Med Stats Exam 1

The Pearson correlation coefficient is not the correct statistic to use in assessing the agreement between multiple measurements of the same continuous variable because: • A. Correlations between the same measurements are always high • B. Correlations do not capture measurement error • C. High correlations does not imply agreement • D. Correlations do not assess bias

C

If the non-parametric test carried out in 9. Above is significant, which test would be best for assessing the pair-wise differences? • A. Wilcoxon rank sum test • B. Wilcoxon signed rank test • C. Kruskall-Wallis test • D. F test

A

If the null hypothesis is rejected, this means that: • A. Not all means are equal • B. Not all variances are equal • C. All means are equal • D. All variances are equal

A

If you compute an odds ratio for the results in a clinical trial with two treatments, you would be computing the: • A. Odds of developing the outcome in one treatment versus the other • B. Odds of previous exposure in one treatment versus the other • C. Odds of a greater percent reduction in one treatment versus the other. • D. Odds of a greater risk in one treatment versus the other

A

In a box and whisker plot, what is the percentage of the data within the "box?" • A. 50% • B. 75% • C. 25% • D. 100%

A

In the Coronary Artery Surgery Study (CASS), patients with coronary artery disease were randomized to receive either surgical or medical care; no patient was left untreated or given a placebo. They were followed to determine several outcomes. The most distinguishing factor in this study that differs from observational studies is: • A. An intervention is used • B. Two treatment groups were used rather than a treated group and a control group • C. Patients are randomly allocated to groups • D. Patients could not be blinded as to what treatment they receive

A

On the basis of only one randomly drawn sample from a normal distribution, we can say that there is a 95% chance that the true mean of the normal distribution of sample means will be contained within the interval __________. Question 16 Options.pdf 21 KB • A. See formula choices in "Question 16 Options." • B. See formula choices in "Question 16 Options." • C. See formula choices in "Question 16 Options." • D. See formula choices in "Question 16 Options."

A

Statistical significance tests or tests of hypothesis have been devised to assess the compatibility of a set of data with a null hypothesis H0 of no treatment differences. The relationship between confidence intervals and hypothesis tests for two means can be described as: • A. The confidence interval is the "acceptance" region (or failure-to-reject region) for testing H0:mu=mu0 in a two sided hypothesis test at alpha=0.05 level of significance • B. The confidence interval is the rejection region for testing H0:mu=mu0 in a two sided hypothesis test at alpha=0.05 level of significance. • C. A value mu0 lying within the 1-alpha confidence interval corresponds to a value of the test statistic z= (x-mu0)/ sigma/sqrt(n) that would lead to rejection of the null hypothesis. • D. A value mu0 lying outside the 1-alpha confidence interval corresponds to a value of the test statistic z= (x-mu0)/ sigma/sqrt(n) that would lead to failure to reject the null hypothesis.

A

Suppose the frequency of the values of serum triglyceride concentration (mg/100 ml) in men have the shape as shown by a smooth curve that is heavily skewed to the right. And suppose this curve represents values from a huge number of men. Suppose repeated samples of size 50 are drawn from the population of triglyceride values, the means of the samples are computed and those means are plotted on a graph. What shape would the graph of the means have? • A. A normal distribution • B. t distribution • C. An F distribution • D. A chi-square distribution

A

The "degrees of freedom" for the numerator of the ratio used in ANOVA: • A. k-1 • B. N-k • C. N-1 • D. n1 + n2 -1

A

The best measurement of risk of a disease or condition in the case-control study is the: • A. Odds Ratio • B. Relative Risk • C. Hazard Ratio • D. Difference in Proportions

A

The central limit theorem is concerned with the variation among • A. Sample means • B. Individual observations • C. Sample variances • D. Populations means

A

The goal of inference is to: • A. Generalize sample results to populations • B. Take samples • C. Randomize subjects • D. Generalize population results to samples

A

The p-value for a statistical test is: • A. The probability of obtaining a result as extreme as or more extreme than the one observed, if the null hypothesis is true. • B. The probability of accepting the null hypothesis when it is actually false. • C. The probability of incorrectly rejecting the null hypothesis when it is actually true. • D. The same as the critical value for the test

A

The primary disadvantage of a study in which patients are divided into environmental exposure groups and followed for evidence of carcinomas is • A. Causation cannot be proved because of no intervention and other events affecting outcome. • B. Many biases cannot be controlled. • C. It does not have the correct time-sequence to infer risk. • D. The course of the disease cannot be followed.

