Depending on the nature of the quantitative data you are investigating, different trend lines may be more appropriate at representing patterns. The platform allows you to choose the method in which trend lines are calculated.

## This article covers:

## To Access Trend Line Options

**1.** Navigate to a column chart such as the **Employee Satisfaction** analytics dashboard.

**2.** Select the **Trend Line Graph** button to turn the trend line off.

**3.** Select the **Chart Settings** button to customise the type of trend line used on the graph.

**4.** You can select a** Moving Average, Linear Regression** or **Polynomial Regression.**

**5.** You can also choose the period the trend line is calculated on (such as every three months).

### Types of Trend Lines

**Moving Average: **is used to show a trend by averaging sets of data points together. It works by creating a rolling average across the specified number of data points to form each point of the trendline. For example, if the period is set to 2then the line will be drawn as follows:

- The first point in the trendline is always drawn to match the first data point as there are no previous data points.
- The second trendline point is an average of the first two data points.
- The third trendline point is an average of the second and third data points.
- The fourth trendline point is an average of the third and fourth data points, and so on.

**Linear Regression:** is a straight line drawn to fit your data set. It will help identify trends in the data set where the data points are increasing or decreasing over time consistently.

**Polynomial Regression:** is useful when the data set has a curved trend rather than a straight line. This can be helpful for drawing a trendline over a cycle of highs and lows.

The order determines how many peaks to draw through the data set. An order of two indicates a single peak in the trendline. An order of 3 can draw one or two peaks and troughs. An order of 4 can draw up to three peaks and troughs.

## Comments

0 comments

Please sign in to leave a comment.