Title: Minitab Lab Session
1Minitab Lab Session 3w/Homework Problems
- t confidence interval for means
- t hypothesis test for mean (one sample)
- Two population hypotheses testing
- Independent samples
- Dependent samples
- Click here for the Minitab worksheet
2t confidence interval hypothesis test mean,
one population, ? unknown
- The following data show the number of the number
of tickets issued by the South Boro Police each
day for a period of a month. - The data is in column Tickets
- Example 1
- Find a 95 confidence interval for the true mean.
- Draw a boxplot that shows the confidence interval
- A motorist claims that the South Boro Police
issue an average of 55 speeding tickets per day.
Perform a hypothesis test to determine whether
there is enough evidence to support the motorists
claim with a .05.
3t confidence interval hypothesis test mean,
one population, ? unknown
- Homework Problem 1
- Jack works at a store that makes copies (e.g.
Kinkos) He has recorded the number of pages
copied by 50 randomly selected customers. - The data is in column StoreMgr
- Find a 90 confidence interval for the true mean.
- Interpret the interval.
- Provide a boxplot that includes the confidence
interval
4t hypothesis Testing mean, one population, ?
unknown, n lt 30
- Homework Problem 2
- From past experience, a teacher believes that the
average score on a real estate exam is 75. Data
from 20 randomly selected students who took the
exam recently appears below. Test the claim
that the mean score is still 75 with ? .05. - Assume scores are normally distributed.
- The data appears above and in Column Teacher
- Testing the claim means doing the steps in a
hypothesis test (e.g. identifying the population
parameter, stating the hypothesis, interpreting
the results). Minitab computes the test
statistic and the P-value for you.
5Comparing two means
- H0 ?1 ?2
- Ha ?1 gt ?2 or ?1 lt ?2 or ?1 ? ?2
- The testing of these hypotheses is similar to the
tests for a single population - The difference is in the computation of the test
statistic (t-value)
6Independent versus paired samples
- Independent samples
- Two samples are independent if the selection of
the individuals or objects that make up one
sample does not influence the selection of the
individuals or objects in the other sample - For independent samples, use the Two-Sample t
Test - Pooled variances when unknown but equal standard
deviations - Unpooled variances when unknown and unequal
standard deviations
7Independent versus paired samples
- Paired samples
- Two samples are paired when observations from the
first sample are paired in some meaningful way
with observations in the second sample - For paired samples, use the Paired t-test
8Two sample t-test two
- Example
- Has the PH level of our nations lakes increased?
Columns PHHist and PHCurnt are randomly
selected samples of the PH values in US lakes.
Perform a hypothesis test to determine whether
the data supports the claim that the PH level of
our nations lakes has increased. Test at the .05
significance level - Homework Problem 3
- A sports analyst wishes to determine whether gate
attendance has increased from the time the San
Francisco Giants baseball team moved to Oakland.
Gate attendance figures (thousands) from the
1958-1967 seasons, when the team was in San
Francisco, and for 1968-1978 when the team was in
Oakland appear in Columns 1958-67 and
1968-78. Perform a hypothesis test to
determine whether the data support a claim that
gate attendance has increase. Use a .10
significance level. - See note on Homework Problem 2.
9Paired t-test
- Example
- Are all discount mail-order companies the same?
Data were collected on the prices of the top ten
business software packages reported by PC
Magazine in August 1997. Two well-know
mail-order companies were sampled. The data are
in Columns Comp and PCC of todays data file. - Set up a hypothesis test to determine whether the
average price for software is different at the
two mail-order companies. Assume the populations
are approximately normally distributed and a .05
significance level.