stud
bud
Sign up
Log in
Home
Library
Explore
SAT
More
EBVM MCQs Chapters 1-10 Flashcards | StudBud
Explore
/
EBVM MCQs Chapters 1-10
EBVM MCQs Chapters 1-10
96 terms
·
StudBud Administration
·
183 plays
Using the MCQs from the Handbook of Veterinary Research Book Chapters (P.Cockroft)
Flashcards
96 cards
F
Practice quiz
P
Host live
L
Questions
96
Answers
1
Given the sensitivity and specificity but not the prevalence it is possible to calculate:
The accuracy
The predictive value of a positive test result
The likelihood ratio of a test result
The predictive value of a negative test result
2
Multiple cohort studies are useful for:
Common risk factors
Rare diseases
Common diseases
Rare risk factors
3
In retrospective cohort studies:
Outcomes are not used
The variables are measured going forwards in time
The variables are measured going back in time
The variables are all measured at a single point in time
4
Chapter 6 MCQs: How many animals/patients are needed? Estimating the sample size
Questions
10
5
What test should be used to compare the means of 2 sets of measurements when the samples are paired?
One-way analysis of variance
Student's t test
Chi-squared test
Pearson r calculation
6
What is a type II error?
A false negative result
A false positive result
7
Which of the following is a strength of retrospective studies?
Recall bias is not a problem
They are more expensive than prospective studies
Measurement of variables rely on historical records/memory
They are useful to study rare conditions
8
Snowballing is:
A selection method that is self-selecting
A randomised method of selecting samples
A method that may be useful in defining contact network structures
A useful method to identify a target population
9
Accuracy of a measurement can best be assessed by:
Comparing it to a gold standard
Repeating the experiment
Repeated measures
Measuring the precision
10
Ordinal categorical variables are:
Classified in unordered categories
Termed dichotomous
Classified into ordered categories
Quantified on an infinite scale
11
Cohort studies are:
A + C
A, C and D
A + B
A, B, C and D
12
Double blinding ensures that:
Investigator does not know the treatment a trial participant is receiving
All of the above
Investigator+owner dont know the treatment a trial participant is receiving
Owner does not know the treatment a trial participant is receiving
13
What is a correlation coefficient a measure of?
The strength of linear association
Statistical power
A type II error
The variability of a measure
14
Continuous variables are:
Classified in unordered categories
Classified into ordered categories
Quantified on an infinite scale
Quantified on a finite scale
15
When selecting a secondary question, which of the following factors should be considered?
They should be specified at the design stage + limited in number
All of the above
There should be a reasonable expectation of obtaining a significant result
They should be chosen on a rational bias
16
Which of the following best defines the terms 'sensitivity' and 'specificity'?
C
B
A
17
Differences between the sample statistic and the target population parameter value are due to:
Individual variation in the parameter
Probabilities of selection
Sample size
Inappropriate selection of samples
18
Sample size calculations are important because:
Neither of the above
Indicates min. number of participants required to achieve a specified power
Indicate the maximum samples that can be recruited for the study budget
Both of the above
19
Which type of variable contains more information, and will allow a reduction in sample size?
Dichotomous variables
Continuous variables
Nominal variables
Discrete variables
20
Which of the following is the least important characteristic of a good clinical research question?
Acceptable financial and time costs
Novel
Technical feasibility
Interesting and relevant
21
Whatinfo's required to calculate sample size for a studycomparing the proportions from a dichotomous measure from2groups
A, D & E
B, D & E
C, D & E
All of the above
22
The most appropriate design to study potential risk factors in rare disease is:
Cohort studies
Case- control studies
Cross-sectional studies
Randomised controlled trials
23
Using the table below indicate what is determined before the start of a cohort study.
