https://datacamp-community-prod.s3.amazonaws.com/e63a8f6b-2aa3-4006-89e0-badc294b179c

This tidyverse cheat sheet for beginners will help you to find your way around the well-known packages dplyr and ggplot2!

Tidyverse Cheat Sheet For Beginners

https://s3.amazonaws.com/assets.datacamp.com/blog_assets/Tidyverse+Cheat+Sheet.pdf

https://datacamp-community-prod.s3.amazonaws.com/48093c40-5303-45f4-bbf9-0c96c0133c40

This Jupyter Notebook cheat sheet will help you to find your way around the well-known Jupyter Notebook App, a subproject of Project Jupyter.

Jupyter Notebook Cheat Sheet

https://s3.amazonaws.com/assets.datacamp.com/blog_assets/Jupyter_Notebook_Cheat_Sheet.pdf

https://www.datacamp.com/community/podcast/critical-thinking-in-data-science

<p>Debbie Berebichez, Chief Data Scientist at Metis, speaks with Hugo about what type of critical thinking skills are necessary in data science.</p>

Debbie Berebichez speaks with Hugo about the importance of critical thinking in data science.

Critical Thinking in Data Science

https://www.datacamp.com/community/blog/critical-thinking-in-data-science

https://www.datacamp.com/community/podcast/credibility-crisis-in-data-science

<p>Skipper Seabold, a Director of Data Science at Civis Analytics, discusses the current and looming credibility crisis in data science.</p>

Skipper Seabold, a Director of Data Science at Civis Analytics, discusses the current and looming credibility crisis in data science.

The Credibility Crisis in Data Science

https://www.datacamp.com/community/blog/credibility-crisis-in-data-science

https://www.datacamp.com/community/news/the-credibility-crisis-in-data-science-with-skipper-seabold-6bbmgpwy1qm

The Credibility Crisis in Data Science (with Skipper Seabold)

https://www.datacamp.com/community/news/video-tidyverse-exploratory-analysis-4cvm3bp8akn

In this video tutorial, you'll be given a brief introduction to the concept of the tidyverse and then walk through an exploratory analysis.

Video: Tidyverse Exploratory Analysis

https://www.datacamp.com/community/news/choosing-r-or-python-for-data-analysis-an-infographic-rhcw8x3m5wa

Wondering whether you should use R or Python for your next data analysis post? Check our infographic "Data Science Wars: R vs Python".

Choosing R or Python for Data Analysis? An Infographic

https://www.datacamp.com/community/news/understanding-blockchains-by-coding-one-in-r-l9k0ynh6ho8

Implement a small blockchain in R and learn more about what a blockchain looks like and some of the core concepts behind it!

Understanding Blockchains by Coding One in R

<h2 id="two-and-a-half-years-ago-ai-singapore-aisg-selected-datacamp-https-www-datacamp-com-community-blog-ai-singapore-partnership-as-your-preferred-online-training-partner-what-milestones-have-been-reached-since-then-">Two and a half years ago, <a href="https://www.datacamp.com/community/blog/ai-singapore-partnership">AI Singapore (AISG) selected DataCamp</a> as your preferred online training partner. What milestones have been reached since then?</h2>
<p>Since the DataCamp collaboration was announced in August 2018, the AI for Students (AI4S)® programme has reached more than 21,000 students and the AI for Industry (AI4I)® programme has allowed over 5,000 working professionals to benefit from DataCamp as a learning resource to gain proficiency in Python and programming basic AI and data applications. As compared to traditional learning styles, DataCamp’s self-directed learning model works very well for working professionals, particularly in today’s digital economy.</p>
<h2 id="how-has-the-covid-19-pandemic-impacted-online-learning-for-aisg-s-learners-">How has the COVID-19 pandemic impacted online learning for AISG’s learners?</h2>
<p>We saw an impact in three areas:</p>
<ul>
<li>There was a heightened increase in requests for AI4S access in the early stages of the pandemic.</li>
<li>There has been a high take-up rate with working professionals too, as many were working from home during Singapore’s COVID lockdown (Circuit Breaker).</li>
<li>In terms of success, we saw an extraordinarily high completion rate of 45% for the first four batches of AI4I participants. This rate is much higher than the <a href="https://www.insidehighered.com/digital-learning/article/2019/01/16/study-offers-data-show-moocs-didnt-achieve-their-goals#:~:text=Among%20all%20MOOC%20participants%2C%203.13,percent%20the%20two%20previous%20years.">average MOOC completion rate</a>.</li>
</ul>

