Test 143  (Statistics) 1. In Student's t-test:: a. it is a parametric test b. critical ratio is independent of the degrees of freedom c. distribution is normal at infinite degrees of freedom d. is especially useful for multivariant analysis e. can be used to study the effect of an eye drop on      intraocular pressure   Answer a.T   b.F   c.T   d.F   e.T The larger the sample size, the closer the distribution to normal and for each number of degrees of freedom,  there is a completely critical ration. It is not suitable for multivariant analysis. 2. In a normal distribution: a. the mean is the same as the mode b. the mean is higher than the median c. 95% of observations lie within one standard deviation of the mean d. Mann-Whitney's test is suitable for analysis e. the coefficient of variation measures the spread of the values Answer a.T   b.F   c.F   d.F  e.T The mean, the mode and the median coincide. 95%  of observations lie within 1.96 standard deviation of the mean. Mann-Whitney's test is a non-parametric test. 3. With reference to clinical trials: a. the power of the trial equals type II error b. type I error describes the probability of rejecting a null hypothesis when it is in fact valid c. type I  error refers to the probability of false positive d. in cross over designs, patients act as their own controls e. the minimum number of patients required is 100 Answera.T  b.T  c.T  d.T  e.F Type I error describes the probability of accepting a null hypothesis when it is in fact invalid. The number of patients required for a trial depends on type I and II errors. 4. Paired t-test : a. applies to normal distribution b. used only on large number of patients c. is not suitable for samples less than 20 d. used for the analysis of quantitative data e. used for two independent samples Answer a.T   b.F  c.F  d.T  e.F It is suitable for small samples. 5. The standard error of the mean (SEM):  a. is the square root of the variance b. measures the spread of observations around the mean c. assess the reliability of the mean d. is always smaller than the standard deviation e. SEM is equal to standard deviation (SD) divided by      the number of samples (n)   Answer a.F   b.F  c.T   d.T  e.F Standard deviation is the square root of the variance and measures the spread of observations around the mean. As SEM = SD/square root of n and n is always greater than 1, SEM is always smaller than SD.