Show Menu
Cheatography

STAT 100 Chapter 5 6 7 Cheat Sheet (DRAFT) by

This is a draft cheat sheet. It is a work in progress and is not finished yet.

Key Terms

Response variable: variable that measures an outcome or result of study.
Explan­atory variable: variable that we think explains or causes changes in the response variable.
Indivi­duals studied in an experiment are often called subjects
A treatment is any specific experi­mental condition applied to the subjects. If an experiment has several explan­atory variables, a treatment is a combin­ation of specific values for these variables.
Lurking Variable: variable that has an important effect on the relati­onship among the variables in the study but is not one of the explan­atory variables studied
Two variables are confounded when their effects on a response variable cannot be distin­guished from each other.
 
the confounded variables may be either explan­atory or lurking variables.

Chapter 6 Key Terms

double blind experi­ment: neither the subject nor the people who work with them know which treatment the subject is receiving
clinical trials: medical experi­ments involving human subjects.
nonadh­erers: subjects who partic­ipate but don't follow experi­mental treatment
experi­ments that continue over an extended period of time also suffer dropouts: subjects to begin the experiment but do not complete it.
 
If a subject drops out because of their reaction to one of the treatm­ents, bias can occur.
a well designed experiment tells us that changes in the explan­atory variable must cause changes in the response variable.
 
make sure that your findings are statis­tically signif­icant, that they are too strong to occur by chance.
Completely Randomized Design (exper­imental design): all the experi­mental subjects are allocated at random among all the treatm­ents.
Matched Pairs Design (matching and random­iza­tion): compares just two treatments
 
choose a pair of subjects that are as closely matched as possible. Assign one of the treatments to each subject by random assign­ment.
Block design: group of experi­mental subjects that are known before an experiment to be similar in some way that is expected to affect the response of the treatments
 
random assignment of subjects to treatments is carried out separately within each block.
 
combines the idea of creating equivalent treatment groups.
 
another form of control. Some outside variables are controlled by bringing those variables into the experiment to form the blocks.
 

Chapter 5

placebo effect: dummy treatment with no active ingred­ients.
Randomized Compar­ative Experi­ment: one that compares two treatm­ents.
 
random assignment into groups, one group for each treatment
 
make sure to include one control group.

Chapter 7

The organi­zation that carries out the study must have an instit­utional review board that reviews all planned studies in advance in order to protect the subjects from possible harm.
 
purpose:to protect the rights and welfare of human subjects recruited to partic­ipate in research activi­ties.
All indivi­duals who are subjects in a study must give their informed consent before data is collected.
 
must be informed about the nature of the study and risk.
All individual data must be kept confid­ential. Only statis­tical summaries for groups of subjects may be made public.
Anonymity: subjects are anonymous - their names are not known even to the director of the study.
Confid­ent­iality:
 

Logic of Experi­mental Design

Random­ization produces group of subjects that should be similar in all respects before we apply the treatm­ents.
Compar­ative design ensures that influences other than the experi­mental treatments operate equally on all groups.
Therefore, differ­ences in the response variable must be due to the effects of treatments
Principles of Experi­mental Design
 
1. Control. The effects of lurking variables on the response, most simply by comparing two or more treatments
 
2. Randomize. Use impersonal chance to assign subjects to treatm­ents.
 
3. Use enough subjects in each group to reduce chance variation in the results.
Statis­tical Signif­icance: an observed effect of a size that would rarely occur by chance
Good studies are compar­ative even when they are not experi­ments.
We can often combine comparison with matching in creating a control group
 
note: matching does not entirely eliminate confou­nding
A good compar­ative study measures and adjusts for confou­nding variables.

Clinical Trials

Clinical Trials: experi­ments that study the effect­iveness of medical treatments on actual patients.
 
Randomized compar­ative experi­ments are the only way to see the true effects of these new treatm­ents. Without them, risky treatments that are no better than placebos will become common
 
Clinical trials produce great benefits, but most of these go to future patients. The trials pose risks which are borne by the subjects. Balance future benefits against risks
 
Both medical ethics and intern­ational human rights standards say that

Clinical Trials

Clinical Trials: experi­ments that study the effect­iveness of medical treatments on actual patients.
 
Randomized compar­ative experi­ments are the only way to see the true effects of these new treatm­ents. Without them, risky treatments that are no better than placebos will become common
 
Clinical trials produce great benefits, but most of these go to future patients. The trials pose risks which are borne by the subjects. Balance future benefits against risks
 
Both medical ethics and intern­ational human rights standards say that the interest of the subject must always prevail over the interests of science and society.
Behavioral and Social Science experi­ments
 
the direct risks to experi­mental subjects are less acute, but so are the possible benefits.
 
invasion of privacy, informed consent are both issues in these studies