<h2 id="what-do-you-like-about-datacamp-">What do you like about DataCamp?</h2>
<p>Firstly, DataCamp’s pedagogy of short two-to-three-minute videos followed by hands-on coding exercises embedded as part of AI4I works, and it’s driving the high completion rate. We like that the DataCamp modules have a very user-friendly interface and are interactive and engaging for learners. DataCamp integrates with LMS (Learning Management Systems) and has a sandbox environment which enhances hands-on practice sessions.</p>
<p>The lesson structure is concise and straightforward and it helps that the content provided is delivered in a bite-sized format, which makes it easy for learners to digest. Other notable mentions include a whole suite of courses available, regular additions of new courses and updates, access to experts from the data science and analytics community, as well as real datasets for learners to work on.</p>
<p>The courses provided cater to a wide range of learners; in particular, they are  relevant to working professionals. As they are accessible online, busy working professionals can learn at their pace and will just need to set aside one or two hours to learn, freeing up more time as compared to having to spend time commuting from one location to another.</p>
<p>We particularly enjoy the gamification of lessons which provides an excellent and experiential way to learn, thus helping the students to progress through the different levels of difficulty.</p>
<h2 id="what-are-some-exceptional-results-you-ve-seen-since-implementing-datacamp-">What are some exceptional results you’ve seen since implementing DataCamp?</h2>
<p>DataCamp is the preferred platform for our learners compared to other content types. To date, feedback has been very positive. Users reported that they continue to use DataCamp as a useful learning and reference tool to keep them updated.</p>
<p>Here are two examples:</p>
<h3 id="ai-for-students-ai4s-">AI for Students (AI4S)®</h3>
<p>This programme takes secondary and tertiary students to the next stage of data fluency by equipping them with programming and data skills. To support their learning journey, educators from the Ministry of Education’s public schools can tap into AISG’s partnership with DataCamp to leverage the latter’s learning platform for teaching purposes. Students can also embark on independent learning in the lessons under AI4S which include learning Python programming and Git, which are necessary skills to get started in working with AI.</p>
<p>Since AI4S was launched in November 2018, it has reached more than 21,000 students across Singapore’s public schools.</p>
<h4 id="feedback-from-an-educator-and-student">Feedback from an educator and student</h4>
<p><center><a href="https://www.aisingapore.org/"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1614606957/2_quotes_1_jlv8k8.png" /></a><br></p>
<h3 id="ai-for-industry-ai4i-">AI for Industry (AI4I)®</h3>
<p>This programme is open to industry professionals such as technical executives, managers, and developers, as well as undergraduates. AI4I introduces participants to AI concepts and use cases, and equips them with the programming skills to build data and AI applications. It is hosted on AISG’s online AI Makerspace platform and leverages DataCamp for the learning resources that participants need to complete the programme.</p>
<h4 id="feedback-from-ai4i-participants">Feedback from AI4I participants</h4>
<p><center><a href="https://www.aisingapore.org/"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1614610004/4_quotes_2_hbsz3c.png" /></a><br></p>
<h2 id="what-would-you-tell-others-who-might-be-considering-partnering-with-datacamp-">What would you tell others who might be considering partnering with DataCamp?</h2>
<p>Go for it, get on the DataCamp platform today! DataCamp is the way to go. It’s a great learning platform for anyone who would like to build or deepen their data skills. It’s tailored for your needs and you get to learn at your own pace.</p>
<p>If you’re interested in learning more about AI Singapore and our talent programmes, please visit <a href="https://www.aisingapore.org/talentdevelopment">https://www.aisingapore.org/talentdevelopment</a>.</p>


