Ew Flashcards
Explanatory variable
The cause of an experiment
Ex: vision
Census
Taking a poll of the entire population
Response variable
The effect of an experiment
Ex: a students grade
Confounded variable
An alternate explanatory variable
Ex: studying
Control group
The group that gets the placebo to compare to the other group
Factor
The explanatory variables in an experiment
Ex: diet, exercise, genetics
Levels
Different values of the exercise
Ex: 2000 calories, 4000 calories
Treatment
All combinations of the levels of the factors
Block design
Separate the individuals in similar blocks but each block is different
Voluntary response bias
People choose to respond to a survey
Convenience sampling
Choose individuals easiest to reach
Undercoverage
When some groups in the population are left out in the process of choosing a sample
Nonresponse bias
When an individual chosen for the sample cannot be contacted or refuses to cooperate
Response bias
Behavior of the respondent or interviewer that results in no truthful answers
Stratified random sample
Divide the population into groups of similar individuals, then choose an Drs of each group and combine them
Simple random sample(srs)
Every set of individuals has the same chance of being selected
Systematic sample
Choose every nth person from the population
Size bias
Dart on a map, it’s more likely to hit the larger states
Double blind experiment
The subjects nor the administrators of the treatment know what they are receiving
Observational study
Does not disturb what you are examining
Experiment
Treatment imposed, cause and effect
What good experiments do
- incorporate control to see the impact of the explanatory variable
- incorporate randomization to reduce bias
- incorporate replication to avoid the chance of variance
Degrees of freedom
N-1
Quantities variables
Numbers
Categorical variables
Words (use pie and bar graphs)
Interpreting r
There is a fairly strong positive relationship between x and y
Bimodal(approximately)
Two high bars
Interpret slope
On average, y increases by b for every increase in 1x
Interpret y-intercept
When x=0 y is predicted to equal ____
Linear model equation
Ÿ=a+bx
Exponential equation
Log ÿ= a+bx
Power model equation
Log ÿ=a+b(log x) or
Ÿ=10^a • x^b
Sample space
All possible outcomes
Binomial distributions
- each part has the same probability
- they are all independent from each other
- only 2 outcomes, success or failure
- fixed # of observations
R^2
The fraction of the variation in values of y that is explained by the least squares regression
Extrapolation
When LRSL is used to make a prediction outside of the domain of x
Common response
A change in z cause a Change in x and y
Causation
Cause and effect, x effects y
Confounding
X and z influence y
Simpson’s paradox
The reversal in trends when data from several groups is combined to form a single group
Gamblers fallacy
The belief that if something happens more frequently than normal it will happen less in the future
Fair game
The outcome equals the money put in
Law of large #s
As the # of observations increases the sample mean gets closer to the population mean
Multiplication principle
If task 1 can be done in a ways, and task 2 can be done in b ways, then both tasks can be done in a•b ways
Mode
that occurs most often
Central limit theorem
When n is large the sampling distribution of x is approximately normal
Parameter
Population
Statistic
Sample
Unbias
Centralized around the true mean
