Biology 180 Study Questions: Nature of Science and Experimental Design Autumn 2018

Do these questions with your study group (~4 people). Follow these rules:
• Everyone must participate in every question that you work on.
• Listen to each person’s contributions.
• At least once during each study session, each group member should challenge another member’s ideas.
• Assign roles: If your group isn’t part of a large session with TAs or at Tribeta or the IC, one person
should serve as timekeeper; one as organizer (keeping people on task and getting a consensus on when to
move on to the next question for discussion); one as group-dynamics checker (making sure everyone is
heard and respectful); one as challenger (making sure that all ideas are carefully considered).
Format:
1. Each person should write answers to the questions on their own, under time pressure—3 minutes per
question, on average (exams usually have about 12 questions, total).
2. Compare your answers for each question and discuss.
3. Exchange papers and use the answer key to grade each other for full credit, partial credit, or no credit.
4. Each member of the group should write 2 exam questions, based on key concepts from this week. Use
these questions to quiz each other.
NOTE: All of these questions are Biology 180 exam questions from previous quarters. But because we
use different examples or cover different topics from year to year, some of these questions may not be
relevant to this fall’s class.

1. A group did the following experiment on trail-following behavior in termites. “Strokes” refers to the
number of passes with a pen.

Procedure:
Bic black (4 strokes) • Create 4 lines as shown to the left. “Pen cap” = no ink.
• Place 5 termites in the middle of the 4 lines.
Bic blue (3 strokes) • Record how many times termites follow each line.

PM blue (2 strokes)

Pen cap (1 stroke)

The hypothesis being tested is that termites prefer to follow lines from Bic pens. Identify 2 problems
with the experimental design. In each case, explain why the problem would confound the interpretation
of the results.

2. In experiments with human drugs or on natural populations of organisms, researchers use

randomization as a way to make sure that only one thing differs between treatment groups. Explain the
logic of this strategy.

1
3. Dr. Janneke Hille Ris Lambers (Go Dawgs!) has found that when tree seeds are at high density, few
germinate. One hypothesis to explain this pattern is that predation and disease increase when seed
density is high.

a) State the null hypothesis.

b) To test her hypothesis, she created many plots in a forest where seed density is high and many plots
where seed density is low. In half of these plots she’s excluding all predators and diseases. The other half
of the plots will be left alone. In terms of the hypothesis being tested, what is the one thing that differs
between the high density plots?

c) State two other conditions that might affect how many tree seeds germinate and that should be controlled
in this experiment.

d) She plans to control for these conditions by putting the plots in random locations. Why is this approach
valid?

e) If the hypothesis is correct, predict what the data should look like (add four bars with the height
indicating the average value—two bars open, two bar solid; one of each in each density condition).

100

No predators, disease
% of seeds
germinating Predators, disease present

0
Low density High density

4. Kevin Ford (Go Dawgs!) is doing experiments on Mount Rainier that involve transplanting plant
seedlings from low elevation habitats and high elevation habitats into the same environment. He plans to
compare plant growth in seedlings from the low versus high elevation habitats.

• All of the seedlings he is planting are two years old. What alternative hypothesis does this control
for?

• Soil conditions—which have a dramatic effect on plant growth—vary throughout the experimental
site. How can he control for the effect of soil conditions on his results? Explain your logic.

2
5. To test the hypothesis that bark beetles in the
Cascades will survive at the high temperatures
predicted under climate change, you maintain10
populations in the lab, each at a different temperature.
10 weeks later, you measure percent survival and
analyze the data.
a. State the null hypothesis and explain your
reasoning.

b. On the graph, draw a dashed line labeled “Null” to illustrate the prediction of the null hypothesis.

c. State 2 additional variables that should be controlled during the experiment.

1)

2)

d. A friend says that your experimental design is flawed because each of the 10 bark beetle populations
that you used came from a different location. How would you re-do your experiment?

6. Dr. Josh Tewksbury (Go Dawgs!) hypothesized that capsaicin molecules in chili fruits are distasteful
to cactus mice but not to curve-billed thrashers. To test this hypothesis he brought cactus mice and
curve-billed thrashers into the lab and presented them with chili fruits that contained capsaicin (“C
fruits”) and chili fruits with no capsaicin (“NC fruits”).

a. A good experimental design should alter only one thing between treatments. In this case, what is the
one thing?

b. To make sure that only one thing differed between treatments, he controlled other variables. Suggest
another variable that should be controlled. Explain why.

c. Suppose that he wanted to do this experiment in the field, under natural conditions, by setting out
trays containing both C and NC fruits. He plans to place the trays in random locations and record which
fruits were eaten by cactus mice versus curve-billed thrashers. Explain why this would or would not be a
good experimental design.