https://www.datacamp.com/community/blog/AI-Singapore-2021

Q&A with Mr. Koo Sengmeng, Deputy Director of AI Innovation, AI Singapore

Blazing the Trail in Developing AI Literacy Across Singapore | DataCamp

Blazing the Trail in Developing AI Literacy Across Singapore

<p>DataCamp Digest is our monthly roundup aimed at providing the most up-to-date insights and news on all things data science. In this edition, we discuss why it’s hard to become data-driven, the traits of future-proof organizations, how to adopt a product mindset to building dashboards, and more.</p>
<p> <br><center><a href="https://www.datacamp.com/groups/business"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1612888906/insights_qhc4kg.png" /></a></p>
<h3 id="-why-is-it-so-hard-to-become-a-data-driven-company-https-hbr-org-2021-02-why-is-it-so-hard-to-become-a-data-driven-company-harvard-business-review"><a href="https://hbr.org/2021/02/why-is-it-so-hard-to-become-a-data-driven-company">Why is it so hard to become a data-driven company?</a> | Harvard Business Review</h3>
<p>In 2021, to be a successful company is to be a data-driven company. This is why it’s no surprise that the majority of fortune 1000 organizations are investing in data and artificial intelligence initiatives. Despite this investment, only a few are reporting actual business impact. Find out how data culture and skills is the primary impediment to scaling value with data science, and what to do about it.  </p>
<h3 id="-how-to-build-data-literacy-at-your-company-https-mitsloan-mit-edu-ideas-made-to-matter-how-to-build-data-literacy-your-company-mit-sloan"><a href="https://mitsloan.mit.edu/ideas-made-to-matter/how-to-build-data-literacy-your-company">How to build data literacy at your company</a> | MIT Sloan</h3>
<p>Keeping with the same theme of building a data culture, this article deep dives into building data literacy programs at an organization. From assessing data skills, to building a culture of learning, find out the key components of a successful data literacy program.</p>
<p> <br><center><a href="https://www.datacamp.com/groups/business"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1612888907/Editor_s_Picks_nj17hh.png" /></a></p>
<h3 id="-the-nine-traits-of-future-ready-companies-https-www-mckinsey-com-business-functions-organization-our-insights-the-nine-traits-of-future-ready-companies-mckinsey"><a href="https://www.mckinsey.com/business-functions/organization/our-insights/the-nine-traits-of-future-ready-companies">The nine traits of future-ready companies</a> | McKinsey</h3>
<p>A fun and interactive infographic on the key traits of future-proof organizations. Hint: Data and learning are fundamental pillars for future-proof organizations to accelerate growth.</p>
<h3 id="-why-are-video-calls-so-tiring-you-might-be-misreading-cultural-styles-https-stackoverflow-blog-2021-02-11-why-are-video-calls-so-tiring-you-might-be-misreading-cultural-styles-the-overflow"><a href="https://stackoverflow.blog/2021/02/11/why-are-video-calls-so-tiring-you-might-be-misreading-cultural-styles/">Why are video calls so tiring? You might be misreading cultural styles</a> | The Overflow</h3>
<p>One of our favorite reads of the month, this article dives into different communication styles and how they can improve or worsen remote work.</p>
<p> <br><center><a href="https://www.datacamp.com/groups/business"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1612889346/Inspiration2_wxr25i.png" /></a></p>
<h3 id="-one-question-to-make-your-data-project-10x-more-valuable-https-blog-narrator-ai-one-question-to-make-your-data-project-10x-more-valuable-narrator"><a href="https://blog.narrator.ai/one-question-to-make-your-data-project-10x-more-valuable/">One question to make your data project 10x more valuable</a> | Narrator</h3>
<p>Find out how to drive better alignment and impact on data analysis requests by asking one important question.</p>
<h3 id="-how-to-make-dashboards-using-a-product-thinking-approach-https-shopify-engineering-make-dashboards-using-product-thinking-approach-shopify"><a href="https://shopify.engineering/make-dashboards-using-product-thinking-approach">How to Make Dashboards Using a Product Thinking Approach</a> | Shopify</h3>
<p>Dashboards are one of the simpler and most impactful data products to build at an organization. In this article, Shopify Data Scientist Lin Taylor outlines a product thinking approach to maximize impact when developing dashboards for the organization.</p>
<h3 id="-why-external-data-should-be-part-of-your-data-strategy-https-mitsloan-mit-edu-ideas-made-to-matter-why-external-data-should-be-part-your-data-strategy-mit-sloan"><a href="https://mitsloan.mit.edu/ideas-made-to-matter/why-external-data-should-be-part-your-data-strategy">Why external data should be part of your data strategy</a> | MIT Sloan</h3>
<p>As organizations start making the most of their data for analytics and machine learning, they will need to begin leveraging external datasets to refine their models and analytics outputs. Find out more about the importance of external datasets in this article.</p>
<p> <br><center><a href="https://www.datacamp.com/groups/business"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1614608698/Training_and_Resources_yh7mmx.png" /></a></p>
<h3 id="-data-observability-building-data-quality-monitors-using-sql-https-towardsdatascience-com-data-observability-in-practice-using-sql-755dc6421f59-barr-moses-ryan-kearns"><a href="https://towardsdatascience.com/data-observability-in-practice-using-sql-755dc6421f59">Data Observability: Building Data Quality Monitors Using SQL</a> | Barr Moses &amp; Ryan Kearns</h3>
<p>In this series of articles, Barr Moses and Ryan Kearns outline how to build data quality monitors in SQL to identify freshness and anomalies (Part 1), and extract the root cause of these data anomalies (Part 2).  </p>
<h3 id="-interpretable-machine-learning-https-christophm-github-io-interpretable-ml-book-index-html-christoph-molnar"><a href="https://christophm.github.io/interpretable-ml-book/index.html">Interpretable Machine Learning</a> | Christoph Molnar</h3>
<p>In this book, Christoph Molnar details the importance of explainable machine learning, techniques data teams can adopt today, and the future of machine learning interpretability.</p>
<p> <br><center><a href="https://www.datacamp.com/groups/business"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1612889508/From_DataCamp_c6kasv.png" /></a></p>
<h3 id="-white-paper-our-guide-to-open-source-in-data-science-https-www-datacamp-com-resources-whitepapers-our-guide-to-open-source-in-data-science-"><a href="https://www.datacamp.com/resources/whitepapers/our-guide-to-open-source-in-data-science">White Paper: Our guide to open source in data science</a></h3>
<p>Just as it did for software development, open-source will empower organizations to make the most of their data and create transformative solutions, processes, and products with machine learning and data science. Find out all you need to know about open-source in data science in our exclusive new guide.</p>
<h3 id="-webinar-path-to-data-fluency-https-www-datacamp-com-resources-webinars-path-to-data-fluency-"><a href="https://www.datacamp.com/resources/webinars/path-to-data-fluency">Webinar: Path to data fluency</a></h3>
<p>Data-driven transformation is a long process requiring organizations to go through different data maturity stages. In this webinar, we outline what these data maturity stages are and what steps you can take to grow from one data maturity stage to another.</p>


