Chapter 2

SCIENTIFIC
EXPERIMENTS
Designing a Scientific Experiment
Some good starting points for designing an experiment:

1. OBSERVE something you are curious about.

2. ALTER a previous experiment to develop

your own experimental plans.

3. REPEAT past experiments to

see if you get the same results.

11
An experiment requires a detailed list of steps,
or a PROCEDURE, and a list of
materials needed to conduct the
experiment. Another scientist should
Procedure
a step-by-step list of
be able to repeat the experiment how to carry out the
based on your procedure steps experiment
alone, no matter where he or she
is. This allows other scientists to
evaluate your results.

Control
You can have a CONTROLLED a trial where all the
EXPERIMENT by running variables are held constant.
A control is used as the
an experiment more
standard of comparison for
than once: first without your experiment.
changing any factors
(this experiment is called
the CONTROL) and then a second time, changing only the
factor you want to observe.
In a controlled experiment, the
factors that are held constant Constants
all the variables in
are called CONSTANTS, and an experiment that
they don’t affect the outcome remain the same.
of the experiment. A VARIABLE
is a factor that can alter your
experiment’s results-a controlled experiment allows you to
test the influence of the variable.

12
In order to test only one factor, all other factors in the
experiment are held constant-this ensures that the changes
you observe are caused by the one variable you changed.

Different variables play different roles:

An INDEPENDENT VARIABLE is the variable

that you change in an experiment on purpose.

A DEPENDENT VARIABLE is the variable that is

influenced by the independent variable, the results
of your experiment.

13
 EXAMPLE: Goldfish Experiment
Every couple of weeks, the teacher has to buy a new goldfish
after the earlier one has died. The class comes up with a
hypothesis that the goldfish is not getting the right amount
of food. They devise an experiment to test this factor alone,
holding all other variables (type
of fish, tank size, water quality, CONSTANTS
water temperature, food type, 1. Type of fish
2. Tank size
and location) as constants.
3. Water quality
4. Water temperature
In this experiment, the 5. Food type
independent variable is how 6. Location
frequently they feed the
fish (either once a day or once every other day), and the
dependent variable is the health of the fish after two weeks.

experimental CONTROL
14
COLLECTING DATA
Good data is specific and detailed. Data with quantitative
descriptions, or measurements, is often helpful. Good data
is also accurate. Observe and measure things carefully. It is
easy to forget things, so in order to ensure accuracy, record
data and observations during
the experiment instead of
after. Without reliable data,
conclusions are meaningless!

ANALYZING and
PRESENTING
DATA
Here are some common ways
to organize and display data:

TABLES present data in rows and columns. Because

the numbers are side by side, tables can be read quickly
and numbers can be easily
compared. A table is the best
way to record data DURING
an experiment.

15
Once you’ve collected data in a table, turn it into a GRAPH to
make the information easier to see.

LINE GRAPHS show the relationship between two

variables-one plot ted on the x -axis (the horizontal line),
the other on the y -axis (the vertical line). A SCALE on each
axis shows the intervals of measurements. The scale should
go up in EVEN INCREMENTS, for example:
2, 4, 6, 8 . . . or 5, 10, 15, 20 . . .-not 2, 5, 7, 15 . . .

This is the
y-axis.

This is the
x-axis.

Line graphs help to show how one variable affects another, or

in other words, how the dependent variable changes because
of the independent variable. The independent variable is
on the x-axis, and the dependent variable is on the y-axis.
Line graphs work best for experiments that show continuous
change over time, such as the growth of a plant or the
acceleration of a race car.
16
A SCATTER PLOT is a type of line graph that shows
the relationship between two sets of data. Scat ter plots
graph the data as ORDERED PAIRS (these are simply pairs
of numbers-but the order in which they appear together
mat ters).

EXAMPLE: After a math test, Ms. Phinney asked

her students how many hours they studied. She
recorded their answers, along with their test scores.

NAME NUMBER OF TEST SCORE

HOURS STUDIED
Tammy 4.5 90
Latril 1 60
Sophia 4 92
Michael 3.5 88
Monica 2 76
Dave 5 100
Eva 3 90
Lance 1.5 72
Becca 3 70
Sarina 4 86

17
 To show Tammy’s data, we mark the point whose
x-value is 4.5 and whose y-value is 90.

ed
100 — Ac it!

95 —

90 —
TA M M
Y

85 —
TEST  SCORES

80 — FIT
E ST
O FB
NE
LI
75 —

70 —

65 — ly y
ea l d
I r to stu
ed re
60 — ne mo

55 —
| | | | | | | | |
1 1.5 2 2.5 3 3.5 4 4.5 5
HOURS  STUDIED
By graphing the data on a scatter plot, Ms. Phinney and her
students can see if there is a relationship between the number
of hours studied and test scores. The scores generally go up
as the hours spent studying go up. This shows that there is a
relationship between test scores and studying.
18
Eva studied 3 hours and got a 90. Becca also studied for 3 hours, but
got a 70. A scatter plot shows the overall relationship between the
data, while individual ordered pairs (like Eva or Becca) don’t show
the general trend. Eva and Becca might be considered OUTLIERS
in this situation because they don’t follow the typical pattern.