3
d. Should the researchers present the fruits in random order to the predators, or present all three fruits at
once? Explain your reasoning.

e. He also offered hackberry fruits, which are about the size of chili fruits but contain no capsaicin and
are readily eaten by mice and thrashers. Why did they bother to offer hackberry?

7. As a follow-up to the feeding experiment with chili fruits, the researchers germinated non-pungent
chili seeds that had been passed through thrashers, mice, or packrats, or had been planted directly into
the ground.
a. The researchers reasoned that capsaicin might be adaptive—that it deters mammals because
mammals destroy the seeds when they eat them, but that birds don’t. What prediction does this make
about the percentage of seeds that will germinate in each of these four treatments?

b. What is the null hypothesis here? What does the null hypothesis predict about the percentage of
seeds that will germinate in each of the four treatments?

8. Nemoria caterpillars either look like oak flowers or oak twigs, depending on whether they hatch
during the spring or during the summer. To determine how individuals from the same species can end up
looking so different, a researcher raised eggs under 8 sets
of conditions. All of the larvae in the boldfaced treatments
developed into caterpillars that looked like flowers. None Temp. Hours of Diet
of the larvae in the other treatments developed into light per day
caterpillars that looked like flowers. 150C 12.5 oak flowers
150C 12.5 oak leaves
a) This experiment was designed to test the hypothesis 15 C
0
14 oak flowers
that Nemoria caterpillars develop into flower-morphs if 15 C
0
14 oak leaves
exposed to spring-like physical conditions and twig- 250C 12.5 oak flowers
morphs if exposed to summer-like physical conditions. 250C 12.5 oak leaves
What is the null hypothesis here? 25 C
0
14 oak flowers
25 C
0
14 oak leaves

b) Do the results support the hypothesis being tested (about the effect of temperature and daylength)?
Explain why or why not.

c) The researcher also raised larvae under constant conditions of temperature and light, on an artificial
food that contained molecules called tannins. Tannins are abundant in oak leaves. Almost all of the

4
larvae developed into caterpillars that look like oak twigs. Does this result support or conflict with the
previous experiment? Explain.

d) He used 50 caterpillars in each treatment. Why didn’t he use 10 caterpillars? Why didn’t he use 1000
in each?

9. Design an experiment to test the hypothesis that Nemoria caterpillars develop into the twig-like
morph if they eat tannins (a molecule in oak leaves). You have a supply of tannins and artificial food
(doesn’t contain tannins), and Nemoria eggs from the spring and summer population. State your
experimental set-up, and the predictions of the “Tannins R Us” hypothesis and null hypothesis.

10. Eelgrass is a plant that grows in shallow ocean water. Design an experiment to test the hypothesis
that juvenile salmon are more common in low-density eelgrass beds versus high-density eelgrass beds.
The experiment will be conducted in a large expanse of eelgrass on the Pacific coast, where juvenile
salmon are common.

a. State the null hypothesis.

b. State the comparison groups that you will create and study.

c. State at least two other variables that might affect the abundance of juvenile salmon during the study,
and explain how you will control for them—so that only one thing varies between your comparison
groups.

d. What will you measure?

e. If the hypothesis being tested is correct,

predict what the data should look like. Be sure to
label the axes on your graph.

5
11. Dr. Freeman claims that large parts of the introductory biology textbook he used in 1976 are wrong.
Based on material covered in class, is the most likely explanation for this observation that the book was
inaccurate, sloppy, or otherwise of poor quality? Explain your reasoning. Include an example or other
data.

12. A molecule called capsaicin makes chili seeds pungent (hot-tasting). The Tewksbury lab (Go
Dawgs!) finds that capsaicin levels tend to be high in chilis from wet areas in Bolivia but much lower
(or even zero) in chilis from nearby dry areas. They hypothesize that capsaicin discourages fungi from
growing on the seeds. The fungus thrives best in wet areas and can destroy seeds. (Note that chili fruits
in Bolivia are not eaten by mice or rats, as they are in the southwest U.S.) Design an experiment,
conducted in the lab with field-collected seeds that have no capsaicin, to test this hypothesis. You have a
supply of purified capsaicin and fungi.

Null hypothesis:

Experimental set-up (treatments):

Name two other variables that you will control. State how each will be controlled and why.