https://www.datacamp.com/community/blog/digest-march-2021

Our favorite articles of the month include this one question that will make your data project 10x more valuable.

DataCamp Digest March 2021: The traits of future-proof organizations | DataCamp

DataCamp Digest March 2021: The traits of future-proof organizations 

<h2>Introduction to Instrumental Variables</h2><p>This chapter will introduce you to Instrumental Variables (IV) analysis, a tricky but powerful method to find causality through indirect inference</p><h2>Instrumental Variables in Practice</h2><p>This chapter covers more details about Instrumental Variables Analysis, and allows you to practice instrumental variables for yourself</p><h2>Regression Discontinuity Design</h2><p>This chapter introduces you to Regression Discontinuity Design (RDD), which is often considered a more flexible relative of Instrumental Variables</p><h2>The Local Average Treatment Effect of IV</h2><p>This chapter explains the treatment effect we are estimating with the instrumental variables method - LATE, or the Local Average Treatment Effect</p>

https://www.datacamp.com/community/open-courses/causal-inference-with-r-instrumental-variables-rdd

A look at using Instrumental Variables and Regression Discontinuity to estimate causal effects

Causal Inference with R - Instrumental Variables & RDD

<h2>Regressions 1: Introduction to Regression As Causality</h2><p>This chapter will introduce you to using regression analysis to find causal effects</p><h2>Regressions 2: Using Regression to Estimate Causal Effects</h2><p>This chapter will introduce you to using regression analysis to find causal effects</p><h2>Regressions 3: Introduction to Matching Methods</h2><p>This chapter will introduce you to using matching methods to find causal effects</p>

https://www.datacamp.com/community/open-courses/causal-inference-with-r-regression

Welcome to the 3rd course in our series on causal inference concepts and methods created by Duke University with support from eBay, Inc. 

