# Lies, Damned Lies and Statistics!

## Charts and graphs are used frequently in Microsoft products such as Excel, PowerPoint, Project andPower BI but all may not be as it first seems!

According to Wikipedia “Lies, damned lies, and statistics” is a phrase describing the persuasive power of statistics to bolster weak arguments. It is also sometimes used to doubt statistics used to prove an opponent’s point.

Graphs and charts are powerful tools for presenting data in a clear and concise manner. However, they can also be manipulated to mislead or deceive the viewer. Here are six common ways this can happen so beware, all that you see may not be as true as you think!!

### 1. Truncated Y-Axis

By starting the y-axis at a value other than zero, the differences between data points can be exaggerated. This can make small differences appear significant, misleading the viewer about the true scale of the data.

#### Example

A bar graph is used to compare the sales of two products, Product A and Product B. The y-axis starts at 90 instead of 0. Product A sold 100 units and Product B sold 95 units. Because the y-axis starts at 90, the bar for Product A appears to be twice as long as the bar for Product B, giving the impression that Product A sold many more units than Product B when the figures are in fact much closer.

### 2. Inconsistent Scales

Using different scales for similar graphs can distort comparisons. This can lead to incorrect conclusions about the relative size or growth of the data points.

#### Example

A company presents two line graphs showing the growth of their user base over time. The first graph shows the number of users in the thousands and the second graph shows the number of users in the millions. If viewed quickly or without careful attention to the scales, it could appear that the company’s user base grew exponentially when in reality the scale of the second graph was just larger.

### 3. Selective Data

Presenting only a selective range of data can skew the viewer’s perception. This can create a false understanding due to omitting relevant data that would provide a more accurate picture.

#### Example

A line graph is used to show the average temperature in a city over the past 10 years. However, the graph only includes data from the winter months. This could lead viewers to believe that the average temperature in the city is much lower than it actually is when considering the entire year.

Labels on a graph can be misleading if they are vague, incorrect, or intentionally deceptive. This can confuse the viewer or lead them to draw incorrect conclusions from the data.

#### Example

Imagine a bar graph showing the sales of a product over time. If the labels for the time axis are vague (e.g., “Before” and “After” instead of specific years or months), it can be unclear what period the data covers.

### 5. Cherry-Picking Data

This involves only selecting data that supports a particular conclusion while ignoring data that contradicts it. This can create a biased view of the data and mislead the viewer into thinking the data supports one conclusion when it may not be the case.

#### Example

A company might create a line graph showing their stock price over the past year, but only include the months where the stock price was increasing. This would give the impression that their stock is a good investment, even though it might have decreased significantly in value during the months that were omitted.

### 6. Distorting Dimensions in 3D Graphs

In 3D graphs, you can distort dimensions to make certain parts of the graph appear larger or smaller than they actually are. This can exaggerate or minimize differences in the data, leading to false perceptions about the data’s significance.

#### Example

A 3D pie chart might be used to represent market share of different companies in an industry. If the pie chart is tilted, the sections at the front can appear larger due to perspective, making those companies seem to have a larger market share than they actually do.

### Conclusion

While graphs are essential for data visualization and cannot lie as such the way the data is presented can certainly bend the truth.  This is used regularly in the media to make people believe something that is not actually true.