# Data Analysis Skills Test and Verified Score > A data analysis skills test on reasoning with real data, plus live case sessions and connected notebooks on GitHub. See what proves data analysis skills. Source: https://getlemma.io/skills/data-analysis A dashboard screenshot shows the output, not the judgement behind it. Lemma records how you frame a question, handle messy data and defend a conclusion, through live sessions, assessments and work you connect. How data analysis is assessed ## What data analysis competence looks like A competent analyst starts with the decision, not the dataset. They clarify what question matters, what would change someone's mind and which metric actually measures it. Then they check the data: missing values, duplicates, odd distributions and definitions that changed partway through the period. Their conclusions come with the right amount of doubt. They know when a difference is noise, when correlation is being mistaken for cause and when a segment is too small to trust. They choose charts that make the point plainly and write a summary a busy stakeholder can act on without reading the notebook. ### Framing the question Turning a vague request into a specific question, a metric and a decision it informs. ### Data cleaning Finding missing values, duplicates, outliers and definition changes before they distort results. ### Exploratory analysis Distributions, segments and trends explored systematically rather than by scrolling until something looks interesting. ### Statistics in practice Sampling, variance, significance and the limits of correlation, applied to real business questions. ### Visualisation Choosing charts that make one point clearly and avoiding scales or encodings that mislead. ### Communicating findings Writing summaries with a clear recommendation, stated caveats and the evidence behind them. ## What counts as evidence of data analysis skills? In data analysis, a polished result can hide a shaky method. Lemma records the evidence behind the method: sessions where someone followed your reasoning, an adaptive test and the repositories you link, notebooks included. Each source is listed below. See the [evidence methodology](https://getlemma.io/methodology#evidence-types) for how each source is weighed. [Live session](https://getlemma.io/product/sessions) ### Working a messy dataset live A practitioner gives you a dataset and a business question. They watch how you check quality, explore and reach a conclusion, and rate the reasoning as well as the answer. [GitHub connector](https://getlemma.io/product/passport) ### Notebooks and analysis on GitHub Connect a repository with notebooks, scripts or a write-up from an analysis you did. Lemma checks the repository exists and that your account can access it. [Adaptive assessment](https://getlemma.io/product/assessments) ### Scenario questions on interpretation Adaptive questions present realistic results, such as a misleading average or a confounded trend, and ask for the right reading. Technical levels add debugging of broken analysis code. [Live session](https://getlemma.io/product/sessions) ### Presenting findings to a peer In a session you explain an analysis and its recommendation to a practitioner who challenges it. How you handle those challenges is rated and recorded. ## How is a data analysis skills test run on Lemma? The data analysis test asks 15 to 25 questions and moves up or down a level with each answer, focusing on judgement calls such as leakage, skew and bad joins. It feeds your data analysis SCI together with sessions and linked work, and the score weighs how fresh that evidence is. [How the SCI score is calculated](https://getlemma.io/product/sci)[How adaptive skills assessments work](https://getlemma.io/product/assessments)[Practise data analysis in live skill swap sessions](https://getlemma.io/product/sessions)[Which evidence types count](https://getlemma.io/methodology#evidence-types)[What your Skills Passport shows](https://getlemma.io/product/passport) 1. ### Sharpen your statistics The AI tutor gives exercises on sampling, distributions and interpreting results, and asks you to explain answers in your own words. Drills add only a small, capped amount to your practice score. 2. ### Analyse with a practitioner Book a live session to work through a real dataset or review an analysis together. Both of you rate the session, and the completed session counts as data analysis evidence. 3. ### Take the analysis assessment Scenario questions on cleaning, interpretation and statistics adapt to your level. At higher difficulty you fix broken analysis code. Time limits and no going back keep the score fair. ## Data analysis skills: common questions ### What should a data analyst portfolio include? A few complete analyses beat many charts. Each should state the question, show how you cleaned and checked the data, explain the method and end with a recommendation and its caveats. Connect those notebooks to Lemma and walk a practitioner through one in a live session to add a rated review. ### How can career switchers prove data analysis skills? Evidence matters more than your previous title. Practise with the AI tutor, then work through real datasets in live sessions where practitioners rate your reasoning. Pass the adaptive assessment and connect your best analysis. Your Passport then shows analytical work done now, not a job you held before. ### Are analysis tools like Excel or Tableau assessed? The skill focuses on reasoning: framing, cleaning, interpretation, statistics and communication. Those apply whichever tool you use. In live sessions you can work in a spreadsheet, a BI tool, SQL or Python. If you use SQL or Python heavily, build evidence in those skills too. ### How do hiring managers check an analyst's judgement? They usually give a case study or ask you to walk through a past analysis. They listen for how you questioned the data and handled uncertainty. Lemma's sessions work the same way, and the practitioner's rating records that judgement so you can share it before the interview. ## Skills related to data analysis [Browse the skills directory](https://getlemma.io/skills) [### SQL Prove you can write correct, fast queries on real schemas, from joins to window functions and query plans. Evidence for SQL](https://getlemma.io/skills/sql)[### Python Show how you write, debug and structure Python, not just that you once finished a course. Evidence for Python](https://getlemma.io/skills/python)[### Machine learning Show how you frame, train, evaluate and debug models, with evidence a practitioner has reviewed live. Evidence for machine learning](https://getlemma.io/skills/machine-learning)[### Data engineering Evidence that your pipelines are reliable, testable and recoverable, reviewed by practitioners in live sessions. Evidence for data engineering](https://getlemma.io/skills/data-engineering) ![Painting of an olive grove in soft sunlight with hills beyond](https://cdn.sanity.io/images/islu4peo/production/613eb3f93566423c9adf45246f2cf1a61437c04c-1680x1120.jpg?w=1080&fm=webp&q=68&fit=max) ## Start your data analysis evidence record. Lemma is in early access. Request an invite and we will email you when your place is ready.