Causal Inference with R - Regression

<p>Plot character or <em>pch</em> is the standard argument to set the character that will be plotted in a number of R functions.</p>
<p>Explanatory text can be added to a plot in several different forms, including axis labels, titles, legends, or a text added to the plot itself. Base graphics functions in R typically create axis labels by default, although these can be overridden through the argument <code>xlab()</code> that allows us to provide our own x-axis label and <code>ylab()</code> that allows us to provide our own y-axis label.</p>
<p>Some base graphics functions also provide default titles, but again, these can be overridden has we see here:</p>
<pre><code class="lang-python">library(MASS)
par(mfrow = c(1, 2))
plot(density(geyser$waiting))
plot(density(geyser$waiting), main = &quot;Estimated density: \n Old Faithful waiting times&quot;)
</code></pre>
<center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1603481962/pch_2_kukmll.png" /></center>


<p>The plot on the left uses the default title returned by the density function, which tells us that the plot was generated by this function using its default options and also gives the R specification for the variable whose density we are plotting. In the right-hand plot, this default title has been overridden by specifying the optional argument main. Note that by including the return character, backslash n, we are creating a two-line title in this character string.</p>
<h2 id="-text-function"><code>text()</code> Function</h2>
<p>Like the lines and points functions, text is a low-level graphics function that allows us to add explanatory text to the existing plot. To do this, we must specify the values for x and y, the coordinates on the plot where the text should appear, and labels, a character vector that specifies the text to be added.</p>
<pre><code class="lang-python">text(x, y, adj)
</code></pre>
<center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/c_scale,f_auto,q_auto:best,w_575/v1603481260/pch3_ubupgd.png" /></center>

<p>By default, the text added to the plot is centered at the specified x values, but the optional argument <code>adj</code> can be used to modify the alignment.</p>
<p>It is noted that the <code>adj</code> argument to the <code>text()</code> function determines the horizontal placement of the text, and it can take any value between 0 (left-justified text) and 1 (right-justified text). In fact, this argument can take values outside this range. That is, making this value negative causes the text to start to the right of the specified <code>x</code> position. Similarly, making <code>adj</code> greater than 1 causes the text to end to the left of the <code>x</code> position.</p>
<pre><code class="lang-python">library(MASS)
plot(UScereal$fibre)
text(5, 28, &quot;&lt;-- Outliers [left-justified text at (5, 28)]&quot;, adj = 0)
text(65, 23, &quot;[Right-justified text at (65, 23)]&quot;, adj = 1, col = &quot;red&quot;)
text(5, 28, &quot;[Centered text (default) at (31, 18)]&quot;, col = &quot;blue&quot;)
</code></pre>
<center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best,w_575/v1603481260/pch5_decn8p.png" /></center>

<h3 id="fonts-orientations-and-other-text-features">Fonts, Orientations, and Other Text Features</h3>
<p>Below we see some of the ways the appearance of added text can be changed through optional arguments to the text function. By default, text is created horizontally across the page, but this can be changed with the <code>srt</code> argument, which specifies the angle of orientation with respect to the horizontal axis.</p>
<p>Next, the text in red angles upward as a result of specifying <code>srt</code> as the positive value 30 degrees, while the text in green angles downward by specifying <code>srt</code> equals -45.</p>
<p>We can also specify the color of the text with the <code>col</code> argument, and the red text shows that we can change the text size with the <code>cex</code> argument. By specifying <code>cex = 1.2</code>, we have specified the text to be 20% larger than normal.</p>
<p>The other text characteristic that has been modified in this example is the font, set by the <code>font</code> argument. The default value is <code>font = 1</code>, which specifies normal text, while <code>font = 2</code> specifies boldface, <code>font = 3</code> specifies italics, and <code>font = 4</code> specifies both boldface and italics.</p>
<pre><code class="lang-python">library(MASS)
plot(Boston$rad)

# &quot;Inner city&quot; with adjusted colour and rotation
text(350, 24, adj = 1, &quot;Inner city? --&gt;&quot;, srt = 30, font = 2, cex = 1.2, col = &quot;red&quot;)

