4.2 Today you will work with your group to design an experiment. But first….There are a few things you need to know.

1) Observational study – The researcher observes outcomes without imposing any treatment.

2) Experiment – The researcher actively imposes some treatment in order to observe the response.3) In every experiment, you must clearly identify your explanatory and response variables.

Designing your experiment

1) You have 5 min. to design an experiment appropriate for the given scenario.*Be as descriptive as possible so that another person could implement your experiment exactly as you plan to.

2) You will join with another group that has your same scenario (same color paper). Each group will have 2 -3 min. to describe their experiment to the other group. The listening group members should take notes as they listen.

3) Once both groups have presented, each group may offer advice on ways to improve the other group’s experiment.

*You will need to select a presenter for your group

Designing your experiment

4) You will then learn about the details of experimental design with the 4.2 notes.

5) During the lesson, keep your experiment in mind as you will be given a chance to make improvements to your design prior to turning it in. *This lesson will take two days. Your experiments are not due until tomorrow, but you will turn them in today at the end of the hour so I can check your progress.

4.2Designing Experiments

After this lesson you should be able to:• Name and describe components of

experiments

• Diagram an experiment

I’ve developed a new rabbit food,

Hippity Hop.

Rabbit Food

Makes fur soft & shiny!

Increases growth!

Better than any other

rabbit food out there

Hippity Hop

I’ll use my pet rabbit, Lucky!

Since Lucky’s coat is softer & he has more energy, then Hippity Hop is a better rabbit food!Can I just make these claims?

What must I do to make these claims?

Who (what) should I test this on?

What do I measure (response variable)?

Do an experiment

Rabbits

Chose one variable that will be “evidence” that your food is better or not.Weight Gain, How soft fur is? Energy level,

etc

I’ll choose Weight Gain

• Experimental unit – the individual (person, animal, object, etc.) to which the treatments are assigned.

• Experimental units who are human beings are called Subjects.

More definitions:

• Response variable – what you are measuring

• Treatment – a specific experimental

condition applied to the units

• Factor – the specific levels of the

explanatory variables

I plan to test my new rabbit food by doing an experiment.

What are my experimental units?

What is the response variable?

Rabbits

How well they grow….(weight gain)

• Control group – a group that is used to compare the explanatory variable against; can be a placebo or the “old” or current item on the market

Definitions continued:

Old Food Hippity Hop

Now I’ll use Lucky & my friend’s rabbit, Flash.

Lucky gets Hippity Hop food & Flash gets the old rabbit

food.

WOW! Lucky is bigger so Hippity Hop is better!WOW! Lucky is bigger so Hippity Hop is better!

Lucky

Flash

Measurements are usually subject to variation. Use enough subjects or units to reduce variation.

• Replication – the practice of assigning each treatment to many experimental units.

Old Food Hippity Hop

The first five rabbits that I catch

will get Hippity Hop food and the remaining five will get the old food.

The Hippity Hop rabbits have higher weights so it’s the better food!

• Randomization – The use of chance to assign experimental units to treatment groups.

Systematic differences among the groups of experimental units in a comparative experiment cause bias.

Randomize to reduce bias.

Old Food Hippity Hop

3

5 7

8 9

41

2 6

10

I evaluated the rabbits & found that the rabbits eating Hippity Hop are gaining more weight than the rabbits eating the old food!

97

975

836

1 2 10

4

583

Number the rabbits from 1 – 10.

Place the numbers in a hat.The first five numbers

pulled from the hat will be the rabbits that get

Hippity Hop food.

The remaining rabbits get the old food.

Principles of Experimental Design

• Control• Replication• Randomization

The ONLY way to show cause & effect is with a well-designed, well-controlled experiment!

- reduces bias by spreading any uncontrolled confounding variables evenly throughout the treatment groups.

• Placebo – a “dummy” treatment

that can have no physical effect.• Placebo Effect – a measurable,

improvement not attributable to a treatment that has been administered.

