The following section is an excerpt from the example laboratory report, titled Proper Inflation of a Basketball. (The entire example report is available in either  Microsoft Word Format for printing, or  HTML format for online viewing.)

Discussion

These results were obtained at a fixed temperature. The ideal gas law (Equation (2)) tells us that as the temperature increases, the pressure in the basketball will go up. Conversely, as the temperature goes down, the pressure will decrease. We observed this firsthand: when we brought the ball in from the cold garage, it was noticeably flat. This is the largest source of uncertainty in this experiment. Our results are not valid for different temperatures. One way to address this problem would be to account for it in our analysis, using the ideal gas law. This would make our result more complicated mathematically, but more generally useful. Another method would be to take data under at different temperature conditions (such as outside during hot and cold weather).

We neglected the effect of the material of the ball on its rebounding when doing this experiment. The material of the ball has some elasticity to it, and a leather basketball would bounce slightly differently than a rubber one. Our results show that this is a more important effect when there is very little air in the ball (see Figure 3). Even a little air dramatically changes the ball’s rebounding, as shown by the initially large slope in the best-fit line. The experiment could be improved by repeating it with other basketballs of different materials to verify this assertion.

Note what's here:

• Interpretation of the results. Are the results valid, and if so, over what limits?
• The largest source of error. Identify the largest source of error or uncertainty in the expreriment, explain why it is the largest uncertainty, and describe what you could do to improve the expreiment.

• A laundry list of errors. A shotgun approach ("It could have been this, it could have been that, etc.") doesn't demonstrate that you understand the expreiment.
• Unsupported opinions. If you can't support what you say with your data, then don't say it.