# &quot;Suburbs&quot; with adjusted colour and rotation
text(100, 15, &quot;Suburbs? --&gt;&quot;, srt = -45, font = 3, col = &quot;green&quot;)

title(&quot;Text with varying orientations, fonts, sizes &amp; colors&quot;)
</code></pre>
<center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1603481012/pch_1_qfa8ff.png" /></center>

<h2 id="adjusting-text-position-size-and-font">Adjusting Text Position, Size, and Font</h2>
<p>In the following example, you will:</p>
<ul>
<li>Create a plot of <code>MPG.city</code> vs. <code>Horsepower</code> from the <code>Cars93</code> data frame, with data represented as open circles (default <code>pch</code>).</li>
<li>Construct the variable <code>index3</code> using the <code>which()</code> function that identifies the row numbers containing all 3-cylinder cars.</li>
<li>Use the <code>points()</code> function to overlay solid circles, <code>pch = 16</code>, on top of all points in the plot that represent 3-cylinder cars.</li>
<li>Use the <code>text()</code> function with the <code>Make</code> variable as before to add labels to the right of the 3-cylinder cars, but now use <code>adj = 0.2</code> to move the labels further to the right, use the <code>cex</code> argument to increase the label size by 20 percent, and use the <code>font</code> argument to make the labels bold italic.</li>
</ul>
<pre><code class="lang-python"># Plot MPG.city vs. Horsepower as open circles
plot(Cars93$Horsepower, Cars93$MPG.city)

# Create index3, pointing to 3-cylinder cars
index3 &lt;- which(Cars93$Cylinders == 3)

# Highlight 3-cylinder cars as solid circles
points(Cars93$Horsepower[index3],
       Cars93$MPG.city[index3],
       pch = 16)

# Add car names, offset from points, with larger bold text
text(Cars93$Horsepower[index3],
     Cars93$MPG.city[index3],
     Cars93$Make[index3],
     adj = -0.2, cex = 1.2, font = 4)
</code></pre>
<p>When we run the above code, it produces the following result:</p>
<center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1603481260/pch2_uumynj.png" /></center>

<p><a href="https://campus.datacamp.com/courses/data-visualization-in-r/adding-details-to-plots?ex=11">Try it for yourself</a>.</p>
<p>To learn more about adding text to plots, please see this video from our course <a href="https://campus.datacamp.com/courses/data-visualization-in-r/adding-details-to-plots?ex=9">Data Visualization in R</a>.</p>
<iframe width="560" height="315" src="https://www.youtube.com/embed/VY6X27P2uAQ" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>

<p>This content is taken from DataCamp’s <a href="https://www.datacamp.com/courses/introduction-to-data-visualization-with-ggplot2">Data Visualization in R</a> course by Ronald Pearson.</p>


https://www.datacamp.com/community/tutorials/plot-character-in-r

(Tutorial) PCH in R

PCH in R

<p>The <code>boxplot()</code> function shows how the distribution of a numerical variable <code>y</code> differs across the unique levels of a second variable, <code>x</code>. To be effective, this second variable should not have too many unique levels (e.g., 10 or fewer is good; many more than this makes the plot difficult to interpret).</p>
<p>The <code>boxplot()</code> function also has a number of optional parameters, and this exercise asks you to use three of them to obtain a more informative plot:</p>
<ul>
<li><code>varwidth</code> allows for variable-width Box Plot that shows the different sizes of the data subsets.</li>
<li><code>log</code> allows for log-transformed y-values.</li>
<li><code>las</code> allows for more readable axis labels.</li>
</ul>
<h2 id="when-you-should-use-a-box-plot-">When you should use a Box Plot?</h2>
<ul>
<li><p>When you have a continuous variable, split by a categorical variable.</p>
</li>
<li><p>When you want to compare the distributions of the continuous variable for each category.</p>
</li>
</ul>
<h2 id="histogram-vs-box-plot">Histogram vs. Box Plot</h2>
<ul>
<li>Below is the comparison of a Histogram vs. a Box Plot. The line in the middle shows the median of the distribution. That is, half the monarchs started ruling before this age, and half after this age.</li>
</ul>
<center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/c_scale,f_auto,q_auto:best/v1601058164/boxplot3_aej9he.png" /></center>


