STAT 100 Chapter 6: Chapter 6.docx

24 views2 pages

For unlimited access to Textbook Notes, a Class+ subscription is required.

Chapter 6 – Experiments in the Real World
“Designing an experiment”
oConcerned with statistical side of designing experiments, ideas that apply to experiments
in psychology, medicine, engineering, and other areas as well
Randomized comparative experiments assume that we treat all subjects alike, aside from
treatments received, but any other unequal treatment can cause bias
Equal treatment for all is to ensure that the placebo effect operates on all subjects b/c placebos
Doctors expectations change how they interact with patients and even the way they diagnose a
patient’s condition
oExperiments with human subject should be double-blind
Double-blind experiment – neither the subjects nor the people who work with them know the
treatment each subject is receiving
oClinical trials and other experiments with human subjects should be double-blind
whenever possible
Clinical trials – medical experiments involving human subjects
Refusal, nonadherers, and dropouts are problems in experiments with human subjects
oBias can result if those who refuse are systematically different from those who cooperate
oMinorities, especially blacks, more likely to refuse
oLack of trust in medical establishment
Remedies are : clear and clear information about the experiment, insurance
coverage for experimental treatments, participation of researchers with different
ethnic backgrounds, cooperation with doctors and health organizations in
different communities
Nonadherers – subjects who participate but don’t follow experimental treatment
oCan cause bias as well
Dropouts – subjects who begin the experiment but do not complete it
oCommon in experiments that continue over an extended period of time
oBias can result if subjects drop out due to their reaction to one of the treatments
treatments, subjects, or the environment of experiment may not be realistic
olack of realism limits ability to generalize beyond environment and subjects in their study
true scope of a new finding must usually be explored by a number of experiments in various
good experiments = statistical principles + understanding of a specific field of study
completely randomized treatments – all the experimental subjects are allocated at random
among all the treatments
ocan have any number of explanatory variables
ooften want to study combined effects of several variables simultaneously  interaction of
several factors can produce effects that could not be predicted from looking at effect of
each factor alone
completely randomized designs are simplest statistical designs for experiments but can be inferior
to more elaborate statistical designs
omatching subjects in various ways can produce more precise results than simple
matched pairs design – compares just two treatments
Unlock document

This preview shows half of the first page of the document.
Unlock all 2 pages and 3 million more documents.

Already have an account? Log in

Get access

$10 USD/m
Billed $120 USD annually
Homework Help
Class Notes
Textbook Notes
40 Verified Answers
Study Guides
1 Booster Class
$8 USD/m
Billed $96 USD annually
Homework Help
Class Notes
Textbook Notes
30 Verified Answers
Study Guides
1 Booster Class