Eight misunderstandings that should be avoided in data visualization

Data is something we often come into contact with , however , Can we visualize these data in the right way ？ We may have to create a lot of visual charts every day , Some will be appreciated by the audience , And some were rejected , Why is that? ？ ok , The answer is to create , Let's try to find out the causes and problems , See how to deal with them . ad locum , I'll summarize some of the best and worst versions of the chart , If you are using the worst version , Stop and correct as soon as possible .

The baseline is 0 Start chart for

One of the most common mistakes I find when drawing a chart is not to 0 Baseline to make charts .

We are Data visualization Horizontal or vertical bar charts are usually used in , Sometimes when we use a simple bar chart for comparison , It does send a message , But vertical or horizontal stacked bars are better . Let's take an example , We have monthly sales figures for women and men , We can use the chart below for comparison , Do not plot positive and negative values on the same side of the chart , It's hard to read the information of the chart .

Multiple colors on the chart

Many people use multiple colors when using charts , This makes the graphics look very strange , Normally , If not to highlight a data , Just use one color for a chart .

If you have 2 More than one chart , Then it would be a good idea to separate the chart by color , Let's look at the diagram below .

Emphasize data

Sometimes emphasizing the data also makes the chart look good . In the chart below , Only 4 A variable , So this little change in the chart won't be very obvious , But when we use 100 Variables , Highlighting the highest value of the column chart will be of great help .

Confusing color choices

typeface 、 Color 、 Axis and other functions are very important when drawing charts , Choosing the color of the chart is a very critical step , Because if there are very bright or very light colors on the chart , Then reading the chart will become very difficult . In this example , We have the following 2 A different chart , Show the highest to lowest sales in this area. The expression in color is different , Let's look at the chart on the left , It's basically hard to tell which is high and which is low . And in the chart on the right , There are blue and pink scales , You can easily distinguish shadows .

Avoid randomness on the chart

Always arrange in ascending or descending order according to the value of the bar chart , For horizontal bars , Put the maximum at the top , For vertical bars , Put the maximum value on the left , This will help the audience find the highest and lowest values from the chart .

Tell a story or answer at least one question

Most initial data visualization tools generate only a single chart , Such as histogram or bar chart , But most of the time, you need to combine multiple charts , Let's see how it is done in intellectual analysis . You can find this case in intelligence analysis . ad locum , I made a for analyzing customers RFM Customer value classification model chart , You can make different charts , Or a bar chart, such as a pie chart , But I put them all together , To summarize and analyze . ad locum , We can see clearly that , Generally, the number of customers retained is the largest , What's the percentage , What are the specific details of the customer name .

Add text as needed

If you think adding other text will help readers better understand the chart , Then just add text . Let's look at a real example , You can find this chart on intelligence analysis , ad locum , I added some text , for example A This category of products represents the cumulative proportion of sales 80% Products within .

Use pie chart

I see a lot of people using pie charts in the wrong way , When using pie charts, remember , The value in the pie chart must not exceed 5 individual . In the following example , We see in the Netflix Proportion of programs watched on , We can see clearly that , Movies are more popular here .

From the above case , We see some common mistakes in the chart , And learned how to overcome these mistakes . Data visualization is an art form , It needs to be mastered over time . Although the tips and techniques of data visualization are not comprehensive , But it will certainly help you get on the right track . Understanding the audience's perspective is the key to creating successful and effective visual effects , No matter what tool you use to create elegant and beautiful charts , It is important to understand the nature behind visual effects .