Graphing basics

Often times, it is appropriate to graph the data/results you obtain from an experiment. This
provides a visual representation of your data so that it can easily be interpreted by the
reader. Look at the examples below to see the proper way for formatting a graph. Both
graphs were made using the same set of data; however the bottom graph does a better job
at conveying the information. Note: The data in these graphs are completely made up.
Example of a poor
graph

The title has a label (i.e.

“Figure 1”) that can be
referenced when the
author writes the
experiment’s
conclusions. Also, the
Example of a good title provides enough
information so that the
graph reader can deduce what
specific hypothesis was
being tested. One
common way to write a
The
graph’s title is to say,
dependent
“The effect of the
variable is
(independent variable)
plotted along
on the (dependent
the y-axis.
variable).”
Also, the
dependent
The identity of
variable is
each data set is
clearly
clearly stated.
stated. If
Don’t leave the
applicable,
reader guessing
the units
as to what each
used to
line represents
measure the
(the person who
variable are
made the “poor
given.
graph” above
The independent variable is plotted along the x-axis. Also, the makes the reader
independent variable is clearly stated. The units used to measure do a lot of
the measure the variable (minutes) are also given. guessing by
writing “Bacteia
1”, etc.).
When to include linear regressions

Sometimes, when making a line graph, it may be useful to include a linear regression (i.e.
a “best fit straight line” through all the data points). This is useful when (1) the data points
suggest a trend (i.e. generally increasing or decreasing), and (2) the experimenter wants to
impress upon the reader that the values for the independent variable can be used to
predict the values for the dependent variable, beyond what’s shown on the graph.

Including a linear regression would be appropriate for the data shown on the previous page
because an overall trend is apparent (the longer bacteria are exposed to UV radiation, the
more mutations occur). Also, the experimenter might want to suggest that she can predict
how many mutant cells she would see in a population of E. coli that have been exposed to
20 minutes of UV radiation. After linear regressions are included, it would look like this:

The equation for each linear regression is given in the form of y = mx + b, where m
represents the slope of the line.
For data sets like the one shown below, it is usually not appropriate to include a linear
regression. This is because there is no overall increasing trend, nor is there an overall
decreasing trend.

When to use line graphs vs. bar graphs

With some data sets, it is more useful to present them using bar graphs, instead of line
graphs. This is the case when:

(1) The independent variable is qualitative (not quantitative).

(2) There is a relatively small data set.
(3) The experimenter is not trying to show a trend over time.

For example, suppose that someone performs an experiment to test whether smiling
during the day affects the amount of endorphins (a type of hormone) produced by people.
Here are the data produced by this experiment (note: These data are completely made up):

Men
Smiles at least once during the day? Amount of endorphins present per liter of
blood
Yes 1.15 pmol
No 0.77 pmol

Women
Smiles at least once during the day? Amount of endorphins present per liter of
blood
Yes 1.33 pmol
No 0.70 pmol
In this experiment, the independent variable (smiling) is qualitative (the people either did it,
or they did not – no number is being used to describe the amount of smiling). Also, there is
a relatively small data set; there are only four data points. So, it would be more useful to
depict these data using a bar graph:
Name ___________________________________________________ Instructor: _________________
Directions: Read through the introduction, then answer the questions below.
1) Suppose you want to test whether an element’s melting point is influenced by its
atomic number. After experimentally determining the melting points for various
elements, you come up with the following results:

Melting point for element with atomic number of 1: 14.06 K

Melting point for element with atomic number of 2: 1.15 K
Melting point for element with atomic number of 3: 453.74 K
Melting point for element with atomic number of 4: 1551.20 K
Melting point for element with atomic number of 5: 2352.20 K
Melting point for element with atomic number of 6: 3773.15 K
Melting point for element with atomic number of 7: 63.34 K
Melting point for element with atomic number of 8: 54.80 K

a) What’s the independent variable in this experiment?

b) Is the independent variable quantitative or qualitative?

c) What’s the dependent variable in this experiment?

d) Is the dependent variable quantitative or qualitative?

e) Consider the hypothesis that’s being tested in this experiment. Based on your data,
would you conclude that this hypothesis is supported? Why or why not?
2) Suppose you want to determine how effective a new vaccine is at protecting monkeys
against infection with Virus X. So, you gather two groups of monkeys; each group
consists of 50 monkeys. You administer the test vaccine to one of the groups (Group
A). The other group (Group B) receives saline solution, instead of vaccine. Next, you
expose both groups to Virus X. You find that 60% of the of the monkeys in Group A end
up showing symptoms of viral infection. In Group B, 65% of the monkeys end up
showing symptoms of viral infection.

a) What’s the independent variable in this experiment?

b) Is the independent variable quantitative or qualitative?

c) What’s the dependent variable in this experiment?

d) Is the dependent variable quantitative or qualitative?

e) Based on your data, would you conclude that this new vaccine is effective at
protecting monkeys against infection with Virus X? Why or why not?
3) Suppose you want to study the following chemical reaction:
Chemical A + Chemical B → Chemical X + Chemical Y

Specifically, you want to determine how the pH of the chemical reaction’s environment
affects the rate at which Chemical X is produced. After performing your experiment, you
get these results:

At pH 2: At pH 7:
Time after the Amount of
reaction begins Chemical X Time after the Amount of
(mins) produced reaction begins Chemical X
(OD units) (mins) produced
0 0.00 (OD units)
1 0.05 0 0.00
2 0.10 1 0.25
3 0.17 2 0.49
4 0.21 3 0.81
5 0.28 4 1.04
5 1.35
At pH 10:
Time after the Amount of
reaction begins Chemical X
(mins) produced
(OD units)
0 0.00
1 0.14
2 0.29
3 0.50
4 0.66
5 0.88

a) What’s the independent variable in this experiment?

b) Is the independent variable quantitative or qualitative?

c) What’s the dependent variable in this experiment?

d) Is the dependent variable quantitative or qualitative?

e) Based on your data, which pH results in the greatest reaction rate?