We can draw a line on the graph that roughly describes

the relationship between the number of hours studied and
test scores. This line is known as the LINE OF BEST FIT
because it is the best description of how the points are related
to one another. None of the points lie on the line of best fit,
but it’s okay! This is because the line of best fit is the line that
best describes the relationship of all the points on the graph.

BAR GRAPHS present the data as rectangles of different

heights. Each rectangle represents a different part of a
category, or variable, such as type of pet or favorite ice
cream flavor. The taller the rectangle, the larger the number.
Number o f People

Favorite Pet
19
You can think of a CIRCLE GRAPH like a pie cut into slices.
(Circle graphs are also sometimes called PIE CHARTS.) MMM.

A graph should be titled and labeled with things like

scale and units so readers can interpret the data.

DRAWING CONCLUSIONS
Did your results support your hypothesis? If not, how would
you change your hypothesis to fit your results? Sometimes
conclusions aren’t immediately apparent and you will have to
INFER, or use observations and facts, to reach a conclusion
about something you may not have directly witnessed.

For example, if you want to find out what a

Tyrannosaurus rex ate, you might observe the
types of fossilized droppings that could be
20
found near a T. rex ’s fossils. If you see crushed bones, you
might infer that the T. rex ate smaller animals or dinosaurs.
When you need to infer, it can help to look at background
information and do further research.

Conclusions are also a place to be critical of the experiment

and findings: Were there any errors in measurements or
otherwise? Was the procedure followed correctly? How
precise is your equipment? Even if you carried out the
experiment error free, the results aren’t always the same.
Constants are hard to hold perfectly constant. Unwanted
variables may be affecting your results. In order to ensure
that findings are accurate, conduct several trials of the
experiment.

EXAMPLE: Plant Fertilizer Experiment

Bob wanted to investigate the effects of plant

fertilizer. He bought three identical plants and gave
plant 1 fertilizer every morning, plant 2 fertilizer
once a week, and he kept plant 3 as the control (he didn’t give
it any fertilizer).

Bob watered each plant once every morning and set all three
on the windowsill so they all would have the same amount of
sunlight. (The sunlight and water are constants.)
21
He measured the height of each plant once a week and
recorded it in a data table. In order to analyze the data,
Bob graphed the results. He plotted height versus time for
each plant :

PLANT HEIGHT
PLANT WEEK 0 (START) WEEK 1 WEEK 2 WEEK 3
1 6 cm 8 cm 10 cm 12 cm
2 6 cm 7 cm 8 cm 9 cm
3 6 cm 6.5 cm 7 cm 7.5 cm

12 —
PLANT 10 —
HEIGHT
(cm) 8—
6—

KEY: 4—
= Plant 1 2—
= Plant 2
= Plant 3
1 2 3
WEEK #

With the help of his data and graph, Bob concluded that plants
given fertilizer every day grow four times as quickly as plants
that are not fertilized. Using the evidence that plant 1 grew
more quickly than plant 2, he also concluded that providing
fertilizer daily instead of weekly makes the plant grow faster.
22
 The Engineering
Design Process
ENGINEERING is a branch
of science that studies the
engineering
design, building, and use of a branch of science that studies
machines and structures in the design, building, and use
of machines and structures to
order to invent new products
solve real-world problems
that solve problems. Just as
scientists use scientific inquiry
to investigate questions, ENGINEERS use the ENGINEERING
DESIGN PROCESS to solve problems through invention,
design, and innovation. For example, engineers are currently
developing pavement that can collect solar energy and use
it to light roadways; this innovation can potentially solve a
number of problems-it can help light up roads for safer
night-time driving, it uses renewable energy to do it, and it
cuts the cost of lighting roadways. To get to a solution like
this one, however, engineers usually follow a certain path.

23
 The major branches of engineering are:
MECHANICAL: deals with mechanical power and
designing mechanical systems, machines, and tools;
studies forces and motion
CHEMICAL: works with raw materials and chemicals;
discovers new materials and processes
CIVIL: includes designing and constructing buildings,
roads, bridges, dams, and more
ELECTRICAL: studies electricity and the design of
electrical systems such as computer chips
And there are many more types of engineering: computer,
aerospace, biomedical, automotive, manufacturing, geological, etc.

