature review you spent weeks doing. If your figures look sloppy, reviewers subconsciously assume your experiments were sloppy too, and it could lead to a rejection. You have two choices to fix this as you continue your research in 2026: 📌 Learn ggplot2 in R to generate vector graphics (I share tips on this weekly). 📌Let Genomac Services and Consult (GSC) handle it for you. At Genomac Services and Consult (GSC), we turn your raw spreadsheets into publication-ready, high-resolution figures. You focus on the discussion; we handle the code. Whichever way you choose to go, always remember that impactful research in 2026 is not just about asking innovative and insightful research questions. How you execute your ideas and present your findings matters too. #DataViz #RStats #Bioinformatics #AcademicPublishing #Genomac #ResearchTips #ScienceCommunication #PhDLife