[Data Analysis] Presenting findings (9/9)
A guide for beginner Data Analysts on presenting analysis results. Covers organising findings, creating visualisations with Python, structuring a presentation, crafting a narrative, and preparing for questions
[This guide is part 9 of an 9-article walkthrough.]
Key concepts:
Visual Studio Code · Python · Jupyter Notebook · Data analysis · Data cleaning · Data portfolio piece · Presentation
Welcome to the final part of the 9-part series! You’ve cleaned your Iris dataset, performed your analysis, and now it’s time to showcase your findings. In this guide, we’ll walk through creating a compelling presentation of your Iris dataset analysis results. Let’s dive in!
To remind ourselves where in the process data cleaning comes into play, here is a general outline of the data analysis process:
- Define Objectives: Clearly understand the goals of your analysis.
- Data Acquisition: Obtain the dataset you’ll be working with.
- Data Exploration: Explore the dataset to get an initial understanding of its structure and content.
- Data Cleaning: Preprocess the data to ensure its quality and consistency.
- Data Visualization: Create visualizations to gain insights into the data.
- Feature Engineering: Create new features or transform existing ones to enhance the dataset’s predictive power.
- Statistical Analysis (if applicable): Conduct statistical tests or analyses to answer specific questions or hypotheses.
- Machine Learning (if applicable): Split the data into training and testing sets, select an appropriate algorithm & train and evaluate the model’s performance using metrics like accuracy, precision, recall, or F1-score.
- ➡️ Present solution: Interpret the findings in the context of your objectives. Document your analysis process and create a report or presentation summarising your analysis.
In this guide, we will be looking at step 9: presentation of results.
Step 1: Organize Your Findings
Before we jump into creating slides, let’s organize our key findings. Open your Jupyter notebook in VS Code and run through your analysis.
Note down:
- The main objective of your analysis
- Key statistics about the dataset
- Any visualisations you created
- Insights or patterns you discovered
- Conclusions drawn from your analysis
If you’ve been following this 9-article walkthrough, you should already have the above prepared.
Step 2: Choose Your Presentation Tool
For this guide, we have used Matplotlib & Seaborn to create our visualisations within VS Code. We will then use Canva to create our presentation, where we will upload all of our visualisations.
❗️Note: You can use any presentation software you’re comfortable with to assemble your slides (PowerPoint, Google Slides, Keynote, etc.).
Step 3: Visualization Best Practices
Creating effective visualizations is crucial for communicating your findings clearly. Here are some best practices to keep in mind when designing your charts:
Choose the Right Chart Type
- Use bar charts for comparing categories
- Use line charts for showing trends over time
- Use scatter plots for showing relationships between two variables
- Use pie charts sparingly, and only for showing parts of a whole
Keep It Simple
- Avoid cluttering your charts with unnecessary elements
- Use clear, easy-to-read fonts
- Remove gridlines if they’re not essential
Use Color Effectively
- Choose a consistent color palette throughout your presentation
- Use color to highlight important data points or categories
- Ensure sufficient contrast between colors for readability
- Consider colourblind friendly palettes
# Example: Using a colourblind friendly palette
sns.set_palette(“colorblind”)
Label Clearly
- Always include a clear title that explains what the chart shows
- Label axes clearly and include units where applicable
- Use legends when necessary, but try to label data directly if possible
# Create figure and set size
plt.figure(figsize=(10, 6))
# Plot data
sns.boxplot(x=’species’, y=’sepal_length’, data=df)
plt.title(‘Distribution of Sepal Length by Species’)
# Set labels
plt.xlabel(‘Species’)
plt.ylabel(‘Sepal Length (cm)’)
# Display datapyth
plt.show()
Scale Appropriately
- Start y-axes at zero for bar charts to avoid misrepresentation
- Use consistent scales when comparing multiple charts
- Consider using logarithmic scales for data with large ranges
Highlight the Data
- Minimise chart junk (unnecessary decorative elements)
- Use data-ink ratio principle: maximize the ink used to present data
Order Data Logically
- Order categories in bar charts by value for easier comparison
- In time series, always order data chronologically
Provide Context
- Include a brief explanation or key takeaway with each chart
- Relate the visualization back to your analysis objectives
Remember, the goal of data visualisation is to communicate your findings clearly and effectively. By following these best practices, you’ll create charts that are not only visually appealing but also informative and easy to understand.
Now, let’s move on to creating the presentation structure
Step 4: Create Your Presentation Structure
Now that we have our visualizations, let’s structure our presentation:
- Title Slide: Include the dataset name, your name, and the date.
- Table of Contents: Show the audience what you plan to cover.
- Project Overview: Briefly explain the Iris dataset and your analysis objectives.
- Dataset Overview: Present key statistics about the dataset.
- Key Findings: List the main insights you discovered. Include the visualizations you created, with brief explanations.
- Conclusion: Summarise your findings and their implications.
- Next Steps: Depending on the nature of your analysis, suggest potential further analyses or applications.
- References: If you have referred to any articles, books or authors whilst making your presentation, make sure to mention them here.
Step 5: Craft Your Narrative
As you assemble your slides, focus on telling a story with your data. Here’s an example of how you might structure your narrative:
- “The Iris dataset contains measurements of 150 iris flowers from three species.”
- “Our objective was to identify distinguishing features between these species.”
- “Looking at the pairplot, we can see clear separation between species, particularly when considering petal length and width.”
- “The box plot shows that sepal length varies significantly between species, with Iris Setosa having the shortest sepals.”
- “The violin plot reveals that petal width is a strong differentiator, with Iris Setosa having notably narrow petals.”
Step 6: Practice Your Presentation
Once your slides are ready, practice delivering your presentation. Focus on:
- Clearly explaining each visualization
- Highlighting key insights
- Connecting your findings back to your initial objectives
- Speaking confidently and at a steady pace
Step 7: Prepare for Questions
Anticipate potential questions your audience might ask. Some examples could be:
- “How did you handle any outliers in the dataset?”
- “What machine learning models could be applied to this data?”
- “How might these findings be applied in botany or horticulture?”
Having thoughtful answers prepared will demonstrate your thorough understanding of the analysis.
Conclusion:
Congratulations! 🎉 Creating an effective presentation is a crucial skill for any Data Analyst. By following these steps, you’ll be able to showcase your analysis in a clear, engaging, and professional manner. Remember, practice makes perfect, so don’t be afraid to refine your presentation skills over time.
Happy presenting, and may your data always tell a compelling story!
Reference material
- Canva presentation for the Iris Dataset Analysis [link]