Just as scientific inquiry has specific steps to carefully

answer a question, the engineering design process has a
system to help guide an engineering project. The engineering
design process begins with a problem or need that can be
solved by design. For example, oceanographers may want
to explore and learn about deep seabeds, but divers have
a hard time moving in the deep, fast-moving currents.
An engineer would do BACKGROUND RESEARCH on the
SHORT FOR “ DESIGN
problem, determine all of the DESIGN SPECS SPECIFICATIONS”

(requirements) needed to
start a design, and identify
Design Specs
CONSTRAINTS (restrictions) the requirements that an
that may affect his or her design. engineer must fulfill in
his or her design
For example, an engineer might
24
 Constraints
research what kind of information restrictions or
the oceanographers are looking limitations (can be
for in the seabeds. Some design physical, social, or
financial)
specs might include how deep divers
must go and how fast the currents
move. The engineer would also find out about constraints,
like how much money they can spend on the solution and
which materials will work far
underwater.
H EEE
ELLP
P P!

After a problem is identified

and all the necessary information is gathered, the next
step is proposing possible solutions. In scientific inquiry, you
formulate a hypothesis, but in engineering you establish a
DESIGN STATEMENT-this defines what it means to solve
the particular problem. Engineers often brainstorm many
ideas and evaluate each solution in order to choose which
is the best option. For example, the engineer who wants to
solve the problem of exploring deep seabeds may come up
with a motor that a diver can wear or an underwater robot
that pushes against the fast-moving currents and transmits
information. They ask, which approach most likely will work
best? And why?

Y!
G WA Y !
O N A
W R NG W
O
WR

25
 How do you choose which solution is best? Designers often think about
these universal design criteria when deciding which is the best choice:
Robustness (strength) • Cost
Aesthetics (looks) • Resources • Time
Skill required • Safety • Elegance

Then, engineers design and build a

PROTOTYPE of the solution, which PROTOTYPE
a preliminary model that
is like the first draft of a paper-it’s can be easily adjusted
a rough idea of what the solution
might eventually be. Engineers make
technical drawings and crunch numbers to construct a simple
prototype that can be easily adjusted depending on how it
performs. The engineer may decide that an underwater robot
that looks and acts like a crab may be the best solution to
the divers’ problems-it can stabilize
R OB O-
itself with its six legs and carry CR AB
cameras and sonar equipment to send
information back to the surface.

Once the design is complete, engineers construct a simple

prototype by using the drawings as a blueprint.

You can design in many different ways—with drawings,

computer models, storyboards, etc. You can also create
prototypes with lots of different materials—scrap wood, toy
blocks, poster board, or with more advanced materials like
metal, plastics, or even by printing out parts with a 3-D printer!

26
 NICE .

Next, it’s time to test how the

prototype holds up in the real world!
Engineers test the product multiple times to see how it does
under different conditions. They collect data on how well the
product solves the problem. If it doesn’t work well, then they
either go back to brainstorming new solutions or redesign the
prototype. Often they troubleshoot the ways that the design
is not meeting requirements or expectations. As a result of
testing the prototype in the real world, they find ways
to improve the design, then adjust their prototype
or make a new one. After going back through the
steps several times and
making improvements each
time, hopefully, they find a
solution that works.

Just like an experiment isn’t a failure if it doesn’t work out

as predicted, a prototype that doesn’t work in the real world
can lead to new discoveries and ideas. Knowing what is NOT
working is an important part of figuring out what WILL work.

Last, engineers construct a final product. Just like the final

draft of a paper, engineers tweak their design until it is
perfect. Then, they use final design
materials in order to create a final
product and present it to the public
(and possibly sell their invention!).
27
 e e r i n g
Engi n 1. Define the problem.

D es i g N 2. Do background research.

c e s s 3. Determine design specs

Pro and co ns traint s.

orm
a d es ig n st a tement : Brainst
4. Create lu tions, and choo
se
te p os si bl e so
ideas, evalua option .
ha t w ill , ho p ef ully , be the best
w

5. Design a prot ot ype.

yp e.
6. Cons truct a prot ot

7. Test the prot ot ype.

8. Evaluate it : Does it
solve the problem well?

u ct
s t r u c t a final prod
n
9B. N O? 9A. YE S? Co nt to t he public.
a n d p re s e
28
w CheckYour Knowledge
Match the term with its correct definition:

  1. P
 rocedure a. T
 his factor depends on the
independent variable. It is
  2. I
 ndependent variable usually the observed outcome
(result) of an experiment.
  3. D
 ependent variable b. A
 trial where all the variables
are held constant
  4. C onstants c. T
 he variable that is changed on
purpose in an experiment by
  5. C ontrol the scientist
d. F actors in an experiment that
  6. I
 nfer
remain the same
e. A
 step-by-step list of how to
carry out an experiment
f. U se evidence to draw
conclusions about things you
may not have directly observed.

In the park, there are 25 pigeons, 15 squirrels, 5 rabbits,

and 5 stray cats.

  7. M
 ake a table for this data.

  8. D
 raw a bar graph to represent this data.

  9. W
 hy can’t you draw a line graph using only this information?

answers 29