Let’s understand why do we need a visual representation of our data. A human brain is not meant to process this amount of unstructured, raw data and transform it into something useful and meaningful. Graphs and charts help us convey data results so that we can spot patterns and trends, gain understanding, and swiftly arrive at smarter conclusions.
There are numerous ways to put information together such that the data can be seen. A variety of different graphs and tables may be used to construct an intuitive dashboard, depending on the data being modelled and its intended use. While some visualisations are made automatically, some are created manually. In either case, there are numerous sorts to satisfy your needs for visuals.
Determining which type of graph to use can be difficult. If you have not read our blog post on Demystifying Data Analysis for Beginners: Mastering Numerical Data, Categorical Data and More! Then we strongly recommend to read it first. It will help you build strong foundation in Understanding the type of data you are working with is crucial for choosing appropriate data analysis techniques and visualizations.
There are generally three categories of visualizations: volumetrics, illustrative, and statistical graphs. However, it’s worth noting that there can be some overlap between these categories, and some visualizations may not fit neatly into any of these categories. Let’s first go over the three different categories of visualisations: Volumetrics, Illustrative and statistical graphs.
Volumes are visualised. Using a volumetric visualisation, you can see the truth. A quick look at the trends, some key figures, and perhaps some target analysis.
Examples of volumetrics visualizations include:
- Line Chart: A line chart is useful for visualizing trends over time. For example, a line chart can show how the number of visitors to a website has changed over the course of a year.
- Area Chart: An area chart is similar to a line chart but fills in the area beneath the line. This can be useful for showing the magnitude of changes over time. For example, an area chart can show how the revenue of a company has changed over the course of a year.
Illustrative graphs are useful for working with unstructured data, such as maps and word clouds.
Examples of illustrative graphs include:
- Choropleth or Geo Map: A choropleth map uses different shades or colors to represent different values of a variable, such as population density or income levels, within a predefined geographic region. The regions are often political boundaries like states or countries, and the intensity of the color represents the magnitude of the variable being measured.
- Geo Map: A geo map displays data points on a geographic map. Each data point is represented by a marker, such as a dot or pin, on the map. The markers can be color-coded or sized based on a variable being measured, such as the number of customers in a particular region or the amount of sales revenue generated.
- Word Cloud: A word cloud is a visual representation of text data where the most frequently occurring words are emphasized. For example, a word cloud can show the most commonly used words in a customer survey.
Choropleth maps and geo maps are similar in that they both represent data geographically. However, they differ in the way they display data.
A choropleth map uses different shades or colors to represent different values of a variable, such as population density or income levels, within a predefined geographic region. The regions are often political boundaries like states or countries, and the intensity of the color represents the magnitude of the variable being measured.
A geo map, on the other hand, displays data points on a geographic map. Each data point is represented by a marker, such as a dot or pin, on the map. The markers can be color-coded or sized based on a variable being measured, such as the number of customers in a particular region or the amount of sales revenue generated.
Therefore, although both types of maps are used to visualize geographic data, they differ in the way they represent and display that data.

Statistical Graphics: There are many conventional statistical graphics that are excellent for giving quantitative data life.
Statistical graphics are used to visualize and summarize data, making it easier to understand patterns, trends, and relationships. In this article, we will explore the different types of statistical graphics and how they can help data analysts and scientists gain insight. Choosing the correct graph allows us to view data in new ways and gain more insight than a noisy table of totals in a conventional report.
Comparison Charts – Comparison charts are used to compare a single or multiple datasets. They can compare things or show how things have changed over time. Comparison charts are useful for identifying patterns and trends in the data.Types of Comparison Graphs.
- Bar Chart: A bar chart is a chart with rectangular bars with lengths proportional to the values represented. Bar charts are used to compare data across categories or show changes over time. For example, a bar chart can be used to compare sales figures for different products.
- Column Chart: A column chart is a chart with vertical bars with lengths proportional to the values represented. Column charts are used to compare data across categories or show changes over time. For example, a column chart can be used to compare the number of customers per day.
