Statistics - definitions Flashcards
Power of a study
Probability of correctly rejecting the null hypothesis (that there is no difference between treatment arm and placebo) when it is false (i.e. not making a type II error)
Standard error
Used in the calculation of confidence interval – many sample means
Type I error + not affected by + increases with..
This is where the null hypothesis (that there is no difference between remdesivir and placebo) is incorrectly rejected (false positive result).
Termed alpha
Not affected by sample size as significance level determined in advance. The chance increases if the number of end points is increase
Type II error
determined by
The failure to reject the null hypothesis (that there is no difference between remdesivir and placebo) when it is false (also known as a false negative result).
Termed beta
Determined by sample size and alpha (type I error)
P Value
Probaility of obtaining a result by chance at least as extreme as te one that has actually observed, assuming that the null hypothesis is true
Properties of Normal distribution
Symmetry = Mean= median = Mode
68.3% of values lie within 1 SD of the mean
95.4% of values lie within 2 SD of the mean
99.7% of values lie within 3 SD of the mean
95% confidence interval
IF a repeat sample of 100 observations are taken from the same group 95 of them would be expected to lie in that range
Standard deviation
IS a measure of how much dispersion exists from the mean
Parametric Data tests
Students t-test – paired or unpaired
Pearsons product-moment coefficient – correlation
Non-parametric tests
Mann-Whitney U test
Wilcoxon signed-rank test
Chi squared tests
Spearman, kendall rank- correlation
Endemic
Persistent, usual, or expected level of disease in a given population
Pandemic
Involves epidemics affecting a large number of people across many countries, continents or regions
Validity
Refers to the extent to which something measures what it claims to measure. First major distinction is internal and external validity
Reliability
Is the extent to which an experiment, test, or any measuring procedure yields the same result on repeat trials
Internal validity
Is the confidence that we can place in the cause and effect relationship in a study. It is the confidence that we have that the change in the independent variable caused the observed change in the dependent variable (rather than due to poor control of extraneous variables
Internal validity
example
A researcher has suggested a causal link between a new designer drug and psychosis. During peer review of the research, this link is challenged by the fact that the temporal relationship between to the two has not been proven. What form of validity has been brought into question?
External Validity
IS the degree to which the conclusions in a study would hold for other persons in other places and at other times i.e. its ability to generalize
external validity example
A PhD student conducted a small study which demonstrated that a new form of psychological therapy successfully alleviated symptoms of depression. They therefore made a claim in a review article that the new therapy is successful in treating depression. They were criticised for making this claim as the therapy had not been widely tested and so the results may not be generalisable.
Content validity
Content validity refers to the extent to which a test or measure assesses the full content of a subject or area.
content validity example
For example if a test is designed to help diagnose depression, it would have poor content validity if it only asked about psychological symptoms and neglected biological ones
Face validity
Face validity refers to the general impression of a test. A test has face validity if it appears to test what it is meant to
Criterion validity
Criterion validity concerns the comparison of tests. You may wish to compare a new test to see if it works as well as an old, accepted method. The correlation coefficient is used to test such comparisons
Criterion validity (concurrent)
In concurrent validation, the predictor and criterion data are collected at or about the same time.
Criterion validity (concurrent) example
An example could be testing a new, shorter test of intellectual functioning against a standard measure
Criterion validity (predictive)
In Predictive validation, the predictor scores are collected first and criterion data are collected at some later/future point. Here you want to know if the test predicts future outcomes.
Criterion validity (predictive) example
An example might be evaluating a new assessment method to select medical students. The test could be compared against the students performance at the end of year one to see if there is a correlation
Construct validity
The extent to which a test measures the construct it aims to
Construct validity (convergent)
A test has convergent validity if it has a high correlation with another test that measures the same construct
Construct validity (divergent)
A test’s divergent validity is demonstrated through a low correlation with a test that measures a different construct
Nominal data types
Observed values can be put into set categories which have no particular order or hierarchy. You can count but not order or measure nominal data (for example birthplace)
Ordinal data types
Observed values can be put into set categories which themselves can be ordered (for example NYHA classification of heart failure symptoms)
Discrete data types
Observed values are confined to a certain values, usually a finite number of whole numbers (for example the number of asthma exacerbations in a year)
Continuos data types
Data can take any value with certain range (for example weight)
Binomial data types
Data may take one of two values (for example gender)
Interval data types
A measurement where the difference between two values is meaningful, such that equal differences between values correspond to real differences between the quantities that the scale measures (for example temperature)
A Rate
is a quantity measured with respect to another measured quantity (e.g. 60 miles an hour).
Attributable risk
is the rate in the exposed group minus the rate in the unexposed group. For example the attributable risk for lung cancer in smokers is the rate of lung cancer in smokers minus the rate of cancer in non smokers. Essentially it tells you what proportion of deaths in the exposed group were due to the exposure.
Relative risk
is the risk of an event relative to exposure. It is also known as the risk ratio.
Population attributable risk
can be described as the reduction in incidence that would be observed if the population were entirely unexposed. For instance how would the incidence of lung cancer change if everyone stopped smoking? It can be calculated by multiplying the attributable risk by the prevalence of exposure in the population.
Attributable proportion
is the proportion of the disease that would be eliminated in a population if its disease rate were reduced to that of the unexposed group.
False-negative rate
Would be the proportion of patients with a disease not picking up by testing
False-positive rate
proportion of patients with a positive test who do not have the condition being tested for
Hazard ratio
The hazard ratio (HR) is similar to relative risk but is used when risk is not constant to time. It is typically used when analysing survival over time
Precision
The precision quantifies a tests ability to produce the same measurements with repeated tests.
Confounding
Refers to a variable which correlates with other variables within a study leading to spurious results
Confounding
Refers to a variable which correlates with other variables within a study leading to spurious results
Confounding occurs when there is a non random distribution of risk factors in the populations. Age, sex and social class are common causes of confounding.
Confounding can be controlled by (design stage versus analysis stage)
In the design stage = randomisation
In the analysis stage of an experiment, confounding can be controlled for by stratification.