Recently, I often talk about visualization with my friends , Found that many novices often do not choose the right chart , As a result, the data analysis report is not satisfactory , Today I'd like to share some tips for chart selection

To make the visual chart achieve the best information transmission effect to the user , We have to seriously consider the various elements of planning and Design

And there are many kinds of charts , How to choose the right chart to achieve “ A picture is worth a thousand words ” The effect of ？

I divide the charts into several broad categories , Respectively 「 Comparative 、 Proportion class 、 Trend class 、 Related classes 、 Geography 」, You can choose the right chart for your own purposes .

# One 、 Contrast class

**1、 Ordinary column chart **

brief introduction ： Ordinary column chart Use vertical columns to show numerical comparisons between categories , One axis of the histogram shows the category being compared , The other scale represents the other axis

characteristic ： Not suitable for more than 10 Compare data from different categories , If the category label is too long, bar chart is recommended

**2、 Compare the histogram **

brief introduction ： Compare the histogram Use forward and reverse columns to show numerical comparisons between categories . One axis of the chart shows the categories being compared , And the other axis represents the corresponding scale value .

characteristic ： Used to show comparisons of data with opposite meanings , If it's not the opposite, it's suggested to use grouped column chart .

**3、 Group column **

brief introduction ： Grouping histogram is often used under the same grouping , Comparison of different types of data . Use the height of the column to show the numerical comparison , Use color to distinguish different kinds of data .

characteristic ： In the same group , You can't have too many categories of data .

**4、 Stacked column **

brief introduction ： Stacked column The total number of groups can be compared , You can also view the size and proportion of each small category contained in each group , Therefore, it is very suitable for dealing with the relationship between the part and the whole .

characteristic ： Suitable for total display size , But it is not suitable to compare the same category under different groups .

**5、 Broken line map of the area **

brief introduction ： Broken line map of the area Can separate multiple indicators , Reflect the tendency of things to change with time or order

characteristic ： Good for comparing trends , Avoid crossing multiple line charts .

**6、 Radar map **

brief introduction ： Radar map It's also called spider web chart , Each of its variables has an axis emitted from the center , All the angles between the axes are equal , At the same time, each axis has the same scale .

characteristic ： Too many radar chart variables will reduce the readability of the chart , Great for showing performance data .

**7、 The word cloud **

brief introduction ： The word cloud It is an important way of text big data visualization , It is often used to highlight high-frequency sentences and words in a large number of texts , Quickly perceive the most prominent words . Commonly used in the statistics of high frequency search fields of websites .

characteristic ： Not suitable for text data with large amount of data , It is not suitable for data processing with little data differentiation .

**8、 Aggregate bubble chart **

brief introduction ： Aggregate bubble chart in , Dimensions define bubbles , Measurement defines the size of the bubble 、 Color .

characteristic ： It's not suitable for data with small discrimination .

**9、 Nightingale rose **

brief introduction ： Nightingale rose The function of a graph is similar to that of a column graph , It is mainly used to compare , The numerical size is mapped to the radius of the rosette .

characteristic ： When the data are similar , Not for pie charts , It's a nightingale rose .

# Two 、 Proportion class

1、 The pie chart

brief introduction ： The pie chart It's usually by color , The magnitude of the comparison , And it can show the proportion relationship between each category and the whole .

characteristic ： The number of categories cannot be too many , And it is not suitable for data with small discrimination .

**2、 Rectangular block **

brief introduction ： Rectangular block Suitable for displaying hierarchical data , Can directly reflect the comparison between the same level . The parent node embeds the child node , Each node is divided into rectangles of different sizes , Use the size of the area to show the attributes of the node .

characteristic ： Very suitable for weighted tree data , Compare the size of each category and its proportion to the whole .

**3、 Percentage stacked column **

brief introduction ： Percentage stacked column Compare the proportion of different categories in the same group of data .

characteristic ： The number of different categories in the same group should not be too many .

**4、 Multi layer pie chart **

brief introduction ： Multi layer pie chart It means having multiple levels , And there are pie charts with inclusion relationships between levels . Multi layer pie chart is suitable for displaying complex tree structure data with parent-child relationship , Such as geographic area data 、 The upper and lower levels of the company 、 Time level of quarter and month, etc .

characteristic ： You can't have too many levels and categories , Too many slices cause too little interference in reading

**5、 The dashboard **

brief introduction ： The dashboard Set the target , And then it's used to show speed 、 temperature 、 speed of progress 、 Completion rate 、 Satisfaction, etc , In many cases, it is also used to express the proportion .

characteristic ： Only suitable for data display of single indicator .

# 3、 ... and 、 Trend class

**1、 Broken line diagram **

brief introduction ： Broken line diagram It's very convenient to reflect the trend of things over time or other ordered categories .1） It can analyze the interaction and interaction of multiple sets of data over time , So we can sum up and get some conclusions and experiences .2） It can compare the size of multiple groups of data at the same time .

characteristic ： The number of broken lines should not be too much , Can lead to poor readability of the chart .

**2、 Area map **

brief introduction ： Area map To show persistent data , It's a good indicator of trends 、 The cumulative 、 Decrease and change .

characteristic ： Show the trend of the difference between two continuous variables .

**3、 Ordinary area map **

brief introduction ： The common area chart is evolved from the broken line chart , It's also convenient to reflect the trend of things changing with time or other orderly categories . Because of the area filling , So it's better than a line chart to show trends .

characteristic ： It's better not to have more than five area lines .

**4、 Scatter plot **

brief introduction ： Scatter plot You can display the shape of the data cluster , Distribution of analysis data . By observing the distribution of scatterers , We can infer the correlation of variables , stay FineBI Can be done by data fitting .

characteristic ： Scatter plot when there is more data , In order to better reflect the data distribution .

**5、 Waterfall Plot **

brief introduction ： Waterfall Plot Displays the cumulative total when adding or subtracting values , It is usually used to analyze a series of positive and negative values against initial values （ for example , Net income ） Influence .

characteristic ： Through the hanging column , It can show the increase and decrease of data more intuitively .

# Four 、 Distribution of the class

**1、 Scatter plot **

brief introduction ： Scatter plot You can display the shape of the data cluster , Distribution of analysis data . By observing the distribution of scatterers , We can infer the correlation of variables .

characteristic ： Scatter plot when there is more data , In order to better reflect the data distribution .

**2、 Thermal zone map **

brief introduction ： Thermal zone map Display the weight of each point in the coordinate range in a special highlight way .

characteristic ： The effect softens , Not suitable for accurate data representation , It's mainly used to look at distribution .

** 5、 ... and 、 Other categories **

**1、 Map **

brief introduction ： The map component even reflects the data in a geographic location ,FineBI Provides a variety of map components , Include Heat map 、 Regional map 、 Flow map 、 Point map, etc .

characteristic ： Very intuitive observation of the data relationship between different regions .

**2、 Funnel diagram **

brief introduction ： Funnel diagram It's also called an inverted triangle , Funnel chart from top to bottom , There is a logical order , Often used in process analysis , For example, the analysis of which link loss rate is abnormal .

characteristic ： There must be a logical order between the top and the bottom , If there is no logical relationship, it is recommended to use column chart for comparison .

版权声明

本文为[The sail is soft]所创，转载请带上原文链接，感谢

https://cdmana.com/2020/12/20201224120550848i.html