If the hypothesis being tested is correct, use

the axes provided to the right to predict what
the graphed data will look like.

6
13. You are correcting lab reports on experiments about trail-following behavior in termites. One group
worked with 3 variables: The orientation of the line the termites are placed on, the color of the ink used,
and the type of line. 20 termites, chosen at random, were placed into each of test groups below. Each
was tested on a fresh line. The average distance that each termite traveled on the line was recorded in
centimeters.

Variable Test Group # 1 Test Group Test Group #3

#2
Orientation of Line Turns to right Turns to left Stays straight
Color of Ink Black ink Red ink Black ink
Type of Line Thick line Thick line Thick line

Results of Experiment:
Average distance traveled on line 5.26 cm 5.24 cm 5.25 cm

Consider three possible conclusions from the experiment. Indicate whether the conclusion is valid or
invalid and explain your reasoning.

Conclusion Invalid/valid Reasoning

The orientation of a line influences the
average distance a termite will travel on a
line.

Whether a line makes a right turn or stays

straight does not influence the average
distance a termite will travel on that line.
The type of line was identical in all 3 test
groups and there was virtually no change
in the average distance traveled. Thus, the
type of line does not influence the
average distance a termite will travel on a
line.

14. Most animals with legs walk on five toes. Wolves and domestic dogs walk on four toes, but have a
fifth toe (a “dew claw”) located well above their foot—on their lower leg.

a. What treatment groups would you create to test the widely accepted hypothesis that the fifth toe has
no function? (Note: many dog breeders routinely remove the dew claw in their puppies, shortly after
birth. The operation is almost painless and risk-free for the puppy.)

b. Suppose your data support the hypothesis that dew claws have no function. What would the data
suggest about the claim, under the theory of special creation, that species are static? Explain your
reasoning. (SAVE THIS QUESTION FOR NEXT WEEK—ONCE WE’VE COVERED THIS
MATERIAL.)

7
15. Some Japanese people state that if a young couple sits in an Alaskan hot spring and watches the
northern lights, they will conceive a son.

a. To test the “hot springs and northern lights” hypothesis, you recruit a large number of young
Japanese couples. Each couple gets a free Alaskan vacation at the same resort at the same time of year.
Name four treatment groups that you would create.

b. Explain how you would make sure that workplace stress, personal history, diet, or other relevant
variables would not affect the outcome of the experiment. Explain why your plan would be effective.

16. In class you analyzed an experiment on how the number of hours of sunlight exposure affects plant
growth. Name two conditions—other than water and fertilizer—that you would need to control during
the experiment. BRIEFLY explain why.

17. Dr. Cynthia Chang (Go Dawgs!) noticed that fireweed plants on Mount St. Helens are shorter in
forested areas and taller in open meadows. She hypothesizes that fireweeds are shorter in forests due to
reduced sunlight.

a. State the null hypothesis:

b. To test her hypothesis, she gathers 100 seeds from the

open meadow and plants them in two treatments (50 seeds
in each treatment) in the same open meadow. Treatment 1
= full sunlight, Treatment 2 = 50% sunlight reduction
under shade cloth. Her results are graphed to the right.
c.
Name the dependent variable:

Name the independent variable:

Treatment 1 Treatment 2

d. Based on these results, she concludes that “Fireweed plants are shorter in the forested areas due to
reduced sunlight.” Is this a valid conclusion?

e. Give two reasons why you answered yes or no.

8
18. You want to test whether termites prefer to follow lines made with blue Bic or blue Papermate pens.
In 10-second tests, you record time-spent-on-line for a termite placed on a 10-cm line made with fresh Bic
ink, fresh Papermate ink, and a dry ballpoint pen (a no-ink control). You propose to test one termite on
each treatment each day for 30 days, for a total sample size of 30 different termites in each treatment. A
lab partner says no—you should test different 30 termites on each treatment on the same day.

Which is the most convincing experimental design? (circle one) Yours Lab Partner’s

Why?

19. Since 1945, the USA has added fluoride to tap water. Fluoride is claimed to prevent dental cavities.
In an experiment, how would you test whether fluoridated water prevents dental cavities?

a. Manipulated (independent) variable b. Outcome (dependent) variable

c. What is another factor (besides your manipulated variable) that

might affect your outcome variable?

d. Explain how you would try to control for the influence of this variable.

e. Draw graphs that show the potential results of your experiment if…

…the NULL hypothesis were supported. …the hypothesis were

supported.

