Reading...
Play button
Play button
Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
14 Cards in this Set
- Front
- Back
|
frequency polygon (graph) with distinctive characteristics
|
Normal Distribution - symmetrical bell-shaped curve stat workhorse
|
|
|
The number of standard deviations a case is away from the mean of the distribution
|
Z score
|
|
|
The Standard Normal =?
|
normal distribution of Z scores
|
|
|
can we infer that trends/relationships in sample data can be generalized to a larger population?
|
Survey/Epidemiological Research
|
|
|
Survery/Epidemiological Research and experiments
|
Statistical Inference
|
|
|
Are the results sufficiently different from chance to conclude a true/effect
Are the results sufficiently different from chance to conclude a non-zero effect. |
Experiments in Statistical Inference
|
|
|
If Ho is true on the average the difference b/2 two sample means will be?
|
zero
|
|
|
the likelihood of the sample data coming from the sampling distribution assuming Ho is true
|
Level of Significance (alpha)
|
|
|
the probability or likelihood below which we will say not very likely and reject the null
|
Level of Significance - in general set at .05 or 5%
|
|
|
When probability of Type 1 error decreases probability of type 2 error?
|
increases
|
|
|
risk of rejecting Ho when Ho is actually true
|
Type 1 Error - False Positive
|
|
|
Risk of failing to reject Ho when Ho is actually False
|
Type 2 Error - False Negative
|
|
|
1-beta
the probability of detecting a difference b/t treatments in a study, when a actually exists. |
Power - Opposite of type 2 error
|
|
|
Desired lvl of power
|
= or > 80 - increase by increasing sample size.
|
