Scribd
Upload
589 views

Stats 6

The document provides information about experimental design, including key terms and concepts. It discusses explanatory and response variables, control groups, randomization, replication, blocking, and blinding. Examples are provided to illustrate experimental design principles and identify potential flaws. The document also includes notes on lurking variables, placebos, and double blinding. Homework includes practice problems and studying for an upcoming quiz.

Uploaded by

api-294156876
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
589 views

Stats 6

The document provides information about experimental design, including key terms and concepts. It discusses explanatory and response variables, control groups, randomization, replication, blocking, and blinding. Examples are provided to illustrate experimental design principles and identify potential flaws. The document also includes notes on lurking variables, placebos, and double blinding. Homework includes practice problems and studying for an upcoming quiz.

Uploaded by

api-294156876
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 25

9/8/2015

Statistics

Mr. DeOms

Experimental Design

SWBAT explain the design process for an experiment


SWBAT analyze the quality of an experiment.
SWBAT build community and understanding by sharing images of self.

A lurking variable is a variable that affects


both of the variables of interest, but is either
not known or is not acknowledged.

1. As ice cream sales increase at a store along the


beach, the number of drowning deaths also
increases.

Lurking Variable: Number of people at the beach

2. There is a strong positive relationship between


the number of firefighters present during a fire
and the amount of damage that the fire causes.

Lurking Variable: size of the fire

3. The following conclusion was presented by Newsweek: Of all precollege curricula, the highest level of mathematics one studies in secondary
school has the strongest continuing influence on bachelor's degree
completion. Finishing a course beyond the level of Algebra 2 (for example,
trigonometry or pre-calculus) more than doubles the odds that a student
who enters postsecondary education will complete a bachelor's degree.

Lurking Variable: study habits of student

Optimism is the faith


that leads to achievement;

nothing can be done


without hope and confidence.
Helen Keller

Please get out a piece of loose


leaf paper for todays Notes

Experimental Design

Name
9/8/2015
Stats 1

Vocabulary in Todays Lesson

Experiment/ Observational Study


Survey
Blinding/ Double Blinding
Lurking Variable
Confounding Variable
Representative sample
Treatment
Experimental unit

Control group
Placebo effect
Randomization
Representative sample
Blocks
Matched pair design
Replication / Sample size
Retrospective Study
Prospective Study

Lets say we are trying to determine if angioplasty or bypass


is the better procedure for prolonging life for heart patients.

The question is: which procedure prolongs life more?


Survival time is the response variable.
The type of surgery performed (angioplasty or bypass) is the
explanatory variable because it may or may not affect the response
variable.
In this example, we have only one explanatory variable (type of
surgery), but in real life you would have several more explanatory
variables, including age, health, weight and other lifestyle factors

A response variable is the particular quantity that we


ask a question about in our study.
An explanatory variable is any factor that can
influence the response variable.
While there can be many explanatory variables, we
will only have a single response variable per study.

The distinction between explanatory and response variables


is similar to another classification. Sometimes we refer to
variables as being independent or dependent. The value of a
dependent variable relies upon that of an independent
variable. Thus a response variable corresponds to a
dependent variable while an explanatory variable corresponds
to an independent variable. This terminology is typically not
used in statistics because the explanatory variable is not truly
independent. Instead the variable only takes on the values
that are observed. We may have no control over the values of
an explanatory variable.

Experimental studies are preferred to observational


studies when showing correlation is not enough; when
we need to show causation between two variables.
Three main elements of great experimental design
- Control
- Randomization
- Replication

Control Include a group in the study that does


not receive the treatment

Placebo a neutral treatment to show the effect


on the response variable simply by including
subjects in the experiment

Placebo Effect - a subject's positive response to a


placebo

Blinding - the practice of not telling subjects


whether they are receiving a placebo or the actual
treatment
Double Blinding both the researchers and the
subjects do not know who is receiving a placebo or
the actual treatment

Randomization Randomly assign subjects


to different groups in order to limit the effects
of possible lurking variables.

Replication Larger sample sizes bring more


validity to experimental results

Blocking dividing subjects into subgroups


called blocks, such that the variability within
blocks is less than the variability between
blocks. Used to eliminate potential effects of
lurking variables.
Match Pair Design special case of
blocking used to control for two different
lurking variables

Group Analysis

Example 1: Decide whether the following experimental designs will lead to valid
results, discuss what could be improved
Part 1 - A company is testing the effectiveness of a new chewing
gum that prevents smoking. The company identifies ten adults
who are heavy smokers. Five subjects are given the new gum and
five subjects are given a placebo.

Example 1: Decide whether the following experimental designs will lead to valid
results, discuss what could be improved
Part 2 - The company now identifies 1000 adults who are heavy
smokers and divide the 1000 into blocks according to gender.
Females are given the new gum and males are given the placebo.

Example 1: Decide whether the following experimental designs will lead to valid
results, discuss what could be improved
Part 3 - The company identifies 240 adults and randomly assigns
each person to be in the treatment group or the control group. Each
subject is also given a DVD featuring the dangers of smoking.

Example #2 - Researchers want to identify whether taking aspirin regularly reduces the
risk of heart attack. Four hundred men between the ages of 50 and 84 are recruited as
participants. The men are divided randomly into two groups: one who takes aspirin
and one who take a placebo. Each participant takes a pill every day for 3 years.
1. Identify the population

___________________________________________

2. Identify the sample

___________________________________________

3. Identify the experimental units ___________________________________________


4. Identify the explanatory variable

___________________________________________

5. Identify the response variable

___________________________________________

6. Identify the treatment.

___________________________________________

7. How could the study use blocking? How could the study use blinding?
8. Is it possible to use double blinding? How or why not?

Example #3 - A running shoe manufacturer wants to test the effects of


its new sprinting shoe on 100-meter dash times. The company sponsors
5 athletes who are running the 100-meter dash in the Summer
Olympics. To test the shoe, it has all 5 runners run the 100 m dash with
a competitor's shoe and then again with their new shoe. The company
uses difference in time to determine their effectiveness.
1. What was the explanatory variable? What was the response variable?
2. What do you see as some flaws of this study?

Homework
Complete the Experimental Design Practice
Problems
Study all notes, definition notecards, and
classwork for our first Quiz on Thursday!

You might also like