A

To determine the chance of successes in independent trials, we use the: • A. Binominal Distribution • B. Normal Distribution • C. F-Distribution • D. t-Distribution

A

Which measure is used to compare the relative variability in two different variables such as heart rate and systolic blood pressure? • A. Coefficient of variability • B. Variances • C. Coefficients of determination • D. Correlations

A

Which of the following factors is the most important to take into account after computing the positive predictive value for a specific disease? • A. Prevalence of the disease • B. Probability of the disease • C. Incidence of the disease • D. Sensitivity of the test

A

Which of the following is NOT used to determine whether a data distribution is "normal"? • A. Carrying out a non-parametric test • B. Fitting a curve to a histogram of the data • C. Carrying out a goodness of fit test • D. Looking at a stem-and-leaf plot

A

Which of the following is not one of the assumptions made in the analysis of variance? • A. Repeated observations for the same subject are normally distributed within each group • B. The dependent variable is normally distributed in each group • C. The population variance is the same in each group • D. Observations are independent

A

Which of the following is used to assess the spread in the data for a continuous variable? • A. Standard Deviations/Variances • B. Correlations • C. Percentiles • D. Coefficients of variation

A

The Wilcoxon signed rank test is used to: • A. Compare means in paired data • B. Compare medians in two groups • C. Compare medians in paired data • D. Compare variances in two groups

C

A 95% confidence interval for a mean indicates that: A.The sample mean is the correct estimate of the population mean with 95% confidence. B.If repeated samples were taken and the 95% confidence interval was computed for each sample, 95% of the intervals would contain the population mean. C.All samples taken from the same group will fall into this interval with 95% confidence D. A 95% confidence interval has a 0.95 probability of containing the population mean.

B

A study is conducted to examine the relationship between histology slides and magnetic resonance imaging to study characteristics of diseased carotid arteries. The goal is to assess the level of agreement between the two measurements. This design of this study is: • A. Prospective cohort • B. Cross-sectional • C. Cross-over • D. Case-control

B

Five different training programs were compared for assessing maximum oxygen consumption during a certain period of running on a treadmill. There were 10 participants in each group. An analysis of variance was used to compare the five training programs and the p-value that resulted from the test was 0.018. At a significance level of 0.05, the appropriate conclusion about mean maximum oxygen consumptions is that it • A. Is the same for the five training programs • B. Differs for at least two (one pair) of the five training programs • C. Is different for each of the five training programs • D. Is significantly better for one of the training programs than for the other four.

B

For two measures, X and Y, a correlation value of -0.80 was obtained. What is the interpretation of this correlation? • A. As X increases, Y increases by 80% • B. As X increases, Y decreases • C. As Y increases by 80%, X increases • D. As X decreases by one unit, Y decreases

B

Histograms are produced by JMP and other statistical software packages. What is the most important reason to examine histograms before analyzing a continuous variable? • A. To determine the median of the data • B. To examine the distribution of the data • C. To check for the widest interval that contains the data • D. To decide which cutpoints to use in making categories of the data

B

If the distribution of the data is normal or "bell-shaped," what is the proportion of the data that falls between two standard deviations on either side of the mean? • A. 99% • B. 95% • C. 75% • D. 68%

B

In a clinical trial, the way to ensure that two treatment groups are as similar as possible is to • A. Assign patients in sequence first to one group and then to the other as they enter the study • B. Randomize the patients to the two groups • C. Create two groups according to severity of disease and select one from each group for the treatments in sequence as patients enter the study • D. Administer the two treatments in exactly the same fashion

B

In computing the confidence interval for the difference in two means and for carrying the hypothesis of the difference in two means, we pool the standard deviations from the two groups if: • A. The two standard deviations are known • B. The two variances (standard deviations) are not unequal • C. The two means are equal • D. Each of the two standard deviations is not zero

B

In statistical inference, we use sample values to estimate population values. The sample estimates are called: • A. Random variables • B. Statistics • C. Parameters • D. Probabilities

B

Paired data analysis for means is appropriate when the question is: • A. Is the difference between means equal to zero? • B. Is the average difference equal to zero? • C. Is the confidence interval for the mean in one group of subjects different from the confidence interval for the mean in another group of matched subjects? • D. Can the difference between means be attributed to pairing?

B

Suppose that 120 people with high cholesterol are randomly divided into four groups of 30 and a different treatment (drug-diet combinations) for decreasing cholesterol is assigned to each group. The response variable is the change in cholesterol level after three months of treatment. An analysis of variance will be used to compare the four treatments. What null hypothesis is tested by this test? (Hint: remember the meaning of "inference.") • A. The sample variances are equal for the four treatment groups • B. The population means (µ1=µ2=µ3=µ4) are equal for the four treatment groups • C. The population variances are equal for the four treatment groups • D. The sample means are equal for the four treatment groups

B

The Wilcoxon rank sum test is used to: • A. Compare means in paired data • B. Compare medians in two groups • C. Compare variances in two groups • D. Compare medians in paired data

