1) A normal population distribution is needed for the following statistical tests:

a. chi-squared

b. variance estimation

c. standard error of the mean

d. Student's t-test

e. Kendall's rank coefficient.

2) With regard to statistics:

a. the mode is the most frequently occurring value

b. the mode is the maximum point on a frequency distribution curve.

c. the median has an equal number of values above and below it

d. in a normal distribution the mean, mode and median coincide

e. about 93.5% of observations lie within two standard deviations of the mean

3) With regard to the chi-squared test:

a. it is used to test the difference between frequencies

b. it is used as an alternative to the t-test to determine the difference between two means

c. the number of degrees of freedom is the number of independent comparisons.

d. the greater the value of the chi-squared test, the less likely it is to be significant

e. the null hypothesis is not required.

4) The prevalence rate of a disease has the following features:

a. it is dependent on the incidence of the disease

b. it is dependent on the duration of illness

c. it measures all the current cases in the community

d. it can be estimated from a cross-sectional study

e. it can not be used to determine the health needs of a community

5) An aetiological factor must satisfy the following before one can say that it is casually related to a disease:

a. exposure to the factor must precede the development of the disease

b. elimination of the factor decreases the risk of the disease

c. the factor is found more frequently among the diseased than non-diseased

d. the factor is not found among persons without the disease

e. the factor is found in all cases with the disease

6) A 95 % confidence interval:

a. can only be used in parametric data

b. it is a test of the null hypothesis

c. it is calculated at ± 1.96 times the standard error of the mean

d. is useful when comparing data with another population

e. if zero difference lies within the 95% when comparing two groups to a treatment, it indicates the treatment has no effect

7) Meta-analysis of randomized controlled trials:

a. should only include peer-reviewed studies

b. is usually performed if individual trials are too small to give a reliable answers

c. should include trials in which patient selection is not randomized

d. provides a more stable estimate of the effect of treatment than individual trials

e. gives conclusion making further controlled trials unnecessary

8) The following are true about evidence-based medicine:

a. does not involve health economic assessment

b. is restricted to randomized placebo-controlled trials

c. is used to cut down waiting list

d. relies only on objective measurements of disease outcomes

e. combines clinical expertise and external evidence

9) The following are true about errors in clinical trials:

a. type I error is wrongly rejecting the null hypothesis

b. type II error is accepting the null hypothesis when it is invalid

c. errors are more common when big samples are used

d. type I error is more likely to occur when multiple t-test are used

e. type II error is reduced by the use of confidence interval.

10) The following are true about a normal (Gaussian) distribution except:

a. the mean, median and mode coincide

b. 95% of observation lie between the mean 2 standard deviation

c. 68% of observations lie between the mean 1 standard deviation

d. the 95% confidence interval may be calculated as the mean ±1.96 times the standard error of the mean for population >29

e. data from a normal distribution is suitable for parametric tests without prior transformation.