9
20. A UW Farm researcher thinks that tomato plants release more pollen, resulting in more tomato fruits
produced, if you release pollen with a tuning fork that vibrates at the same frequency as a bee’s wings in
addition to letting pollination occur by wind or other types of insects.

a. What is the hypothesis?

b. What prediction follows from this hypothesis?

c. What is the null hypothesis?

d. Design an experiment to test your prediction.

i. What are your treatment groups?
• Treatment 1:

• Treatment 2:

ii. Name three conditions you need to control for between the experimental and control treatments:

iii. What variable will you measure?

e. Assuming that your findings support the

null hypothesis, draw a graph showing
the results of your experiment. Be sure
to label the axes.

10
Answers
1. The answer must identify authentic issues (max 2); AND each explanation must articulate the concept that the
issue is a legitimate alternative explanation/hypothesis (“a critic could claim …” “the results could be due to
…”)
Some possibilities:
• Each line has a different number of strokes; thus, termites could be responding to differences in line
weight/depth or amount of ink—not brand.
• The termites are released in a group instead of individually, so the “counts” could represent the same or a few
individuals—not all.
• The termites are released in a group instead of individually, so termite response could be influenced by social
interactions instead of the nature of the lines.
• The termites are released in the center, so they could simply respond to the first line(s) encountered (e.g. never
encounter the outside lines) instead of making a choice among 4.
• The pen cap line is shorter; thus, termites could be responding to differences in line length—not brand.
• 5 termites is a small sample; the experiment would be more convincing (less susceptible to bias caused by a
few unusual individuals) with a larger sample of individuals.
• The “counts” are subjective—there is no way for a third-party to replicate measuring the response (“following
the line”) in a precise manner.
• If the hypothesis is that termites prefer Bic, then the use of both blue and black Bic confounds the design by
introducing a contrast with color. It would be better to compare blue Bic with blue PM.
• There is only 1 PM line and 2 Bic lines, so just following lines at random would predict that Bic would have a
higher chance of being followed.
• One line is shorter than the others; since experimenters are only recording how many times termites follow a
line, a pass over a shorter distance would count equally with a pass over the long lines.
2. In situations like this, it is not possible to control all variables that are relevant to the quantity you are trying to
measure. If you randomize treatments among experimental subjects or samples, then any differences among
those individuals or samples average out—except for the one factor hypothesized to affect the result.
3. a. Predation and disease do not affect seed germination.
b. The presence of predators and disease.
c. light availability, moisture, nutrients, soil quality/texture, slope, temperature, exposure to wind … others possible
but they must be relevant to seed germination
d. If plots are in random locations, then there is no bias in the physical conditions experienced by the seeds in
different treatments. On average, seeds in each treatment will experience the same range of conditions.
e.
100

% seeds germ.

0
Low density High density
4. The observed differences in growth rates are due to differences in the age of transplanting seedlings—not
elevation.
By randomizing the location of the seedlings planted in the experimental area and using a large number of seedling,
he can ensure that there is no systematic differences in the soil conditions experienced by seedlings from the two
source habitats. (They are each equally likely to end up in good or bad soil.)
5. a. There is no difference between treatments. Or Beetles (ants) have no difference in survival rates at the
different temperatures.
b. flat line
c. Any 2 reasonable variables (ones above are examples), e.g. environmental variables (light; temperature; humidity;
food source; water etc); experimental conditions (habitat type; volume (density); etc).

11
d. To control for where the population originates either a source population is harvested and split equally among the
treatment groups or each population is tested at the different temperatures and multiple experiments are conducted.

6a. Presence of capsaicin in fruits.

