Title: Warm Up
1Warm Up
2Which percent confidence do we use in this
course?
3To reduce bias we
- Increase the sample size.
- Take a simple random sample.
- Cry about it.
4Which of the following is a random sampling error?
- Processing error
- Response error
- Margin of error
- Convenience sampling error
5Ideally what is the sampling frame?
- Sample
- Population
- Phone book
6Chapter 5
7Components of an Experiment
- Recall that experimental studies are
characterized by an active intervention to impose
some treatment. All experiments and many
observational studies are interested in the
effect one variable has on another - Response Variable a variable that measures the
outcome of the experiment (i.e. the Y variable in
an equation) - Explanatory variable the variable that we think
explains or causes the changes in the response
variable (i.e. the X variable in an equation) - Subjects individuals studied in an experiment
- Treatment A specific experimental condition
applied to the subjects. A treatment can be a
combination of specific values of explanatory
variables.
8EXAMPLE
- Does nicotine gum reduce cigarette smoking?
- Response variable cigarettes smoked
- Explanatory variable nicotine gum
- Does exercise help with weight loss?
- Response variable
- Explanatory variable
- NOTE Explanatory variable is also referred to as
independent variable and the response variable is
also referred to as dependent variable.
9Example
- An administrator at the university wants to know
if drinking different beverages in the morning
affects the students performance in class. He
decides to explore this situation. He randomly
assigns one group of 45 students to drink orange
juice every morning for one semester and another
group of 45 to drink coffee every morning. He
records their final grade in their earliest class
at the end of the study. - Who are the subjects?
- What is the explanatory variable?
- What are the treatments?
- What is the response variable?
10Hidden Components
- Lurking Variable
- a variable that has an important effect on the
relationship between the variables in a study but
is not one of the explanatory variables studied. - Confounded variable
- We say that two variables are confounded when
their effects on a response variable cannot be
distinguished from each other. - the confounded variables may be explanatory or
lurking variables. -
11Example
- A study is done to compare the progress of
students taking a course online versus taking the
course in the classroom. In order to measure the
progress of students, a test is given to the
students after taking the course and the test
scores are compared. - Explanatory variable course setting (online vs.
classroom) - Response variable test score after course
- Lurking variable Student preparation
- Confounded variables course setting and
Student preparation
12An article in a womens magazine says that women
who nurse their babies feel warmer towards their
infants than those that bottle-feed. However,
women choose whether to nurse or bottle-feed.
What is the explanatory variable?
- The way women feel about their babies
- Whether the women bottle-feed or nurse.
- The prior attitude of the women towards their baby
13An article in a womens magazine says that women
who nurse their babies feel warmer towards their
infants than those that bottle-feed. However,
women choose whether to nurse or bottle-feed.
What is the response variable?
- The way women feel about their babies
- Whether the women bottle-feed or nurse.
- The prior attitude of the women towards their baby
14An article in a womens magazine says that women
who nurse their babies feel warmer towards their
infants than those that bottle-feed. However,
women choose whether to nurse or bottle-feed.
What is the lurking variable?
- The way women feel about their babies
- Whether the women bottle-feed or nurse.
- The prior attitude of the women towards their baby
15Randomized Experimental Design
- In an experiment we want to make conclusions
about the effects of our explanatory variable. - Randomization is important so that we do not have
biased groups before we apply the treatments - The comparative design of the experiment allows
us to eliminate some of the confounding effects
that may be present. This allows us to make
conclusions about the effects of the explanatory
variables on the response variables. - Sample size is always important! Use a large
enough sample! -
16Randomized comparative experiments
- Best way to conduct an experiment is to compare
two or more treatments. -
- Reasons Shows us the effect of the explanatory
variable(s) on the response variable(s) by
eliminating a great deal of confounding. - This is exactly what randomized comparative
experiments do. - To conduct a randomized comparative experiment,
we use random assignments to divide our subjects
into groups. Each group receives a different
treatment. (We should try to keep our groups
about the same size.)
