For all hypothesis tests, write out 6 steps, including a p-value statement.
For all confidence intervals, justify the distribution used, calculate the interval, and write out a confidence interval statement.
1.
7.10 Do the calculations in a), b), c), d) and e) by calculator and
by SPSS.
Circle
the answers for
a), b), c), d)
and e) on the SPSS output.
In addition, do a boxplot of the difference scores (in lbs) in SPSS.
Does the data look like it could have come from a normal distribution?
For e), use α = .01
and do a two-tailed test. Do d) either by calculator or SPSS.
2. Using the data in 7.84 and α = .05,
a)
compare men and women using SPSS to construct side-by-side boxplots. Is
the use of the pooled t procedure appropriate for this data?
b)
Most studies have found that the mean SSHA score for men is lower than
the mean score in a comparable group of women. Test this supposition here.
Do by hand (use the pooled technique) and by SPSS. Circle the test statistic
and correct p-value on SPSS output.
c)
Construct a 90% confidence interval for the mean difference between the SSHA scores of male and female first-year students at this college. Do
by hand and by SPSS. Circle the confidence interval on the SPSS output.
3.
8.22 Remember to do the 6 steps by hand. Remember assumptions/reasons
for distribution used in hypothesis test and confidence interval.
c)
What sample size would you take if you wanted to estimate the proportion
who preferred freshly brewed coffee to within .05 (5%)? First use
sample proportion as the best estimate. Then redo assuming no estimate
of p is known.
********************** The following kind of problem will be on the exam, so do it on your own and check answers from HW answer sheet **************************.
4. 8.43, 8.44 Do both problems by hand, and also find the test statistic and p-value by SPSS. Remember assumptions/reasons for distribution used in hypothesis test and confidence interval. Circle the test statistic and p-value on the SPSS output.