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;
49 Cards in this Set
- Front
- Back
Meaning of stratified
|
divide into groups with similar characteristics. choose a sample from group
|
|
Meaning of Cluster
|
dissimilar groups are already divided. choose entire group.
Ex) an airport. pick 5 flights and sample them. |
|
process of matched pair
|
Subjects -->paired--> treatment 1
-->treatment 2 |
|
Def & Process of block design
|
similar subjects formed then randomly assigned to treatment groups.
Subjects--> block (randomized) --> treatment 1 -->treatment 2 |
|
Mild Lower fence formula
|
Q1 - 1.5 (IQR)
|
|
Mild upper fence formula
|
Q3 + 1.5 (IQR)
|
|
Extreme lower fence formula
|
Q1 - 3 (IQR)
|
|
Extreme upper fence formula
|
Q3 + 3 (IQR)
|
|
Measure of center for symmetric distribution
|
Mean
|
|
Measure of spread for symmetric distribution
|
Standard Deviation
|
|
Measure of center for Skewed Distribution
|
Median
|
|
Measure of spread for skewed distribution
|
IQR
|
|
Mean > median
|
Skewed right
|
|
Mean < Median
|
Skewed left
|
|
Mean = median
|
Approximately symmetric
|
|
Interquartile Range
|
IQR = Q3 - Q1
|
|
When to use Z = (X-U)/ O
|
When its normal distribution
|
|
Correlation Coefficient (r)
Positive weak |
0 < r < .3
|
|
Correlation Coefficient (r
Negative weak |
-.3 < r < 0
|
|
Correlation Coefficient (r)
Positive, moderate |
.3 < r < .7
|
|
Correlation Coefficient (r)
Negative, moderate |
-.7 < r < -.3
|
|
Correlation Coefficient (r)
Positive Strong |
.7 < r < 1
|
|
Correlation Coefficient (r)
Negative Strong |
-1 < r < -.7
|
|
Residual formula
|
e = real (y) - Yhat (predicted)
|
|
Meaning of negative residual
|
Prediction is too high
|
|
Meaning of positive residual
|
Prediction is too low
|
|
When to use P(A U B)
|
to find the probability of both A AND B or both occuring
|
|
If A and B are disjoint (CANNOT occur at the same time) then P(A U B) formula is
|
P (A) + P (B)
|
|
When to use P (A n B)
|
to find the probability of both A AND B occurring (the overlap)
|
|
If A and B are independent then P (A n B) =
|
P(A) * P(B)
|
|
For independent events P(A U B) =
|
P(A) + P(B) - P(A) * P(B)
|
|
Properties of Binomial Distribution (4)
|
Fixed number of trials/observations, trials are independent, only two outcomes: success or failure, probability of success is the same for each trial
|
|
Properties of Approximate normal distribution (for NUMBER) (3)
|
Above properties for a binomial must be satisfied, NP and NQ > 10, SRS
|
|
Properties of approximate distribution for sample PROPORTIONS (4)
|
Sampled values are independent,NP and NQ > 10, SRS, sample < 10% of population
|
|
If it is number and NP and NQ is NOT greater than 10, then the sampling distribution is:
|
X ~ B( n,p)
|
|
If it is a number and NP and NQ is greater than 10, then the sampling distribution is:
|
X~AN (p, sqrt((pq)/n)
|
|
If it is a proportion and NP and NQ are satisfied then the sampling distribution is:
|
Phat ~ AN(U, O)
|
|
Normal distribution notation
|
X~N(U,O)
|
|
Empirical Rule
|
68-95-99.7
|
|
if it is normal approximation to binomial. how to find mean and standard deviation
|
Mean = np
Standard deviation= sqrt(np(1-p)/n) |
|
Margin of error formula
|
Z* (sqrt((pq)n))
|
|
If given C.I, find Point of estimate
|
(upper + lower) / 2
|
|
If given C.I, find ME
|
(upper - lower) /2
|
|
alpha =
|
1 - confidence level
|
|
Confidence interval =
|
1 - alpha
|
|
Type I error
|
mistakenly reject the null. Ho is TRUE but we mistakenly said it was FALSE.
|
|
Type II error
|
failed to reject the null. Ho is FALSE but we mistakenly said it is TRUE.
|
|
P (A n B) means
|
probability of A AND B
|
|
P ( A U B) means
|
probability of A OR B
|