- Line Chart: A line chart is a chart that uses lines to connect data points. Line charts are used to show trends over time or to compare data sets. For example, a line chart can be used to show the trend in the stock market over time.
Column charts and bar charts are both used to compare different categories or groups of data, but they differ in the orientation of the bars or columns. In a column chart, the columns are arranged vertically, while in a bar chart, the bars are arranged horizontally.
Column charts are best used when the categories being compared are few in number and have long category names or labels. The columns in a column chart are thin and tall, making it easier to display long category labels along the horizontal axis.
Bar charts are better used when the categories being compared are many and have short labels or names. The bars in a bar chart are short and wide, making it easier to display many categories along the vertical axis.
Both types of charts are useful in different situations, and the choice between them depends on the specific data and the purpose of the visualization.
Here’s an example: Let’s say you want to compare the sales performance of different products in a store. If you have a small number of products with long names, a column chart may be more appropriate. But if you have many products with short names, a bar chart may be a better choice.
“Understanding the Differences between Histograms and Bar Charts for Effective Data Visualization”
Column chart and histogram chart are different types of charts used for data visualization.
A column chart is used to represent categorical data, where each column represents a category and the height of the column represents the value. It is suitable for showing discrete data with a few categories.
On the other hand, a histogram chart is used to represent continuous data that is grouped into intervals or bins. It is suitable for showing the distribution of a large dataset, as it shows the frequency of data falling within a range of values.
While both charts use vertical bars to represent data, the key difference lies in the type of data they are best suited for representing. A column chart is used for categorical data with a few categories, while a histogram is used for continuous data with many data points.
Here’s an example to illustrate the difference between the two:
Suppose we have data on the ages of students in a class. The data ranges from 15 to 20 years, and we want to represent the distribution of ages using a chart.
If we use a column chart, we would have a separate column for each age, with the height of each column representing the number of students of that age. This may not be the most effective way to represent the data, as it would result in too many columns.
On the other hand, if we use a histogram chart, we would group the ages into intervals or bins, such as 15-16, 16-17, 17-18, 18-19, and 19-20. We would then represent the frequency of ages falling within each bin using a vertical bar. This would provide a more effective representation of the distribution of ages in the class.


Table with Embedded Charts: A table with embedded charts is a combination of a table and charts. It is used to show data in a more structured format. For example, a table with embedded charts can be used to show sales figures for different products over time.
Relationship Charts– Relationship charts are used to demonstrate the relationship or correlation between two or more variables. These charts are useful for identifying patterns and trends in the data.
- Box Plot: A box plot is a chart that shows the distribution of data. It is used to identify outliers and patterns. For example, a box plot can be used to show the distribution of salaries in a company.
- Candlestick Chart: A candlestick chart is a financial chart used to represent the price movements of an asset, such as a stock, over time. It displays four pieces of information: opening price, closing price, high price, and low price.
- Scatter Plot: A scatter plot is a chart that uses dots to represent data points. It is used to identify the relationship or correlation between two variables. For example, a scatter plot can be used to show the relationship between age and income.
- Bubble Chart: A bubble chart is a chart that uses bubbles to represent data points. It is used to identify the relationship or correlation when there are more than two variables. For example, a bubble chart can be used to show the relationship between age, income, and education level.
Box plot and candlestick chart are similar in some ways, but they are not the same. Both are used to display numerical data and provide information on the distribution of the data.
A box plot, also known as a box and whisker plot, displays the distribution of data through five statistics: minimum value, maximum value, median, first quartile (Q1), and third quartile (Q3). It is useful for identifying outliers and comparing the distribution of multiple datasets. A box plot consists of a box with lines extending from the top and bottom, representing the interquartile range (IQR), and whiskers that extend from the box to the minimum and maximum values.
A candlestick chart is a financial chart used to represent the price movements of an asset, such as a stock, over time. It displays four pieces of information: opening price, closing price, high price, and low price. The candlestick chart is named after the candle-shaped elements that represent each trading day. The upper and lower shadows of the candle represent the high and low prices, while the body of the candle represents the opening and closing prices.
In summary, while both box plot and candlestick chart are used to display numerical data, they have different purposes and are used in different contexts.

