A t-test for a mean uses a sample of 15 observations. Find the $t$ test statistic value that has a p-value
Problem 1: A t-test for a mean uses a sample of 15 observations. Find the $t$ test statistic value that has a p-value of $0.05$ when the alternative hypothesis is
- mu not equal to 0
- mu greater than 0
Problem 2: A study has a sample of 20 subjects. The test statistic for testing the null hyp., \(\mathrm{mu}=100\) is \(\mathrm{t}=2.40\). Find the \(\mathrm{p}\) -value for the alternative hyp.
- mu not equal to 100
- mu less than 100.
Problem 3: In an anorexia study, 29 girls received cognitive therapy. The resulting weight changes had a mean of \(3 \mathrm{lbs}\) and a standard deviation of \(7.32 \mathrm{lbs}\). Test the null hypothesis that the therapy has no effect on the weight of girls.
Problem 4: A study of 13 children with asthma compared single inhaled doses of formoterol (F) and salbutamol (S). The outcome measured was the child's peak expiratory flow (PEF) eight hours after treatment.
Let mu denote the population mean of the difference between the PEF values for the F and \(S\) treatments. Perform a significance test for the null hyp. that \(m u=0\) against the two-sided alternative. Give a \(95 \%\) confidence interval for mu.
Problem 5: Do question 4 again, this time using SPSS
Problem 6: Proportions. For a test of null: \(\mathrm{p}=0.5\), the \(\mathrm{z}\) -test stat. is 1.04. Find the \(\mathrm{p}\) -value for the alt. hyp. that:
- \(p>0.5\)
- \(\mathrm{p}\) is not equal to 0.5
Problem 7: For a test of null: \(\mathrm{p}=0.5\), the sample proportion is 0.35 based upon a sample size of 100.
- Find the value of the test statistic
- Find the \(\mathrm{p}\) -value for the alt. hyp. \(\mathrm{p}<0.5\)
Problem 8: In a crossover study, 30 children were given Sumatriptan and a placebo when suffering with separate migraines. Of the 30 children, 22 had more pain relief with the drug and 8 with the placebo.
- Test the null hyp. that \(\mathrm{p}=0.5\) against the two-sided alternative.
- What is a crossover study?
- What are the assumptions for this test to be legitimate?
Problem 9: In a study to assess the effect of a drug on PMS symptoms, 7 out of 10 women reported success.
- Find a \(95 \%\) confidence interval for the population proportion
- Is it plausible that it is successful for only half of the population?
Problem 10: Suppose that \(19 \%\) of workers at Jewish base abstain from alcohol. To estimate the proportion at Jewish Kenwood, how large a sample would you need to estimate the proportion within 0.05 with probability 0.95
- If you use the info. from Jewish base as a guideline.
- If you are unwilling to predict a proportion
Problem 11: Do women do more housework?
Find a \(95 \%\) confidence interval comparing the p
What does it mean if 0 is not in the interval? Note, for large samples such as this, \(t\) scores can be replaced by z scores.
Problem 12: Teenagers were assessed for nicotine dependence. The mean "Hooked on Nicotine
Checklist" score for 150 females was 2.8 (s =3.6). The score for 182 males was 1.6 (s=2.9)
Find the test statistic and P-value for the null hyp. mul=mu2 against the two sided alt.
Would you reject the null hyp. at the 0.01 level?
P roblem 13: What is a matched-pairs study? In practice, is the analysis of means from matched-pairs data a one-sample or two-sample test?
Problem 14: Compare 2 means with an example of your choice in SPSS
Deliverable: Word Document