b. Present fruits in random order to control for tendency to eat first thing. or Light, temperature, time of day, nature
of arena (i.e. constant physical conditions) to control for differences in predator behavior that depend on conditions.
or Appearance and condition of fruits (color, size, age/ripeness) so that differences were minimal (other than
capsaicin), to control for other cues that predators might use to choose fruit.
c. This is a strong experimental design, in part because it tests predator behavior under natural conditions.
Randomizing locations and placing fruits together ensures that predator choice will not be biased by differences in
physical conditions or in tendency to encounter one fruit versus another. (As an aside, the downside of this design is
that it will be very difficult to measure predator response; also susceptible to bad luck like storms blowing fruits all
over)
d. It is important to randomize the presentation of the fruits to address the hypothesis that the order of presentation,
and not the type of fruit, affects the predator’s response. For example, predators might simply eat the first fruit
presented to them.
e. To test the hypothesis that the predators were hungry. To see why this is important, look at the data for cactus
mice. If hackberry hadn’t been included, a critic could say that cactus mice didn’t eat chilis because they weren’t
hungry.
7a High germination from seeds planted directly or passed through thrashers; no germination from seeds passed
through mice or packrats.
b The null hypothesis is that which predator eats a seed has no bearing on its germination. The prediction is that
there will be no differences among treatment groups.
8a Physical conditions have no effect on the development of Nemoria caterpillars.
b No—some caterpillars raised under the same physical conditions developed into different types of morphs, while
some caterpillars raised under different physical conditions developed into the same type of morph. The results
support the null hypothesis. (As an aside, the results also support the hypothesis that development depends on
biological conditions—specifically, the type of diet.)
c It supports and extends the hypothesis that development depends on diet by suggesting that tannins are a specific
molecule in the diet that affects development.
d Small sample sizes are susceptible to sampling error; large sample sizes are inefficient—they do not provide
additional useful information.
9. Randomly select a large and equal number of spring and summer eggs. Raise half the larvae on the artificial food
and half on the artificial food with tannins in it (same concentration as oak leaves). Raise all the caterpillars in a
location where all other conditions—light, humidity, temperature, amount of food available—are identical. Record
what % of larvae develop into the twig-like morph.
Tannins R Us predicts that all or most larvae on the + tannins diet develop into twig-like caterpillars.
Null hypothesis predicts no difference between treatments (either none are twig-like or equal numbers become twig-
like in the 2 treatments)
10a. Eelgrass density has no effect on the abundance of juvenile salmon.
b. Manipulate eelgrass via removal or planting to create study plots with high density of eelgrass vs. study plots with
low density of eelgrass.
c. Possible variables include tide strength, salinity, density of other plants, predator abundance, water depth, light.
Control for these by randomizing the location of plots throughout the study site, so that differences in other factors
besides eelgrass density average out.
NOTE: Students lost points even if they mentioned two factors and stated a correct way each could be controlled
(e.g. fencing to exclude predators). This is because randomizing the locations of study plots allows you to control for
ALL variables effectively (so that only one thing varies between comparison groups), and not have to painstakingly
try to control for every possible factor that might affect your results.

12
d.
Number of juvenile salmon
in each type of plot Juvenile
e. Salmon
abundance

Eelgrass density

11. No – scientific knowledge changes and improves as new data are obtained. New data are arriving at an
increasing rate, as the graph of pages published in PNAS show. Another example: the concept of blending
inheritance was once widely accepted in scientific circles, but is now considered inaccurate based on the research of
Mendel and subsequent geneticists.
12. This question was on Midterm 1 in Autumn 2009; notes from the grader are below:
• The question clearly indicates that this experiment is to be done in the lab; but no deductions were taken if you
ignored this and proposed a field experiment. In the future, points will be deducted for errors like this.
• Null: 0 points if you state the null as a prediction—meaning, something you would measure if the hypothesis being
tested is NOT correct. 0 points if you mixed up cause and effect in terms of the hypothesis you were asked to test,
stating that fungi cause differences in capsaicin. 0 points if you said that the hypothesis was about capsaicin production
(or fungal growth) in wet versus dry conditions. Deduction for stating that the null claims fungi increase with
capsaicin. Full credit for stating that capsaicin would have no effect on fungal growth.
• Exp set-up: for full credit, you must state a capsaicin/no capsaicin contrast and clearly articulate that both will be
exposed to fungi. I deducted points if you added a wet/dry contrast but then did not graph data that would have
resulted from this part of your experiment.
• Other controls: I accepted a wide range of proposed conditions (moisture, temperature, light, nutrients or soil
conditions, time of exposure to fungi, seed density, starting fungal density)—the key was to clearly articulate that they
must be controlled (same for both treatments) because it would be reasonable for a critic to argue that they affect
fungal growth—the response variable in this experiment. No credit for controlling for predators, as the question clearly
states that predators are not an issue in Bolivia and that the experiment is conducted in the lab, where predation is
automatically controlled.
• Graph (For full credit, the relationship between the predictor and response variables must be correct and the
relationship indicated in the data must support the hypothesis being tested. I deducted points if your experimental set
up indicated that you were only testing seeds with and without capsaicin but you drew a scatterplot or line suggesting
that you had tested a continuous range of capsaicin values. Full credit if you indicated that you were testing a range of
values and then drew a regression line.
13.
Groups 1&3 are well-controlled for line orientation but show little/no difference in distance
Invalid traveled. Also, you’ve only tested right vs. straight.
Valid Groups 1&3 are well-controlled for line orientation and show little/no difference in distance
travelled.
Invalid The conclusion is confounded by the differences in orientation of line and color of ink.