(A + C) and (B + D)
(A + B) and (C + D)
n
A + B
24
Nested case-cohort studies are useful for:
When studying rare events
Cost reduction
Investigating more than one variable + outcome
Measuring the prevalence of risk factors
25
Chapter 3 MCQs: sampling, subject, selection criteria and recruitment
Questions
10
26
Applicability (generalisability) is:
The extent the results of a study are relevant to a given population
The usefulness of the results in general terms
The results expected if the outcomes were applied to the general population
The application of the results to all species
27
Prospective studies are:
Usually use records
Do not monitor outcomes
Are conducted forwards in time
Collect non-specific data
28
What is meant by a two-tailed hypothesis?
Answers split across two boxes
2. Accepts that a difference may only be due to an increase in the value...
1. value measured in the experimental group compared to the control group
2. ...measured in the experimental group compared to the control group
1. Accepts that a difference may be due to either an increase or a decrease
29
What is the relationship between sample size calculations for one-tailed and two-tailed hypotheses?
2. ...same as for a two-tailed test at twice that level.
1. ... the same as for a two-tailed test at half that α level
1. The sample size for a one-tailed test at a particular α level is...
2. The sample size for a one-tailed test at a particular α level is the...
30
Chapter 8 MCQ's: Cross-Sectional and case control studies
10
Questions
31
Matching in case-control studies:
Reduces the size of the control group
Reduces the number of variable measurements that need to be made
Reduces the cost of the study
Reduces sampling bias
32
Clusters are:
Randomly selected subpop that're clearly defined + often located in 1 place
Subjects that are interdependent
A group of subjects that are randomly dispersed in the population
Subjects that share a characteristic
33
A strategy that will increase precision but not accuracy is:
Training observers
Refining instruments
Repetition
Automating instruments
34
Proportionate sampling means: (unfinished sentence: in the.. population)
Selecting every nth subject from the population
Randomly select from each stratum based on its relative proportion in the..
Randomly selecting equal numbers from each stratum
Randomly selecting unequal numbers from each stratum
35
Bias may be introduced by:
E
C + D
B
A
36
What is the value of a power calculation (PC) performed after a study has been performed?
A study result is not statistically significant in the absence of a PC
Journal editors will not accept a paper for publication without a PC
A PC provides evidence that a negative result is unlikely to be a FN result
A positive result is not statistically significant in the absence of a PC
37
How can the variance be estimated in advance of collecting the study data?
Using historical data from the same or a similar population
Any of the above
Using a variance value from results published in a similar study
By performing a pilot study
38
Superiority trials are:
Trials that have larger sample sizes
Trials that have a better design
Trials that are performed in a controlled environment
Trials designed to investigate if one treatment is superior to an other
39
Retrospective studies:
Only include animals that have been exposed to the RF
Conducted backwards in time
Do not usually use records
Only include animals that have developed the disease of interest
40
What adjustments should be made when testing multiple hypotheses from the same data set?
A chi-squared test adjustment
A Bonferroni adjustment
Divide desired overall P value threshold by the no.hypotheses being tested
A Pearson adjustment
41
Receiver operator curves are useful because:
They indicate which is the most accurate test
They indicate the cut-off point with the least diagnostic error
All of the above
They indicate the sensitivity and specificity at a given cut-off point
42
Nominal categorical variables are:
Classified in unordered categories
Termed dichotomous
Classified into ordered categories
Quantified on an infinite scale
43
Chapter 4 MCQs: Variables
Questions
10
44
Chapter 2 MCQ's: Establishing the Hypothesis
5
Questions
Side note: explanation of q6 (Q4 in Chap2) answer in the book
45
Which of the following is a weakness of retrospective studies?
All of the above
The cost of the study is difficult to determine
They are more expensive than prospective studies
The type and quality of the measurements cannot be predetermined
46
What is the odds equivalent to a 20% probability?
1:4
0.25
0.2
1:5
47
Inclusion criteria are:
The inverse of the exclusion criteria
Subjects who want to join the study
The requirements that a subject must fulfill to enter a study
The parameter that is being measured
48
Which of the following is a weakness of prospective studies?
