Tests/Conditions/test statistic Flashcards
(11 cards)
one-proportion z-test
Independence, Randomization, 10%, Success/Failure Condition
SD(p)=√(pq/n)
Statistic: z=(phat-p)/SD(p)
ME=z*√(phat x qhat/n)
Power
The probability that a hypothesis test will correctly reject a false null hypothesis.
Power increases with sample size, alpha level, the difference between the statistic and hypothesized parameter in the opposite direction, and decreasing the standard error. Equal to 1-beta.
Type I vs Type II
Type I: False positive; rejecting a true null hypothesis. Probability is alpha.
Type II: False negative;
failing to reject a false null hypothesis. Probability is beta
Two-proportion z-test
Independent Groups Assumption, Independence Assumption, Randomization, 10% Condition, Success/Failures
(two-proportion Z-interval) = SE(p1hat-p2hat) =√((p1hat x q1hat)/n1 + (p2hat x q2hat)/n2)
SEpooled(p1hat-p2hat) = √(phatpooledqhatpooled)/n1+(phatpooledqhatpooled)/n2)
z = (p1hat-p2hat)/SEpooled(p1hat-p2hat)
One-sample t-test
µ
Randomization
10% Condition
Nearly Normal
_
SE(y)=s/√n
_
t = y -y0/SE(y)
Two-sample t-test
µ1-µ2
µd = difference
Independence
Randomization
10%
Nearly Normal
t=(y1-y2)-(µ1-µ2)/SE(y1-y2)
SE(y1-y2) = sqrt((S1)^2/n1)+(S2)^2/n2)
Matched pairs t-test
µ
Paired Data Condition
Independence
Randomization
Nearly Normal Condition
_ _
t = d-0/SE(d)
_
SE(d) = Sd/√n
Chi-square test for goodness of fit
Counted Data Condition
Independence
Randomization
Expected Cell Frequency Condition
x^2 = Σ(Obs-Exp)^2/Exp
Chi-square test for homogeneity
Counted Data Condition
Independence
Randomization
Expected Cell Frequency Condition
x^2 = Σ(Obs-Exp)^2/Exp
Chi-square test for independence
Counted Data Condition
Independence
Randomization
10% Condition
Expected Cell Frequency Condition
x^2 = Σ(Obs-Exp)^2/Exp
Regression Slope t-test
β for hypotheses
Straight Enough Condition
Independence
Does the plot thicken? Condition
Nearly Normal Condition