B

The confidence interval for comparing means in paired data is concerned with: • A. Determining whether the difference in pre- and post-means are significantly different from zero • B. Determining whether the average difference is significantly different from zero • C. Determining whether the difference between the means of two matched conditions are significantly different from zero • D. Determining whether the confidence interval for the difference between two means contains zero

B

The relationship of an attribute to the disease is examined by comparing the diseased and non-diseased with regard to how frequently the attribute was present or, if quantitative, the levels of the attribute in each of the groups. This statement characterizes which type of study? • A. Clinical trial • B. Case-control • C. Historical cohort • D. Retrospective Cohort

B

The test used in ANOVA is a ratio of: • A. Means of groups to overall mean • B. Variances among means to the variance within each group • C. Variance within each group to the variance among means • D. Overall mean to means of groups

B

The variance of the distribution of sample means will decrease as the ___________ increases. • A. Mean • B. Sample size • C. Population size • D. Standard deviation

B

To determine the probability of being between two values of a continuous variable with a frequency distribution that resembles a "bell-shaped" shaped curve, we use the: • A. F-Distribution • B. Normal Distribution • C. t-Distribution • D. Binominal Distribution

B

Two studies are carried out to determine the effect of low levels of wine consumption on cholesterol level. The first study measures the cholesterol levels of 100 volunteers who have not consumed alcohol in the past year and compares these values with their cholesterol levels after 1 year, during which time each volunteer drinks one glass of wine daily. The second study measures the cholesterol levels of 100 volunteers who have not consumed alcohol in the past year, randomly picks half the group to drink one glass of wine daily for a year, while the others drink no alcohol for the year. Serum cholesterol levels are measured again in both groups after one year. Which of the following is a true statement? • A. The first study is an observational study, while the second is an experiment. • B. Both studies are experiments. • C. The first study is an experiment, while the second is an observational study. • D. Both studies are observational studies, but only one uses randomization and a control group.

B

When comparing means, the t distribution is used instead of the z distribution when: • A. When the sample size is very small • B. The population standard deviation is not known • C. The population mean is not known • D. The population standard deviation is known

B

When comparing three or more means, using ANOVA, the distribution used to test the significance of the difference among groups is the: • A. Z (normal) distribution • B. F distribution • C. t-distribution • D. Binomial distribution

B

Which is the most appropriate numerical summary to use in examining the relationship between two continuous (numerical) variables? • A. Relative risk • B. Correlation coefficient • C. Percentage difference • D. Coefficient of variation

B

The "degrees of freedom" for the denominator of the ratio used in ANOVA: • A. N-1 • B. k-1 • C. N-k • D. n1 + n2 -1

C

Having Blood type A and having blood type AB is an example of which type of event: • A. Non-mutually exclusive • B. Independent • C. Non-independent • D. Mutually exclusive

D

After entering data into JMP, the most important next step to ensure that analysis is carried out correctly is to: • A. Name each variable appropriately • B. Enter the value labels for coded data • C. Assign the data type and modeling type • D. Make sure that there are no missing values

C

Eighty percent of the variability in systolic blood pressure can be explained by age. What measure is computed to obtain this value? • A. Correlation coefficient • B. Risk coefficient • C. Coefficient of determination • D. Coefficient of variation

C

If A is an event and B is a mutually exclusive and complementary event, then the sum of the probability of A and the probability of B is: • A. 0.75 • B. 0.25 • C. 1.00 • D. 0.50

C

If A={serum cholesterol 250-299} and B={serum cholesterol > 275} then the events A and B are: • A. Mutually exclusive • B. Independent • C. Non-mutually exclusive • D. Non-independent

C

If the null hypothesis is rejected, the next logical step would be to: • A. Compare all means with the overall mean • B. Compare variances in pair-wise fashion • C. Test differences between pairs of means • D. Compare medians in paired data

C

In which of the following ways are numerical variables summarized descriptively? • A. Percentiles • B. Correlations • C. Means and standard deviations • D. Frequencies and percentages

C

Independent sample t-tests are used to: • A. Compare medians in two groups • B. Compare variances in two groups • C. Compare the means in two groups • D. Compare means in more than two groups

C

Suppose a model is fit to determine whether there is a difference among 3 groups. The parameter estimates are: intercept = 89.62, Group A=3.26 and Group B=-0.87. What is the mean of Group C? • A. 92.01 • B. 86.36 • C. 87.23 • D. 88.75

C

Suppose we have serum albumin value for a 100 patients. The mean serum albumin level is 34.46 g/l and the standard deviation is 5.84 g/l. We could like to know the probability of a value's being above 42 assuming that the true distribution is normal. How many standard deviations above the mean is the value 42? • A. 1.00 • B. 1.18 • C. 1.29 • D. 1.50

C

The most important factor that distinguishes a prospective cohort study from a retrospective (historical) cohort study is: • A. The ability to follow the course of the disease • B. The time sequence of condition and outcome • C. When the trial was initiated • D. The type of conclusions reached

