|
Once the points are plotted on a graph a decision must be made about whether to join them up with a straight line, a curve or with separate straight lines. |
|
Straight line graphs
Plotted points will generally give a good indication of whether or not they can be joined by a straight line. It is quite probable that any line will not pass through all of the points. In this case a line of best fit must be drawn
This means a line that can be drawn smoothly, going through as many of the data points as possible.
Curved graph lines
If the plotted points appear to form a curve then they can be joined by a smoothly sketched curve. Once again it is quite likely that not all of the points lie on the curve. This is normal experimental deviation and can often be resolved by using error bars (see below).
Error bars
These are bars drawn on the graph showing both the maxmum and minimum values that a data point could have. It is perfectly feasible to have data bars in both the x direction and in the y direction to compensate for experimental error in both measured variables.
|
