Stat 121 - Final Exam
A list of all possible values of a variable together
Definition of Distribution
with the frequency (or probability) of each value.
A measure of the "average" or typical deviation of
Standard of Deviation
the observations about the mean; measures
variability of data about the mean
- matched pairs t for means
Type of Procedures for C-Q - Two-sample t for means
data - Analysis of variance (Anova)
-Two-sample z for proportions
Types of Procedures used for
C-C data - Chi-square test
Types of procedure used for Regression Inference
Q-Q data
-one-sample z for proportions
One Sample procedures
-one-sample t for means
Hypothesis for one-sample z Ho: u=uo(mu-
procedure for means not) u>,<, or
= uo
Randomization (SRS)
Conditions to be met for one Normality of sampling distribution of x-bar (as
sample z procedure stated in problem, or sampling distribution appears
normal) or CLT n >30
Sigma known
Ho: u=uo(mu-
1 sample t for means
not) u>,<, or
= uo
Randomization (SRS)
Normality of population as stated in the problem or
Conditions for 1 sample t for
means no outliers or extreme skewness in plot of data
df = n-1
, Hypothesi
Matched pairs t for mean of s: Ud=0
differences Ud >,<, not = 0
Randomization (SRS)
Normality of population as stated in the problem or
Conditions for Matched no outliers or extreme skewness in plot of data
pairs t for mean of df = n-1
differences Uo = 0 in formula
Use mean of differences
Two Randomization (SRS) or RAT
Normality of population as stated in the problem or
2 sample t for difference of no outliers or extreme skewness in plots of data
means
Equal population standard deviations, stated in
problem or largest s/smallest s <2 df = n1+n2 - 2
Hypothesis for 2 sample t Ho: u1-u2 = 0; u1=u2
for difference of means Ha: u1-u2 does not = 0; u1 <,>, or does not =u2
Ho: u1 = u2 = u3...
ANOVA (Analysis of
variance) hypothesis Ha: at least one u is different (not all u's are the same)
Randomization: Multiple (SRS) or RAT
Normality of population as stated in the problem or
Conditions to be met for no outliers or extreme skewness in plots of data
ANOVA
Equal population standard deviations, stated in
problem or largest s/smallest s <2 Output - F and p-
value
Ho = Beta = 0 (there is no linear relationship between the two
Linear Regression Hypothesis
variables)
Ha = Beta <,>, or not equal to 0 (there is a linear relationship
between two variables)
Linearity: residual plot (no
smiley faces)
Independence: SRS or RAT
Conditions to be met for
Linear Regression Normality: histogram or stemplot of residuals
appear normal (no outliers and bellshaped)
Equal Std Dev. Variance (residual
plot - no megaphone) df = n-2
Ho: p = po(p-not)
1 sample Z for proportions
Ha: p <, >, or not equal to p-not
Randomization: SRS or RAT
Normality of sampling
Conditions for 1 sample Z for
proportions distribution of p-hat Np
checks - np0 > 10
n(1-p0) > 10
Ho: p1 - p2 = 0, or p1 = p2
2 sample z for proportions
Ha: p1 - p2 = <, >, or does not equal 0, or p1 <,>, or not = p2
A list of all possible values of a variable together
Definition of Distribution
with the frequency (or probability) of each value.
A measure of the "average" or typical deviation of
Standard of Deviation
the observations about the mean; measures
variability of data about the mean
- matched pairs t for means
Type of Procedures for C-Q - Two-sample t for means
data - Analysis of variance (Anova)
-Two-sample z for proportions
Types of Procedures used for
C-C data - Chi-square test
Types of procedure used for Regression Inference
Q-Q data
-one-sample z for proportions
One Sample procedures
-one-sample t for means
Hypothesis for one-sample z Ho: u=uo(mu-
procedure for means not) u>,<, or
= uo
Randomization (SRS)
Conditions to be met for one Normality of sampling distribution of x-bar (as
sample z procedure stated in problem, or sampling distribution appears
normal) or CLT n >30
Sigma known
Ho: u=uo(mu-
1 sample t for means
not) u>,<, or
= uo
Randomization (SRS)
Normality of population as stated in the problem or
Conditions for 1 sample t for
means no outliers or extreme skewness in plot of data
df = n-1
, Hypothesi
Matched pairs t for mean of s: Ud=0
differences Ud >,<, not = 0
Randomization (SRS)
Normality of population as stated in the problem or
Conditions for Matched no outliers or extreme skewness in plot of data
pairs t for mean of df = n-1
differences Uo = 0 in formula
Use mean of differences
Two Randomization (SRS) or RAT
Normality of population as stated in the problem or
2 sample t for difference of no outliers or extreme skewness in plots of data
means
Equal population standard deviations, stated in
problem or largest s/smallest s <2 df = n1+n2 - 2
Hypothesis for 2 sample t Ho: u1-u2 = 0; u1=u2
for difference of means Ha: u1-u2 does not = 0; u1 <,>, or does not =u2
Ho: u1 = u2 = u3...
ANOVA (Analysis of
variance) hypothesis Ha: at least one u is different (not all u's are the same)
Randomization: Multiple (SRS) or RAT
Normality of population as stated in the problem or
Conditions to be met for no outliers or extreme skewness in plots of data
ANOVA
Equal population standard deviations, stated in
problem or largest s/smallest s <2 Output - F and p-
value
Ho = Beta = 0 (there is no linear relationship between the two
Linear Regression Hypothesis
variables)
Ha = Beta <,>, or not equal to 0 (there is a linear relationship
between two variables)
Linearity: residual plot (no
smiley faces)
Independence: SRS or RAT
Conditions to be met for
Linear Regression Normality: histogram or stemplot of residuals
appear normal (no outliers and bellshaped)
Equal Std Dev. Variance (residual
plot - no megaphone) df = n-2
Ho: p = po(p-not)
1 sample Z for proportions
Ha: p <, >, or not equal to p-not
Randomization: SRS or RAT
Normality of sampling
Conditions for 1 sample Z for
proportions distribution of p-hat Np
checks - np0 > 10
n(1-p0) > 10
Ho: p1 - p2 = 0, or p1 = p2
2 sample z for proportions
Ha: p1 - p2 = <, >, or does not equal 0, or p1 <,>, or not = p2