Candlestick Chart for the Sales figure

Distribution Charts – Distribution charts are used to display how variables are distributed across time. These charts are useful for identifying outliers, patterns and trends in the data.
- Single Variable – Column Histogram Chart: A column histogram chart is used to show the distribution of a single variable with a few data points. For example, a column histogram chart can be used to show the distribution of test scores in a class.
- Single Variable – Line Histogram Chart: A line histogram chart is used to show the distribution of a single variable with many data points. For example, a line histogram chart can be used to show the distribution of annual income in a city.
- Two Variables – Dual Axis Chart: A dual axis chart displays two different sets of data on the same graph, each with its y-axis. This allows for the comparison of two different variables that may have different scales or units of measurement. For example, a dual axis chart can be used to compare the number of visitors to a website and the revenue generated.
- Three Variables – 3D Area Chart: A 3D area chart is used to show the distribution of three variables within the data

Composition charts are used to show how components of a whole change over time. These charts are useful for illustrating how parts of a whole are changing or shifting in relation to one another. There are two types of composition charts: static and dynamic.
Static composition charts, such as waterfall charts and pie charts, provide a snapshot of how components of a whole are distributed at a specific point in time. Waterfall charts are useful for visualizing changes in financial data, while pie charts are commonly used to show how different categories contribute to a total.
For example, a company may use a pie chart to show how their revenue is divided among different product lines. The chart would display each product line as a slice of the pie, with the size of each slice indicating its contribution to the company’s overall revenue. This would provide a snapshot of how the company’s revenue is composed at a specific point in time.
Dynamic composition charts, such as stacked column charts, allow us to see how the components of a whole are changing over time. Stacked column charts are useful for illustrating changes in the composition of a whole, such as changes in the number or proportion of different categories over time.
For example, a healthcare provider may use a stacked column chart to show how the number of patients in different age groups has changed over time. The chart would display the number of patients in each age group as a separate column, with each column representing a specific point in time. The columns would be stacked on top of one another to show the total number of patients over time, while the size of each segment of the column would indicate the proportion of patients in each age group.
Overall, composition charts are useful for showing how different components of a whole are changing over time, and can provide valuable insights for data analysts and scientists.
In addition to the conventional statistical graphics mentioned above, there are also more advanced and complex visualizations that can be utilized for in-depth data analysis. These advanced visualizations can help data analysts and scientists identify patterns, trends, and anomalies that might otherwise go unnoticed.
Heatmaps – Heatmaps are often used to visualize data in a 2D matrix or table format. They are excellent for comparing data across multiple categories or dimensions. Heatmaps use color to represent the values of data points, with different colors indicating different ranges of values.
Chord Diagrams – Chord diagrams are used to visualize the relationships between different categories or dimensions in a data set. They are often used to show how different variables are interconnected, or how they influence each other.
Sankey Diagrams – Sankey diagrams are used to visualize the flow of data through a system or process. They are excellent for identifying bottlenecks, inefficiencies, or other areas where improvements can be made.
Treemaps – Treemaps are used to visualize hierarchical data structures. They are often used to show how different categories or dimensions are related to each other, or to identify areas of the data that might require further analysis.
All of these advanced visualizations can be very useful for data analysts and scientists, but they can also be quite complex to create and interpret. It is important to have a solid understanding of the underlying data and the relationships between different variables before attempting to create and analyze these visualizations.
So, choosing the correct visualization for your data is crucial for gaining insights and making informed decisions. Conventional statistical graphics like bar charts, line charts, and scatter plots are excellent for giving quantitative data life, while advanced visualizations like heatmaps, chord diagrams, and treemaps can help data analysts and scientists identify patterns and anomalies in their data. By utilizing a combination of these different visualization techniques, data analysts and scientists can gain a deeper understanding of their data and make more informed decisions.
