Memory Quiz Flashcards
p^
Sample statistic
p
Population parameter
p0
Parameter used in null hypothesis
P-value
Probability of test statistic or more extreme happening by natural variation
Z
Standardized score
Z*
Upper p critical value (right tail)
-Z*
Lower p critical value (left tail)
H0
Null hypothesis
HA
Alternate hypothesis
n
Sample size
N
Population size
np^
successes expected in sample
nq^
failures expected in sample
np0
successes observed
nq0
failures observed
Square root (p^q^/n)
Standard error of sample population
Square root (p0q0/n)
Standard deviation of parameter in 1propZtest
Z*square root(p^q^/n)
Margin of error for 1propZtest
p^(+/-)Z*square root(p^q^/n)
1PropZInterval
(p^-p0)/square root(p0q0/n)
Test statistic for Z distribution|Z value in 1PropZTest
How do you find outliers in a set of data?
Lower: Q1-1.5IQR
Upper: Q3+1.5IQR
IQR=Q3-Q1
What is the most important difference between an SRS and a stratified random sample?
SRS: every element of population has same probability of being drawn
Stratified random sample creates homogeneous groups, then takes SRS
How do you tell the difference between an observational study and an experiment?
Observational study observes only- no treatment
Experiment imposes a treatment on subjects
Formula to find the slope of the LSRL
slope^=r(Sy/Sx)
Template answer to describe the slope of the LSRL in context
For every 1 unit increase in x-variable, y-variable is predicted to increase/decrease according to slope
Template answer to describe the y-intercept of the LSRL in context
The predicted value of y (y^) when x-variable=0
Template answer to describe what R2 is
The % of the variability of the dependent variable (y^) that can be explained by the LSRL relating x and y^.
What are the Z-scores for 90%, 95%, 98%, and 99% confidence intervals?
90%: 1.645
95%: 1.96
98%: 2.326
99%: 2.576
What is a type I and type II error?
Type I: rejecting the null when it’s true
Type II: failing to reject the null when it’s false
What does the p-value mean in context?
The probability the observed statistic- or one more extreme- will happen by natural variation, if the null is true
What is meant by the power of the test?
The probability of correctly rejecting the null hypothesis when it’s false
What are the two most effective ways to increase the power of a test?
Increase sample size
Increase significance level (alpha)
Formulas for ALL confidence intervals
statistic(+/-)critical value(standard deviation of statistic)
Formulas to find the test statistic for ALL tests
(statistic-parameter)/standard deviation of statistic
What is ALWAYS the null hypothesis for a matched pairs t test?
Mdiff=0
What are the null hypotheses for the 3 different x2 tests?
GOF= H0: the sample data fit/match the population Independence= H0: the two variables are independent of one another Homogeneity= H0: there is no difference in group proportions