
# 22.8: J1.08: Section 3


## Section 3: Choosing a trial model based on the pattern of the data graph

In most cases, it will not be difficult to decide which type of model will best fit the trendline of the data. Each type of model has its own characteristics, which usually means that only one of them, if any, will be capable of fitting the data well. If you identify a good candidate for a model, you can test it by using the appropriate worksheet to find its best-fit parameters and see how good the match is. If no model you know of fits the data, explain what aspect of the data pattern lets you know that model won’t work.

### Example 7—for each graph below, is a linear, exponential, quadratic, or some other model appropriate?

 [a] [b] [c] [d] [e] [f]