- How to choose the right statistical chart for data analysis
- Best practices for creating effective statistical graphics
- Using statistical graphics to gain insights from data
- Types of statistical charts for data visualization
- How to interpret statistical charts for data analysis
- Statistical graphics for beginners: tips and tricks
- Advanced statistical graphics for data scientists
- Comparing and contrasting statistical chart types
- Tips for designing compelling statistical graphics
- Statistical graphics in the age of big data: challenges and opportunities
- Understanding Logarithmic Scales in Data Visualization: When and How to Use Them
Data Visualization FAQs: Questions on Chart Selection for Data Visualization
Question: Which chart to select when comparing total sales for different products?
Answer: Bar Chart
Question: Which chart to select when showing the distribution of ages in a population?
Answer: Histogram
Question: Which chart to select when comparing the market share of different companies in an industry? Answer: Pie Chart
Question: Which chart to select when showing the trend of a variable over time?
Answer: Line Chart
Question: Which chart to select when displaying the relationship between two variables?
Answer: Scatterplot
Question: Which chart to select when comparing the performance of different teams in a sports league? Answer: Stacked Bar Chart.
Question: Which chart to select when showing the composition of a budget or expenditure?
Answer: Stacked Bar Chart or Stacked Area Chart
Question: Which chart to select when displaying the frequency distribution of a categorical variable? Answer: Bar Chart or Pie Chart
Question: Which chart to select when comparing the performance of different candidates in an election? Answer: Grouped Bar Chart
Question: Which chart to select when showing the correlation between two variables?
Answer: Scatterplot or Bubble Chart
Question: Which chart to select when comparing the change in a variable over time for different categories? Answer: Multiple Line Chart or Area Chart
Question: Which chart to select when displaying the parts of a whole?
Answer: Pie Chart or Stacked Bar Chart
How to Use Axes, Series, and Other Key Elements for Effective Data Visualization and Statistical Analysis
When plotting a chart, two important concepts to consider are axes and series.
Axes: An axis is a line that represents a quantitative scale used to measure and display data. In a chart, there are two types of axes: X-axis and Y-axis. The X-axis, also known as the horizontal axis, represents the independent variable, while the Y-axis, also known as the vertical axis, represents the dependent variable. The values on each axis are often displayed in equally spaced intervals to help interpret and analyze the data more easily.
Series: A series is a collection of related data points that are plotted together in a chart. A series can consist of one or more data points and is typically displayed as a line, bar, or point on the chart. Each data point in a series represents a specific value for the dependent variable at a given point along the independent variable axis.
For example, consider a line chart showing the sales performance of a company over the last five years. In this chart, the X-axis would represent the years, while the Y-axis would represent the sales figures. The sales figures for each year would be plotted as a data point in the series. By analyzing the chart, we can easily see the trend in sales over time and make informed decisions based on the data.
In summary, understanding the concepts of axes and series is essential for effective data visualization and statistical analysis. By selecting the appropriate chart type and plotting the data accurately, we can gain valuable insights and make informed decisions.
There are several other important elements for statistical analysis and data visualization. Some of these elements include:
- Titles and Labels: Clear and descriptive titles and labels are essential for communicating the purpose and content of the visualization.
- Legends: Legends provide additional context to the visualization by identifying the different elements represented in the chart or graph.
- Annotations: Annotations can be used to highlight specific data points, provide additional information or context, or explain patterns or trends in the data.
- Gridlines: Gridlines help to visually align and separate data points, making it easier to read and interpret the visualization.
- Color Schemes: Color can be used to highlight different elements in the visualization or to indicate different categories or groups of data.
- Axes and Ticks: In addition to the x- and y-axes, additional axes and tick marks can be used to provide more detailed information about the data being visualized.
- Scale and Range: The scale and range of the axes can have a significant impact on the interpretation of the data. It is important to choose appropriate scales and ranges to accurately represent the data and highlight important trends or patterns.
Overall, there are many different elements that can be used to enhance the clarity and effectiveness of statistical analysis and data visualization. The specific elements used will depend on the type of data being analyzed and the goals of the visualization.
Understanding Logarithmic Scales in Data Visualization: When and How to Use Them
Linear scales and logarithmic scales are two ways to represent numerical data on a graph. A linear scale is a regular numeric scale where the distance between each tick mark is equal. On the other hand, a logarithmic scale is a scale where the distance between tick marks represents a power of ten.