14. a. Group 1: With dew claw: domestic dogs (or wolves) from the same breed (or population) (would also be good
to mention similar age, health status, gender mix) that have intact dew claws, but that have been sedated and
operated on (the dew claws removed and replaced, if practical, or simply operated on but not removed).
Group2: Without dew claw: domestic dogs (or wolves) from the same breed (or population) (would also be good to
mention similar age, health status, gender mix) that have had their dew claws removed.
As an aside, note that many breeders routinely have dew claws surgically removed before puppies are sold, to
reduce the chance that the dew claws snag on branches and tear. So this experiment does not pose ethical problems.
b. Functionless (vestigial) traits support the claim that species have changed through time—that wolves and dogs are
derived from ancestors that had five functioning toes.
15. a.
Sit in hot springs and watch northern lights Watch northern lights only (no hot springs)
Sit in hot springs only (no northern lights) No treatment—no northern lights, no hotsprings

13
b. Assign couples to the treatment groups at random. As a result, on average there would be no difference in the
variables listed among the treatment groups.

16.
Condition to be Why
controlled
Time of year If treatments were run at different times of year, the quality of the light might be different.
Size of pot If roots do not have enough room, plant growth can be affected. So both treatments should have the
same amount of room.
Soil quality The texture and composition of soil can have an enormous impact on plant growth, so both treatments
should be planted in the same soil.
Presence of disease Diseases (bacterial, viral, fungal, or other) can reduce plant growth, so both treatments should be
disease-free or exposed to the same disease-causing agents.
Presence of Herbivores (animals that eat plants) can reduce plant growth, so both treatments should be herbivore-
herbivores free or exposed to the same numbers and types of herbivores.
Presence of Competitors (e.g. weeds) can reduce plant growth, so both treatments should be weed-free or exposed
competitors (e.g. to the same numbers and types of weeds.
weeds)
Presence of fungi on (this is something we’ll learn later in the course) The presence of beneficial fungi can have dramatic
roots effects on plant growth, so should be the same in both treatments.
17. Fireweeds are not shorter in forests due to reduced sunlight. OR Sunlight has nothing to do with height of
fireweeds. Dependent: Height; Independent: amount of sunlight. No.
• Using seeds from only the open meadow confounds the conclusion. Plants from the open meadow may have
undergone different adaptations than plants from the forest (be different genetically).
• Only growing plants in the open meadow confound the conclusion. Forested areas and open areas could have
different environmental conditions OTHER THAN reduced sunlight (e.g. more nitrogen, more herbivores).
• No replication—you should be skeptical of results until they are replicated.
• It is not clear that she planted the seeds in random locations, to control for local differences in soil, moisture,
etc.
• There is no control for the shade cloth manipulation (it could’ve changed temperature as well, or attracted
herbivores).
18. Lab partner’s. In my experiment, we wouldn’t know if the result was due to actual differences in termite
behavior or due to variation over time (in test conditions or the animals tested, from day to day)
19. a. fluoridation of water; b. # of cavities in 5 years; c. diet, dental hygiene, genetics, other reasonable hypotheses;
d. control it (only include people who report to brush regularly & visit dentists) OR pair-match (for each person in
group1 is matched to a similar person in group2) OR randomize (randomly assign a large number of people to each
group, so little average difference); e.

…the hypothesis were supported. …the NULL hypothesis were supported.

# cavities

fluoridated unfluoridated fluoridated unfluoridated

14
20. a. Tuning fork pollination influences pollen release.
b. If tomato plants/eggplants are induced to release pollen using a tuning fork, then those plants will produce more
tomatoes/eggplants than tomato plants/eggplants whose pollen is released only by the wind or insects.
c. Tuning fork pollination has no effect pollen release.
d. i. Tuning fork release of pollen; No tuning fork release of pollen
ii. All the same type of tomato plant/eggplant; equal access to light/water/fertilizer; randomize location of
individuals from the two treatments; same soil; same exposure to wind/insect pollinators
iii.Number of tomatoes/eggplants produced or amount of pollen released
e.
8

Tomatoes Per Plant

Average Number of
NOTE: on your exam, we’d also expect you to add standard
error bars, but you won’t learn about these until next week. 6
4
2
0
Hand No Hand
Pollination Pollination
Treatment

15