Can't be used to estimate the incidence of a specific disease in the popn
It is impossible to determine the time sequence of exposure and outcome
They are not useful to study rare conditions
Measuring the exposure to the putative factor is difficult
49
What is a type I error?
A false negative result
A false positive result
50
Which of the following isn't a suitable strategy when a study is unfeasible because not enough subjects can be recruited
Expand the sources of patients
Reduce the exclusion criteria
Lengthen the timeframe of the study
Perform the study & recruit a statistician to help prove stat. significance
51
What statistical analysis is used for comparing multiple means from more than one group?
Bonferroni test
Student's t test
Analysis of variance (ANOVA)
Chi-squared test
52
Which of the following treatment factors are important to consider when designing a clinical trial?
All of the above
Sample population + generalisability of the results
If different dose rates will be studied
Effectiveness + safety of drug
53
Chapter 9 MCQ's: Clinical Randomised Controlled Trials
10
Questions
54
What is meant by the null hypothesis in a study seeking to establish that there is a difference between two groups?
The results from two groups are statistically different
The research hypothesis is true
That there is no difference between the two study groups
That the study hypothesis is un-true
55
Chapter 10 MCQ's: Studies on Diagnostic Tests
Questions
10
56
What's the correct formula to derive the standard deviation from a standard error of the mean (
N=
no. subjects in mean)?
SD = Standard error of mean x N
SD = Standard error of mean x N2
SD = N x √standard error of the mean
SD = Standard error of the mean x √N
57
An important feature of a RCT is:
It can be published
There are at least 2 groups (treatment + control group)
It is easy to perform
Subjects are readily available
58
Chapter 5 MCQs: Introducing Statistics
Questions
10
59
What criteria have to be satisfied for groups to be 'independent'?
answers split across two boxes due to length
2. That animals are matched only in some regards, but not...
1. ...of an individual in another group.
2. ... for the outcome variable of interest
1. That the selection of an individual has no influence on the selection...
60
How would you best use a diagnostic test with a high specificity but a low sensitivity?
To rule a diagnosis in
To rule a diagnosis out
61
Which of the following is a disadvantage of a clinical RCT:
C
E
D
B
62
Which of the following is an advantage of the cross-over design?
The number of animals can usually be reduced compared to a RCT
Statistical appraisal is simple
Treatments with persistent actions confound the results
It has greater validity than parallel RCT
63
The design providing the strongest evidence about causation is:
Cohort studies
Case-control studies
Randomised controlled trials
Cross-sectional studies
64
Stored samples from subjects in a nested case-control study:
Removes the need for further measurements
Enable all the cohort to be batch tested
Enable selective testing of the cohort once the outcomes are known
Reduce confounding
65
Discrete variables are:
Quantified on an infinite scale
Classified in unordered categories
Quantified on a finite scale
Classified into ordered categories
66
What is the formal statistical description of the research question?
The statistical hypothesis
The research hypothesis
The formal hypothesis
The alternative hypothesis
67
What value for α equates to a
P
value of 0.01?
α = 0.01
α = 1%
α = 99%
α = 0.99
68
How else can the variance be estimated in advance of collecting the study data?
2. ... of measuring the variable of interest
Neither of these, only those in the previous question
2. By seeking advice from colleagues or laboratories that have experience..
By estimating on pathophysiological principles from laboratory data.
69
Which of the following would not increase compliance?
Detailed treatment instructions
Motivational meetings
A short study period and a simple treatment regimen
A long study period and a complex treatment
70
What test is used to test the distribution of individuals to an expected series of categories?