To convert a linear scale to a logarithmic scale, you can use the following formula:
Logarithmic value = log(base 10) of linear value
For example, suppose you have a linear scale with values ranging from 1 to 1,000. To convert these values to a logarithmic scale, you can take the logarithm of each value using base 10:
Logarithmic value of 1 = log(base 10) of 1 = 0 Logarithmic value of 10 = log(base 10) of 10 = 1 Logarithmic value of 100 = log(base 10) of 100 = 2 Logarithmic value of 1,000 = log(base 10) of 1,000 = 3
When visualizing data, a logarithmic scale can be useful when the range of values is large and spans several orders of magnitude. For example, if you’re visualizing data on the size of planets in the solar system, the range of sizes could be from a few miles to several thousand miles. Using a linear scale may result in a graph where the smaller planets are difficult to see. However, by using a logarithmic scale, the smaller planets can be more clearly distinguished from the larger ones.
Another use case for logarithmic scales is when working with data that has exponential growth or decay. In these cases, a logarithmic scale can help visualize the trend more clearly.
A case study where converting to a logarithmic scale helped to visualize the data better:
Problem: A company wants to analyze the growth rate of their sales over the past year. However, the sales data is highly skewed towards a few large sales, making it difficult to see the changes in smaller sales.
Solution: By converting the sales data from a linear scale to a logarithmic scale, the company is able to better visualize the growth rate of their sales. With the logarithmic scale, the smaller sales are more visible, and the larger sales are still distinguishable due to the scale’s properties.
For instance, suppose the sales data for the past year was as follows:
Month | Sales |
---|---|
Jan | 100 |
Feb | 150 |
Mar | 200 |
Apr | 500 |
May | 1000 |
Jun | 1500 |
Jul | 3000 |
Aug | 5000 |
Sep | 7500 |
Oct | 10000 |
Nov | 15000 |
Dec | 20000 |
On a linear scale, it would be difficult to visualize the growth rate of sales, as the differences between the smaller sales would be almost indistinguishable compared to the larger ones. However, by converting the data to a logarithmic scale, the growth rate of the smaller sales becomes more apparent while still allowing the larger sales to be distinguished.
After converting to a logarithmic scale, the data would look like:
Month | Sales (Log Scale) |
---|---|
Jan | 2.00 |
Feb | 2.18 |
Mar | 2.30 |
Apr | 2.70 |
May | 3.00 |
Jun | 3.18 |
Jul | 3.48 |
Aug | 3.70 |
Sep | 3.88 |
Oct | 4.00 |
Nov | 4.18 |
Dec | 4.30 |
By visualizing the data on a logarithmic scale, the company can better see the growth rate of their sales and make informed decisions based on the insights gained from the data.
Handling Data Variability: Choosing Between Standard Deviation and Log Scale
Whether to use standard deviation or log scale depends on the type of data and the specific problem being analyzed.
If the data is skewed or has extreme outliers, using standard deviation may not accurately represent the variability of the data and may not be an appropriate measure. In such cases, a log transformation can be used to adjust the data to a more normal distribution and reduce the impact of outliers.
On the other hand, if the data is already normally distributed or does not have any extreme outliers, using standard deviation may be sufficient to describe the variability of the data.
It is important to carefully consider the nature of the data and the research question being addressed before deciding on which approach to use.
Tableau Data Blending: The Key to Combining Disparate Data Sources
Data blending is a feature in Tableau that enables users to integrate data from multiple sources into a single visualization. This technique enables analysis and display of information from various sources that cannot be combined into a single source.
When using data blending, one data source acts as the primary source, while another serves as a secondary source. Tableau identifies a shared field between both sources and merges the data using it. The primary source is responsible for creating the initial view, while the secondary source introduces extra dimensions and measurements to the visualization.
Data blending is ideal when using sources with different levels of detail or granularity. For instance, when there is sales data in one source, and demographic data is in another source, using data blending enables the analysis of sales data by customer demographics without merging the two sources.
Find Insightful Data Vizzes on My Tableau Public Profile
