Once you've completed your experiment and
recorded your results, make sure they are put together in an organized manner.
Keeping your records organized will not only impress the judges, but will also
give you a clear picture on what you've done. Once you have put your data
together, you will have to ascertain whether the results support your
hypothesis or not. If the results do not support your hypothesis, you should also
state the reasons for this. Think carefully why this is the case, using your
data to support your reasons.
Ensure that you spend
enough time analyzing the data and results of your experiment. Many scientists
find it extremely useful to use tables, charts and graphs to visually represent
the data collected from the experiment. One of the easiest ways to do this is
to use Microsoft Excel or Google Docs (Spreadsheet). Ensure that the unit of measurement is clearly
stated in your tables and graphs.
Where you have repeated
your experiment several times (as mentioned, you should typically perform at
least 3 trials of the same experiment), a table/chart can be very useful to
help you summarize and present your data in a meaningful manner. Sometimes, it
may be appropriate to provide a calculation of the average measurements.
For graphs, it's
probably best that the X-axis represent your independent variable whilst the
Y-axis represents the dependent variable. In our example, the independent
variable is the sunscreen protection factor whilst (grouped by brands, as we
performed this experiment with 5 brands of sunscreen lotion) the dependent
variable is the UV index (UV reading).
Take a look at the graph here http://www.all-science-fair-projects.com/project1260_130_3.html
As you can see, we used
a different color for each SPF level (but re-used that color for that
particular SPF level regardless of the brand). Different types of graphs are
appropriate for different experiments.
There are many types of graphs that you can use, but the most common and
effective types are the bar chart, as well as the line graph.
What to do if the results do not support your hypothesis
A scientist never gives up, even when the results
of the experiment do not support his or her hypothesis! A scientist realizes
that experiments with negative results are just as important as experiments
with positive results - because they both provide objective, valid information
about the topic that is being researched. You will notice that in some of the
science projects contained on this website, some of them actually have a hypotheses
that us rejected. Again, that's perfectly acceptable and the science fair
judges will not penalize you for failing to prove your hypothesis.
What should you do if the results do not support your hypothesis?
Do not change your hypothesis, even if you are tempted to do so. Instead, you
should provide possible explanations of why the results do not support your
hypothesis. You should also suggest an alternative experiment to solve your "problem".
What to do if the results support your hypothesis
If the results do support your hypothesis,
you should then summarize your findings in a short, clear narrative. Here, you
should also re-state the problem, the hypothesis and some ways you can improve
your experiment. In the example above, you could say:-
- My hypothesis is that Sunscreen
lotions with a higher SPF rating are
able to block a larger amount of UV radiation from the sun, and therefore
provide better protection".
- My experiments show sunscreen lotions with a higher SPF rating does in fact block
larger amounts of UV radiation, but only up to a certain point. I discovered
this by testing how much UV radiation was blocked by sunscreen lotions with
various SPF levels. We tested lotions with SPF 15, 30 and 50, and also
performed a control experiment with no sunscreen lotion. Increasing the SPF level
from 15 to 30 resulted in a lower UV radiation reading. However, increasing the
SPF level from 30 to 50 did not result in a further reduction in UV radiation.
- I performed my experiment with 5 brands of sunscreen lotion.
- To improve on my experiment, I would try the same process using a UV lamp instead,
for a more constant/stable source of UV radiation.