---

1) Advantages of random sampling for studying the general human population include:

Choose 3 of the following options.

   A. it can be applied to any population.

   B. likely errors can be estimated.

   C. estimates obtained are not biased.

   D. it is easy to do.

   E. the sample can be referred to a known population.

---

2) In a case-control study to investigate oral contraception and breast cancer:

One answer only.

   A. cases would be women with breast cancer.

   B. controls would be given oral contraception.

   C. cases would be given oral contraception.

   D. women with breast cancer would be randomly allocated to be cases or controls.

   E. controls would be observed for several years to see how many developed breast cancer.

---

3) In a case-control study, patients with lung cancer had a highly significantly lower cholesterol level than did controls. This provides strong evidence that:

One answer only.

   A. low cholesterol causes lung cancer.

   B. there is evidence for a relationship between low cholesterol and lung cancer in the sampled population.

   C. low cholesterol is not related to lung cancer.

   D. low cholesterol and lung cancer always go together.

   E. low cholesterol is a risk factor for the development of lung cancer.

---

4) The paired t test is:

Choose 2 of the following options.

   A. impractical for large samples.

   B. equivalent to a chi-squared test.

   C. suitable for very small samples.

   D. used for independent samples.

   E. requires the assumption that differences between paired observations follow a Normal Distribution.

---

5) For a t test for two independent samples to be valid:

Choose 2 of the following options.

   A. the numbers of observations must be approximately the same in the two groups.

   B. the standard deviations of observations must be approximately the same in the two groups.

   C. the means must be approximately the same in the two groups.

   D. the observations must be from distributions which are approximately Normal.

   E. the sizes of samples the must be small.

---

6) A simple linear regression equation:

Choose 2 of the following options.

   A. always describes a line which goes through the origin.

   B. always describes a line with zero slope.

   C. is not affected by changes in the units in which variables are measured.

   D. describes a line which goes through the point defined by the means of the two variables.

   E. is affected by the choice of dependent variable.

---

7) The product-moment correlation coefficient between two variables, r:

Choose 2 of the following options.

   A. must lie between -1 and +1 inclusive.

   B. can only have a valid significance test carried out when one variable follows a Normal Distribution.

   C. is expected to be zero when there is no relationship between the variables.

   D. depends on which of the two variables is chosen to be the dependent variable.

   E. measures the magnitude of the change in one variable associated with a change in the other.

---

8) The standard chi-squared test for a 2x2 contingency table is not valid unless:

One answer only.

   A. all the expected frequencies are greater than five.

   B. both variables are continuous.

   C. at least one variable is from a Normal Distribution.

   D. all the observed frequencies are greater than five.

   E. the sample size is at least 100.

---

9) To analyse the relationship between blood pressure and height in a sample of students, we could use:

Choose 2 of the following options.

   A. paired t test

   B. correlation coefficient

   C. chi-squared test

   D. regression

   E. two sample t test

---

10) In a case-control study, 101 stroke patients were compared with 137 healthy controls. Among the results were: Cigarette smoking by stroke patients and healthy controls - Ever smoked? Cases: 71 yes, 30 never, 101 total; Controls: 36 yes, 101 never, 137 total; Chi-squared = 45.5, 1 degree of freedom, P less than 0.0001, Odds ratio = 6.6 (95% confidence interval = 3.8 to 11.8)

Choose 4 of the following options.

   A. These data would be unlikely if smoking and stroke were unrelated.

   B. It is estimated that the relative risk of stroke for ever-smokers compared to never-smokers is 6.6.

   C. In the population from which cases and controls come, the odds ratio is estimated to lie between 3.8 and 11.8.

   D. We can conclude that smoking causes stroke.

   E. There is good evidence that smokers have an increased risk of stroke.