Definitions…almost done

• blinding - method used so that

units do not know which treatment they are getting• double blind - neither the

units nor the evaluator know which treatment a subject received

Suppose we wish to test a new deodorant against one currently on the market.

10 people

Random

A

ssig

nm

ent

Treatment A

New Deodorant

Treatment B

Old Deodorant

Group 1

Group 2

Compare

Perspiration

amount

Completely Randomized Experiment

Treatment B

Randomly assign experimental units

to treatments

Treatment A

Completely randomized design

20 total people

Treatment A New

Deodorant

Treatment B Old

Deodorant

Treatment A

New Deodorant

Treatment B

Old Deodorant

Random

A

ssig

nm

ent

Random

A

ssig

nm

ent

Block A

10 Female

Block B

10 Male

Compare

Perspiration amount

Block Design– units are blocked into groups and then randomly assigned to treatments.

Compare

Perspiration amount

Blocking also helps reduce variability.

–match up experimental units according to similar characteristics & randomly assign one to one treatment & the other automatically gets the 2nd treatment

–OR have each unit do both treatments in random order

•Matched pairs - a special type of

block design

Pair experimental units according to

specific characteristics.

Next, randomly assign one unit from a pair to Treatment A. The other unit gets Treatment B.

Treatment A Treatment B

This is one way to do a matched pairs design – another way is to have the individual unit do both treatments (as in a taste test).

Example 1: A farm-product manufacturer wants to determine if the yield of a crop is different when the soil is treated with three different types of fertilizers. Fifteen similar plots of land are planted with the same type of seed but are fertilized differently. At the end of the growing season, the mean yield from the sample plots is compared.

Experimental units?

Factors?

Response variable?

How many treatments?

Plots of land

Fertilizer types A, B, & CYield of crop

3

Diagram the experiment using a Completely Randomized Design

15 Plots

Group 1

5 plots

Group 2

5 plots

Group 3

5 plots

Fertilizer A

Fertilizer B

Fertilizer C

Compare

Yield of crop

Random

Allo

cati

on

Why is the same type of seed used on all 15 plots?

What are other potential extraneous (lurking) variables?

It is part of the “control” in the experiment.

Type of soil, amount of water, etc.

Example 2: A consumer group wants to test cake pans to see which works the best (bakes evenly). It will test aluminum, glass, and plastic pans in both gas and electric ovens.

Experiment units?

Factors?

Response variable?

Number of treatments?

Two factors - type of pan & type of oven

Type of pan has 3 levels (aluminum, glass, & plastic) & Type of oven has 2 levels (electric & gas)

How evenly the cake bakes

6

Cake batter

Example 3: Four new word-processing programs are to be compared by measuring the speed with which standard tasks can be completed. One hundred volunteers are randomly assigned to one of the four programs and their speeds are measured.

Is this an experiment? Why or why not?What type of design is this?

Factors?

Response variable?

Yes, a treatment is imposed.

Completely randomizedone factor: word-processing

program with 4 levels

speed

Example 4: Four new word-processing programs are to be compared by measuring the speed with which standard tasks can be completed. One hundred volunteers are randomly assigned to one of the four programs and their speeds are measured.

Is there a potential confounding variable? Can this

design be improved?

Explain.

YES, the speed at which they could type already

You could do a block design where each person uses each

program in random order.

Example 5: Suppose that the manufacturer wants to test a new fertilizer against the current one on the market. Ten 2-acre plots of land scattered throughout the county are used. Each plot is subdivided into two subplots, one of which is treated with the current fertilizer, and the other with the new fertilizer. Wheat is planted and the crop yields are measured.

What type of design is this?

When does randomization occur?

Matched - pairs design

Randomly assigned treatment to first acre of each two-acre plot

Larger samples produce statistics with less variability. Blocking also helps reduce variability.

Name two ways to reduce variability?

4.2 Designing Experiments• I can identify components of

experiments• I can diagram an experiment

Homework: # 45–56, 61–63, 67, 72, 77, 78, 84