To the left is a very poor, but not unique, example of a graph. This graph is (intentionally) riddled with problems that are commonly seen in students' work.
Find out how to avoid these costly mistakes below.
Graphs - learn to love them!
I hear this moan more than any other - it also is one of the most exasperating:
"Why do we have to draw graphs in Science - it's not math!"
However, it is important that you do not see subjects as cut off from each other. The skills learned in math, English, and social studies are readily and necessarily transferable to the Scientific arena.
Graphs are one of the most important tools available to scientists (young and old) to display data. They are:
- Easy to interpret
- Display a large amount of information in a small (ish) space
- Easy to draw!
Common Mistakes, and how to avoid them
When making a graph, ensure you have followed the 6 Graph Commandments
- Thou shalt draw your graph in with a ruler.
- Thou shalt use all thy graph paper.
- Thou shalt label your axes.
- Thou shalt always give units.
- Thou shalt not draw bar graphs when showing trends in quantitative data.
- Thou shalt not play dot-to-dot with thy data points!
Taking a closer look at the graph above - this person has broken five of the six commandments. This makes it almost completely useless.
If you have ticked off each of the Commandments, you are halfway to achieving a good graph. But now it is time to make the graph specific to your project.
- Give your graph a specific, descriptive title. (e.g., The Effect of Temperature on Plant Growth After Fertilizer Application)
- Ensure you have put your graph the right way around. Your x-axis should always show the independent variable - this is the variable you are changing. Your y-axis should always plot the dependent variable - this is the variable you are measuring. For example, when looking at the effect of temperature on rate of reaction, you change the temperature and measure the rate. As such, temperature goes on your x-axis (it is independent) and rate goes on your y-axis (it is dependent)
- Ensure you plot your data carefully. Mark your data point with a small point. If you are plotting multiple data sets on one graph, then use a small o or x or similar to distinguish between data sets.
- If plotting multiple data sets, WRITE OUT A KEY/LEGEND to indicate what data set each line on the graph is referring to.
- Do not play dot-to-dot. I know I have stated this before, but I so often see jagged connected lines on graphs from my students. Only very rarely are data points connected in this way. More often, we are seeking the trend or pattern that our results show, for that we need...
- DRAW A LINE OF BEST FIT. These lines pass through or near as many data points as possible. They can either be straight lined, or a smooth curve. Look for the pattern to decide which is most appropriate.
The graphs above are near-perfect graphs! Following all of the commandments and many of the higher tier specs too. Beautiful.