<ul>
<li>The box in the Box Plot extends from the lower quartile to the upper quartile. The lower quartile is the point where one-quarter of the values are below it. That is, one-quarter of the monarchs started ruling before this age, and three-quarters after it. Likewise, the upper quartile is the age where three quarters of the monarchs started ruling below this age. The difference between the upper quartile and the lower quartile is called the inter-quartile range.</li>
</ul>
<center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1601058164/boxplot4_sjwpbm.png" /></center>

<p>The horizontal lines, known as &quot;whiskers&quot;, have a more complicated definition. Each bar extends to one and a half times the interquartile range, but then they are limited to reaching actual data points.</p>
<p>The technical definition is shown in the image below, but in practice, you can think of the whiskers as extending far enough that anything outside of them is an extreme value.</p>
<center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1601058084/Screen_Shot_2020-09-25_at_2.20.50_PM_xizuqo.png" /></center>

<p>As mentioned before, the power of Box Plots is that you can compare many distributions at once. Here, the royal houses are ordered from oldest at the top to newest at the bottom.</p>
<center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best,w_400/v1601058084/Screen_Shot_2020-09-25_at_2.20.59_PM_ksobqy.png" /></center>

<p>A trend is visible: since the Plantagenets in the fourteenth century, the boxes gradually move right, showing that the ages when new monarchs ascend to the throne have been increasing.</p>
<p>Godwin and Blois appear as a single line because there was only one king from each house. The Anjou house only had three kings, and forms a box with one whisker, not two.</p>
<p>Notice that the Box Plots for the houses of Denmark and Windsor show some points. These are extreme values, that is, values that are outside the range of the whiskers. Windsor&#39;s left-most outlier is Elizabeth the second, who ascended at age 26.</p>
<h2 id="-boxplot-function"><code>boxplot()</code> Function</h2>
<p>In the following example, using the formula interface, you will create a Box Plot showing the distribution of numerical <code>crim</code> values over the different distinct <code>rad</code> values from the <a href="https://www.rdocumentation.org/packages/MASS/topics/Boston">Boston</a> data frame. Then, use the <code>varwidth</code> parameter to obtain variable-width Box Plots, specify a log-transformed y-axis, and set the <code>las</code> parameter equal to 1 to obtain horizontal labels for both the x and y-axes.</p>
<p>Finally, use the <code>title()</code> function to add the title <code>&quot;Crime rate vs. radial highway index&quot;</code>.</p>
<pre><code class="lang-r"># Create a variable-width Box Plot with log y-axis &amp; horizontal labels
boxplot(crim ~ rad, data = Boston,
        varwidth = TRUE, log = &quot;y&quot;, las = 1)

# Add a title
title(&quot;Crime rate vs. radial highway index&quot;)
</code></pre>
<p>When we run the above code, it produces the following result:</p>
<center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best,w_400/v1601058164/boxplot2_rjcbyw.png" /></center>

<p><a href="https://campus.datacamp.com/courses/data-visualization-in-r/different-plot-types?ex=7">Try it for yourself</a>.</p>
<p>To learn more about Box Plots, please see this video from our course <a href="https://campus.datacamp.com/courses/data-visualization-for-everyone/visualizing-distributions?ex=7">Data Visualization for Everyone</a>.</p>
<iframe width="560" height="315" src="https://www.youtube.com/embed/rs0P8wLLDEQ?start=90" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>

<p>This content is taken from DataCamp’s <a href="https://www.datacamp.com/courses/data-visualization-for-everyone">Data Visualization for Everyone</a> course by Richie Cotton and our <a href="https://learn.datacamp.com/courses/data-processing-in-shell">Data Visualization in R</a> course by Ronald Pearson.</p>


https://www.datacamp.com/community/tutorials/boxplot-in-r

community

datacamp

r programming

News (123)

The Credibility Crisis in Data Science (with Skipper Seabold)

Choosing R or Python for Data Analysis? An Infographic

The Credibility Crisis in Data Science (with Skipper Seabold)

Tutorials (115)

PCH in R

Box Plot in R

Cheat Sheets (4)

Tidyverse Cheat Sheet For Beginners

Open Courses (25)

Causal Inference with R - Instrumental Variables & RDD

Causal Inference with R - Regression

Podcast (20)

Critical Thinking in Data Science (with Debbie Berebichez)

Blog Posts (213)

Blazing the Trail in Developing AI Literacy Across Singapore

Tech Thoughts (0)