17Example
- Simplest randomized comparative experiment
divide subjects into two groups. - Control Group usually the group receiving the
placebo - Treatment Group the group actually receiving
the treatment - Additional types of randomized comparative
experiments - Type 1 Use an existing treatment as the control
group. Compare this group with a group receiving
a new treatment. (Example Is medicine A better
than medicine B?) - Type 2 Compare multiple new treatments by using
multiple groups. (Example Which works betters
medicine A, medicine B, medicine C or medicine D?
there is No control group.)
18Common Experiments
- Clinical Trials
- Clinical trials experiments that study the
effectiveness of medical treatments on actual
patients - Placebo a dummy treatment that is made to look
exactly like the true treatment (There are no
active ingredients in a placebo.) - (If the placebo is a pill, sometimes it is
referred to as a sugar pill.) - Main Reason to use a placebo Eliminate the
placebo effect - Placebo effect Positive response to a dummy
treatment. - The idea of receiving a treatment can affect the
outcome of a clinical trial.
19Example
-
- In order to study the effectiveness of vitamin C
in preventing colds, a researcher recruited 200
volunteers. She randomly assigned 100 of them to
take vitamin C for 10 weeks and the remaining 100
to take a placebo. The 200 participants recorded
how many colds they had during the 10 weeks. The
two groups were compared, and the researcher
announced that taking vitamin C reduces the
frequency of colds.
20Logic of experimental design
- Randomized comparative experiments are one of the
most important ideas in statistics. - They give us the opportunity to draw
cause-and-effect conclusions. - How do they accomplish this goal?
- By using randomization, we are able to form
groups that are similar in all respects before we
apply the treatment. - By using comparative designs, influences other
than the experimental treatment(s) operate
equally on all groups. - By using this type of design, it is easy to see
that differences in the response variables must
be due to the effects of the treatments.
21Principles of design of experiments
- Control the effect of lurking variables on the
responses, by comparing two or more treatments. - Randomize- use impersonal chance to assign
subjects to treatments so that we do not have
biased groups before we apply the treatments - Use enough subjects - reduces chance variation
in the results
22Statistically Significant
- To conclude treatments are different, the
differences in the treatments must be large
enough to conclude that they do come from the
variation that is always present in an
experiment. - Statistically significant - when an observed
effect is so large that it would rarely occur by
chance. - We will discuss it more fully in later chapters.
23Observational Studies
- There are many cause-and-effect questions we
would like to have answers to, but we cant
always perform experiments - Examples
- Do cell phones cause drivers to have more
accidents? - Does smoking cause lung cancer?
- To get answers to these and other questions, we
will need to perform observational studies.
24Observational Studies
- Even though observational studies are not as good
as experiments, they can still be helpful if they
follow some basic guidelines. - Comparative Observational studies should also
try to compare groups. The difference is that
the groups cannot be randomly assigned. - Matching This can be used in conjunction with
comparison. Here, we try to match up a person in
one group with a person in another group who has
similar characteristics. (i.e. age, gender,
education, number of children, and other possible
lurking variables.) - NOTE Comparison Matching cannot eliminate
confounding completely. - Solution Measure and adjust for confounding
variables.
25Example
- This example refers to a study that length of
life is explained by church attendance. However,
it is found that churchgoers were more likely to
be nonsmokers, physically active, and at their
right weight. Before drawing any conclusions the
study must measure and adjust for these
confounding variables. - Final Remarks
- Randomization is able to create groups that are
similar in ALL variables (known and unknown). - Matching and adjustment are only able to work
with variables that the study thought to measure.
If a lurking variable is not thought of
beforehand, observational studies cannot handle
them.
26Homework
- Chapter 4
- Suggested problems
- 1,2, 3, 4, 7, 9, 11, 12, 15, 17, 18, 19, 20,
21, 23, 25, 27, 29 - To turn in
- 2,4, 12, 18, 20
- Chapter 5
- Suggested Problems
- 1, 2, 3, 5, 7, 9, 10, 11, 13, 15, 16, 17, 18, 19,
21, 23 - To turn in
- 2, 5, 10,16,18
- Chapter 6
- Suggested Problems
- 1, 2, 3, 4, 5, 7, 9, 10, 11, 13, 15, 16, 17, 19,
20, 21 - To turn in
- 1, 2, 10, 13,16