1.
Using Studies Wisely
Two types of inference:
- inference about the population
- inference about cause and effect
Randomization helps you decide which type of inference is
appropriate!!
Random selection of individuals allows for inferences to be made
about the population.
Random assignment of individuals to groups allows for inferences
to be made about cause and effect.
2.
Using Studies Wisely
Two examples: what can or can’t we infer….???
• The U.S. Census Bureau carries out a monthly Current
Population Survey of about 60,000 households. Their goal is to
use data from these randomly selected households to determine
the percent of unemployed individuals in the population.
• Scientists performed an experiment that randomly assigned 21
volunteer subjects to one of two treatment groups: sleep
deprivation for one night or unrestricted sleep. The experiment
hoped to show that sleep deprivation causes a decrease in
performance two days later.
3.
Using Studies Wisely
Were individuals randomly
assigned to groups?
YES
NO
Inference about the population:
YES
Inference about cause and effect:
YES
Inference about the population:
YES
Inference about cause and effect:
NO
NO
Were individuals
randomly selected?
YES
Inference about the population:
NO
Inference about cause and effect:
YES
Inference about the population:
NO
Inference about cause and effect:
NO
4.
Using Studies Wisely
A small town dentist wants to know if a daily dose of 500 mg of
vitamin C will result in fewer canker sores in the mouth than taking no
vitamin C.
Design 1: Get all dentists in town with appointments in the next two
weeks to take part in a study. Give each patient a survey with two
questions: a) Do you take at least 500 mg of vitamin C each day? b)
Do you frequently have canker sores? Based on patients’ answers to
Question a, divide them into two groups, those who take at least 500
mg of vitamin C daily and those that don’t.
Suppose she compares the proportions of patients in each group who
complain of canker sores and she finds statistically significant
differences compared to those that take at least 500 mg of vitamin C.
What can the dentist conclude?????
5.
Using Studies Wisely
Design 2:
Get all dental patients in town with appointments in the next two
weeks to take part in a study. Randomly assign half of them to take
500 mg of vitamin C each day and the other half to abstain from taking
vitamin C each day for three months.
Suppose she compares the proportions of patients in each group who
complain of canker sores and she finds statistically significant
differences compared to those that take at least 500 mg of vitamin C.
What can the dentist conclude?????
6.
Using Studies Wisely
Design 3:
Select a random sample of dental patients in town and get them to
take part in a study. Divide the patients into two groups, those who
take at least 500 mg of vitamin C daily and those that don’t.
Suppose she compares the proportions of patients in each group who
complain of canker sores and she finds statistically significant
differences compared to those that take at least 500 mg of vitamin C.
What can the dentist conclude?????
7.
Using Studies Wisely
Design 4:
Select a random sample of dental patients in town and get them to
take part in a study. Randomly assign half of them to take 500 mg of
vitamin C each day and the other half to abstain from taking vitamin C
each day for three months.
Suppose she compares the proportions of patients in each group who
complain of canker sores and she finds statistically significant
differences compared to those that take at least 500 mg of vitamin C.
What can the dentist conclude?????
8.
Using Studies Wisely
A well-designed experiment tells us that changes in the
explanatory variable causes changes in the response variable.
The lack of realism can limit our ability to apply the
conclusions of an experiment to the settings of greatest
interest.
Brake Lights
9.
Using Studies Wisely
In some cases, it isn’t practical or ethical:
- Does texting while driving increase risk of accident?
- Does going to church help you live longer?
- Does smoking cause lung cancer?
So we can’t do an experiment on this idea, so now what…??
10.
Using Studies Wisely
Criteria for establishing causation when you can’t experiment
• Association is strong (smoking & cancer – very strong)
• Association is consistent (many countries)
• Larger values of the explanatory variable are associated
with stronger responses (smoke more or longer, more cancer)
• The alleged cause precedes the effect in time (30 years later)
• The alleged cause is plausible (experiments with animals)
Evidence for causation is overwhelming, BUT it is not as
strong as evidence by an experiment
A particular slide catching your eye?
Clipping is a handy way to collect important slides you want to go back to later.
Be the first to comment