C

The observational study design that best estimates risk of a disease or condition is the: • A. Historical cohort study • B. Case-control study • C. Prospective cohort study • D. Cross-sectional study

C

The primary goal of randomization is to: • A. Ensure different groups prior to intervention • B. Reduce fraud • C. Eliminate bias • D. Create special clusters

C

When continuous data is skewed, which of the following is the best measure of the centrality of the data? • A. Mean • B. Percentile • C. Median • D. Standard deviation

C

Which of the following is NOT true of the receiver operating characteristic (ROC) curve when used to assess the overall effectiveness of a diagnostic test? • A. It is the plot of test sensitivity (y) versus its 1-specificity or false positive rate (FPR) (x) • B. It can take on any value between 0 and 1 • C. It depends on prevalence • D. It can be used to assess the performance of a test independently of the decision threshold

C

Which of the following is not true of a discrete random variable and its distribution? • A. The probability associated with each value lies between 0 and 1. • B. The sum of the probabilities of all values equals 1. • C. It can assume any one of a countless number of values • D. The probabilities are additive

C

ANOVA divides variation in all subjects into two parts, which are: • A. Variation among subjects and variation between each subject and the overall mean • B. Variation between each subject and the overall mean and variation between each subject and one of the group means. • C. Variation between each subject and the overall mean and variation between each subject and all the group means. • D. Variation between each subject and the group mean for that subject and variation between each group mean and the overall mean

D

F-tests are used to: • A. Compare the means in two groups • B. Compare means in more than two groups • C. Compare medians in two groups • D. Compare variances in two groups

D

If the assumptions for the test used in ANOVA are not met, the non-parametric test that should be carried out instead is the: • A. Wilcoxon signed rank test • B. Wilcoxon rank sum test • C. F test • D. Kruskall-Wallis test

D

If the correlation between two variables is 0.70, what is the value of the coefficient of determination? • A. 0.70 • B. 0.35 • C. 1.40 • D. 0.49

D

If you did not have JMP or other software to give you the p-value for the test used in analysis of variance, what would you use to determine whether a test statistic was significant or not? • A. t-distribution table with n1 + n2 -1 degrees of freedom • B. F-distribution table with n1 + n2 -1 degrees of freedom • C. t-distribution table with N-1 and k-1 degrees of freedom • D. F-distribution table with N-k and k-1 degrees of freedom

D

In which of the following ways are nominal and ordinal variables summarized descriptively? • A. Means and standard deviations • B. Variances • C. Odds ratios • D. Frequencies and percentages

D

One of the assumption(s) for performing the two-sample t-test: • A. Observations are not normally distributed • B. The test for unequal variances must equal 1 • C. The diffference between means must equal 0 • D. Observations occur independently

D

The alpha-level for a statistical test is: • A. The same as the p-value for the test • B. The probability of accepting the null hypothesis when it is actually false. • C. The same as the critical value for the test • D. The probability of incorrectly rejecting the null hypothesis when it is actually true

D

The area between two standard deviations of the mean of the normal distribution is • A. 0.68 • B. 0.90 • C. 0.99 • D. 0.95

D

The best measurement of risk of a disease or condition in the retrospective cohort study is the: • A. Percent reduction • B. Hazard Ratio • C. Odds Ratio • D. Relative Risk

D

The primary reason that a clinical trial may be considered a type of prospective study is that • A. The investigator decides the exposure for each patient • B. Both are designed to determine causation • C. Patients are identified on the basis of their disease and compared for previous exposures • D. Comparison groups are selected in advance and followed to determine an outcome

D

The probabilities computed using Baye's Theorem (sensitivity, specificity, etc) are called • A. Simple • B. Composite • C. Joint • D. Conditional

D

The variability of the sample means around the population mean is called • A. Sigma squared • B. Sigma • C. The standard deviation • D. The standard error

D

To study a disease or condition that has a rare outcome, the most efficient design is: • A. Retrospective Cohort • B. Prospective Cohort • C. Cross-sectional • D. Case-control

D

Which is the most appropriate graphic to use in examining the relationship between two continuous (numerical) variables? • A. Stem-and-leaf plot • B. Box-and-whisker plot • C. Histogram • D. Scatter-plot

D

Which of the following is the correct definition for "incidence"? • A. One quantity divided by another quantity in which the population in the numerator is a subset of the population in the denominator. • B. The number of deaths in a defined population over a specified period. • C. The proportion of persons in a given population that has a particular disease at a point or interval of time. • D. A rate giving the proportion of people who develop a given disease or condition within a specified period of time.

D

Assuming that the height of adult males has a normal distribution, what proportion of males will be more than 1.96 standard deviations above the mean height? • A. 0.025 • B. 0.010 • C. 0.050 • D. 0.100

idk