Be able to
. define and distinguish between
samples and populations
. describe these probability sampling
designs: simple random sample,
stratified random sample,
multistage sample
. understand the problems or biases
involved in: voluntary response
sampling, undercoverage,
non-response and response bias due to
interviewer, respondent
or wording of question
. understand the difference between
obtaining the probability of an event from a known population
(section 4.1) AND
making a statistical inference from a sample statistic to a
population parameter (section 3.4)
. know the concepts of sampling
distributions and unbiased
statistics
. know the rule about the sample
size needing to be 1/10
population size
. be able to define: sample space,
event
. know 5 probability rules
. draw a Venn diagram with complements,
intersections and unions
. be able to
write out a table of probabilities for a discrete
distribution and calculate means, variances and standard
deviations from the table
. define random variable
. draw a probability histogram
. be able to calculate probabilities
under a discrete histogram or a
continuous distribution.
. know the law of large numbers,
the rules for combining means and
variances, and know the
general probability rules
. know how to add the probabilities
of many disjoint events
. know how to calculate conditional
probabilities of events which are not
independent and which are
independent
. know the multiplication rule
. learn the Bayes Rule and how
to apply it in a problem.
. know about sampling distributions
versus population distributions
. know the 4 criteria for a binomial
setting
. know what B(n,p) means; know
what N(mu,sigma) means
. know the mu and sigma parameters
for a distribution of counts and
for a distribution of proportions
. know how to apply the normal
approximation
. know when to apply the normal
approximation (np>10; n(1-p)>10)
. be able to calculate the binomial
probability of an outcome
combining simpler events.
. if the population is normal,
and you know the mean and std.
dev. for the random variable in the population, what is the
distribution of the sample means? What if it isn't
normal but n is large? (know central limit theorem)
. be able to calculate confidence
intervals, know how to find z* from a
table, given confidence
level
. be able to define confidence
intervals
. be conceptually familiar with
what is happening in a confidence interval
. be able to calculate sample size
for a margin of error
and margin of error for
a given sample size (knowing
confidence level and standard
deviation)
. be able to pick out null and
alternative hypotheses in an experimental
setting
. be conceptually familiar with
and able to define statistical significance,
. Be familiar with its uses and
abuses
. be able to conduct a z test for
a null hypothesis with one mean.
. be familiar with how to conduct
a two-sided test, and how
to use a confidence interval
to do a two-sided test
. be familiar with how to decide
when to reject a null hypothesis using a
confidence interval, comparing
p to alpha, or comparing the value of Z or t to
a critical value
. be familiar with type I errors,
type II errors and power.
. know how to increase power in
an experiment
. know that the difference between
z and t tests is in whether you know
sigma ahead of time or have
to use the sample standard deviation as an
estimate of the population
standard deviation
. be able to do a one-sample t-test
of a null hypothesis, know how to get
degrees of freedom from
sample size, how to use the table
. be able to do a one-sample confidence
interval for mu when you don't know
sigma