One-way analysis of variance
Mann-Whitney U test
Pearson r calculation
Chi-squared test
71
Prevalence can be determined using:
A cross-sectional study
A randomised control trial
A case-control design
A retrospective prognosis study
72
Chapter 1 MCQ's: Introduction
MCQ's
:)
No
Easy Start
73
Subjective tests using observers should:
A, B + D
E
A, B + C
A, C + D
74
Simple random sampling means:
Each subject in the target population may be selected
Each subject in the target population has an equal chance of selection
Each subject in the target pop has an = and independent chance of selection
Each subject in the target pop has an independent chance of selection
75
Accuracy is:
Increased by random error
Reduced by systematic error
The same as precision
The reproducibility of replicate measures
76
Measurement bias can be reduced in case-control studies by:
Blinding the owners + observers
Ensuring the observer knows whether the records are cases or controls
Ensuring the observer knows the putative risk factor in the study
Using the same observer throughout the study
77
The target population:
Is the sample to be taken
Should be the same as the study population
Is different from the study population
Is a sample of the study population
78
What information is required to calculate the sample size for a study comparing the means of a measure from two groups?
A, B, C, E
A, B + D
A or B; C, D and E
All of the above
79
In cross-sectional studies:
The variables are measured going back-wards in time
The variables are measured going forwards in time
The variables are all measured at a single point in time
Outcomes are not used
80
An important criterion for exclusion would be:
Subjects in whom treatment is unlikely to have any beneficial effect
Subjects who're unlikely to comply with treatment/recording of measurements
All of the above
Prior knowledge that allocation to placebo group would have adverse effect
81
Incidence can be measured using:
Case-control studies
Cohort studies
Cross-sectional studies
Randomised control trials
82
In case-control studies:
The variables are measured going forwards in time
The variables are all measured at a single point in time
No variables can be measured
The variables are measured going backwards in time
83
Chapter 7 MCQ's: Cohort Studies
Questions
10
84
What is meant by the power of a study?
The probability that the study will produce a true positive result
The probability that the study will NOT produce a false negative result
The probability that the test will not produce a false positive result
The probability that the study will produce a false negative result
85
What does the beta value represent?
The probability of obtaining a false positive result
The probability of making a type I error
The probability of obtaining a false negative result
The probability of making a type II error
86
When a P value is reported at <0.05 what does this mean?
That there is a 95% or more chance of obtaining a FN result
That there is a 5% or less chance of obtaining a FP result
That there is a 95% or more chance of obtaining a FP result
That there is a 5% or less chance of obtaining a FN result
87
Which of the following is a strength of prospective studies?
Important variables can be accurately measured- no reliance on records/mems
All of the above
Useful to study conditions with a long incubation period
Useful to study rare conditions
88
Precision is:
The reproducibility of replicate measures
Increased by random error
Increased by systematic error
Reduced by systematic error
89
In prospective cohort studies:
The variables are measured going forwards in time
The variables are measured going backwards in time
No variables can be measured
The variables are all measured at a single point in time
90
Randomisation of participants in a trial ensures that:
Owners do not know which treatment their pet will receive
Investigator does not know the allocation of subjects to treatment
Patients are allocated to each intervention without selection bias
Investigators are randomly allocated to treatment groups
91
Tandem testing can measure:
Which of the two tests is the most accurate
The sensitivity of a test
The prevalence of disease
The predictive values of the tests
92
Sampling is used because:
It gives the same answer as measuring the target population
It gives reliable results
It provides an estimate whilst reducing the time and cost
It is quick and easy to do
93
What does the likelihood ratio for a negative test result tell us?
The likelihood of the animal being disease free, given a positive result
The likelihood of the animal being disease free, given a negative result
The likelihood of the animal having a disease, given a positive test result
The likelihood of the animal having a disease, given a negative test result
94
Nested case-control studies are advantageous when:
Measurement of the variable is expensive
Outcomes are not easily identified
Cases cannot be identified
More power is needed in the study
95
What is the definition of a 95% confidence interval?
Answer spread across two boxes
2. ... an answer within the confidence intervals
2. ...answer within the confidence intervals
1. There's a 5% probability that a repeat of the trial would produce...
2. There's a 95% probability that a repeat of the trial would produce an...
96
Precision can be best assessed by:
Using instruments
Measuring the sensitivity and specificity
Measuring the accuracy
Repeated measures
