1. A well-designed clinical trial is

Always ethically sound

Mostly not ethically sound

Need not be ethically sound

Occasionally ethically sound

2. Translational research is

Cycling of research and feedback from laboratory to clinic to laboratory

Development of potentially new therapeutic modalities

Evaluation of safety and efficacy of treatment

Research conducted in an area with predominantly different language


3. An independent variable in a study is

A parameter that can be determined or manipulated at the outset

An outcome that is determined by an unbiased observer

That cannot be determined by inclusion and exclusion criteria

That can only be a single variable

4. Which of the following is not included in the Bradford-Hill criteria for assessing casuality?

Consistency

Sensitivity

Specificity

Temporality

5. A random error is

Transient, inconsistent and cannot be corrected

Not affected by sample size

Controlled by increasing accuracy

Corrected by adjusting for the error in measurement

6. Which of the following is an example of an analytic study?

Case series

Case control

Cross-sectional

Ecological

7. The use of serially numbered opaque sealed envelopes is a method for

Allocation concealment

Randomization

Blinding

Interventional compliance

8. Analysis by intention to treat is carried out without considering all of the following except

Protocol deviation

Allocation assignment

Compliance

Withdrawal

9. The appropriate graphical representation for the continuous data is

Bar diagram

Histogram

Line chart

None of the above

10. The variables which can be experimentally manipulated by an investigator are called

Dependent variables

Independent variables

Confounding variables

Extraneous variables

11. Which one of the following is not a measure of dispersion?

Range

Percentile

Variance

Interquartile range


12. Descriptive statistics deals with

Description of decision-making process

Organising, displaying and describing the data

Probability distribution

None of the above

13. If a statistical test does not reject null hypothesis, it is known as

Type-I error

Type-II error

Both the above

None of the above

14. Which of the following is a measure based on every item of obseration?

Mode

Standard Deviation

Range

Quartile Deviation

15. Which of the following is the strongest correlation?



0·98

0·0

0·5

16. In a single-factor ANOVA, the computed value of F will be zero when

There is no difference in the treatment means

There is no difference in the block means

The data are skewed left

F will never be zero

17. The appropriate distribution to test whether or not two population variances are equal is

Z-distribution

F-distribution

Chi-square distribution

Standard distribution

18. Error deviations measure distances

Within groups

Between groups

Both and

None of the above

19. The representative sample is used so that the results of a study are

Reliable

Generalisable

Convenient

Limited

20. The population from which the study sample is selected is known as

Accessible population

Target population

Total population

Universal population

21. A sampling distribution is the probability distribution for which one of the following?

Sample

Sample statistics

Population

Population parameter

22. Which of the following is the most common example of a situation for which the main parameter of interest is a population proportion

Binomial Experiment

Normal Experiment

Randomized Experiment

Observational Study

23. The expected value of a random variable is the

Value that has the highest probability of occurring

Mean value over an infinite number of observations of the variable

Largest value that will ever occur

Most common value over an infinite number of observations of the variable

24. Positive confounder causes over-estimation of an association and negative confounder causes under-estimation of an association.

The statement is true

The statement is false

Only the first part is true

Only the second part is true

25. In a statistical test, if the mean scores from two populations differ significantly, then

Accept Null Hypothesis

Reject Null Hypothesis

Accept Alternative Hypothesis

None of the above

26. Randomization is a process that assigns participants

By chance

By choice

Alternatively to either arms

None of the above

27. With large sample size, an estimate will be

Closer to population parameter

Equal to population parameter

With more standard error

None of the above

28. Double-Blinding means

The participant and the statistician do not know to which group the participants are assigned

Neither the participant nor the investigator knows to which group the participants are assigned

Either the participant or the investigator knows to which group the participants are assigned

The participants and the investigator close their eyes before administrating treatment

29. If a statistical test does not reject null hypothesis, then it is possible that we made

Type-I error

Type-II error

Both Type-I and Type-II errors

The power of the test is high

30. Sub-group analyses are indicated

When RCT is done on a large sample

To look for potential heterogeneity of treatment effect related to risk

Always

Never to be done

31. James Lind carried out the first controlled clinical trial. The disease he studied was

Scurvy

Pellagra

Rickets

Beri Beri

32. Clinical trial monitoring is done to

Ensure favourable results

Avoid bias

Control confounders

None of the above

33. Regression is used

To measure the association between two variables

To estimate dependent variables using independent variables

To estimate independent variables using dependent variables

None of the above

34. In a regression equation y 10 5x, which of the following is correct

Correlation coefficient between x and y is 5.

For a unit increase in there is 10 units increase in y.

For a unit increase in there is 5 units increase in x.

For every unit change in there is 5 units change in y.

35. A variable that changes in response to another variable is a/an

Independent variable

Dependent variable

Random variable

Manipulated variable

36. In a regression of Birth weight on Gestational age

Birth weight is the predictor

Gestational age is the predictor

Both Birth weight and Gestational age are predictors

Information is insufficient to answer

37. The 95% confidence interval for the prevalence of hypertension is (0·12, 0·29). The true representation is

Probability that this interval may hold the true value, is 0·95

The population parameter will be between 0·12 to 0·29

The estimated value is higher than the expected level in the population

None of the above

38. P-value is a statistical test, i.e.,

Probability of getting extreme values under null hypothesis

Probability of getting extreme values under alternate hypothesis

Probability of rejecting null hypothesis when it is true

None of the above

39. Translational research refers to

Paradigm to biomedical research

Focuses as interactive feedback loops

Accelerate knowledge transfer from bedside to bench

All of the above

40. Statistical test on hypothesis is

Always based on null hypothesis

Always based on alternate hypothesis

Based on both and

Based on the power of the test

41. Factorial experiment helps in

Studying several factors at a time

Studying a single factor multiple times

Evaluating the effects of interaction

Both and

42. In a study of comparison of the mean ages at marriage of women in two different States of the country, the null hypothesis will be

Significant difference in the mean ages of marriage

Same mean marriage ages in the two States

The mean age of marriage in one State is higher than the other

There is difference in the mean ages of marriage

43. Institution Review Board

Is empowered to approve the study

Cannot insist on modification

Concerns only with the ethical aspect of study

All of the above

44. Case-control study design provides

Maximum evidence

Greater evidence than Cohort study

Minimal evidence

Better evidence than case series

45. Incidence rates can be best calculated in

Case series

Case-control studies

Cohort studies

Randomized controlled trials

46. A Case-control study

May begin with exposure

Always begins with exposure

May begin with disease

Always begins with disease

47. Multiple outcomes are best studied in

Randomized controlled trial

Case-control study

Cohort study

None of the above

48. The advantages of a cohort study are all of the following except

Ability to assess incidence

Ability to assess rare exposures

Ability to assess rare outcomes

Ability to study the relation between exposure and outcome

49. A phase-4 trial evaluates

Dosage

Delivery mechanisms

Short-term safety

Long-term safety

50. T-tests are most useful in which type of data?

Continuous

Ordinal

Nominal

Binary

Write short notes on any four of the following in about 200 -300 words each. Each carnes five marks.

51. Phases of Clinical Trials

52. Binomial Distribution

53. Sampling Methods

54. Measurement of Central Tendency

55. Case-Control Studies

56. Double Blinding Trials