stats Flashcards
what is a type 1 error
A type I error is to reject the null hypothesis when it should be accepted. (false positive, or alpha error)
what is a type 2 error
A type II error is to accept the null hypothesis when it should be rejected (when a genuine difference exists).
false negative, or beta error
what is the power of a study
The power of a study is the probability of rejecting the null hypothesis when it is false, that is, the probability of concluding a result is statistically significant. a test needs enough data to minimise risk of faalse +ve or false -ve
1-B
where B is the B error (type II error)
Acceptable power is 80-90% , which equates to B value of 10-20%. This means a 10-20% chance of a false negative.
when should a null hypothesis be rejected
The null hypothesis is rejected if a significant difference is found.
what are the levels of evidence for studies?
The level of evidence hierarchy is:
Level 1 - High-quality randomised controlled trial with statistically significant difference or no statistically significant difference but narrow confidence intervals (prospective controlled)
Level 2 - Prospective comparative study (prospective uncontrolled)
Level 3 - Case-control study, retrospective comparative study (retrospective controlled)
Level 4 - Case series (retrospective uncontrolled)
Level 5 - Expert opinion.
whats the relationship between SD and variance
SD is the square root of variance
what is SD
SD is a measure of spread of data around a central point. it is also the square root of variance as seen in the equation
what is variance
variance is a measure of the spread of data around a central point. it is the square of SD
what percentage of data will fall within
1SD of the mean
2 SD of the mean
3 SD of the mean
1SD of the mean =68%
2 SD of the mean= 95.7 %
3 SD of the mean= 99.7%
what is Standard error of the mean?
SEM= measure of a group of sample means around the true population mean.
as sample size increases, SEM decreases.
what is confidence interval?
The range of values that will contain the true population mean with a
stated percentage confidence. Used in parametric tests.
A95%confidenceinterval is ±1.96SD andis themost frequently quoted. There is a
95% certainty that this range of values around the mean will contain the popula
tion mean
explain a positive skew and draw the graph
explain a negative skew and draw the graph
what is p-value
define number needed to treat
The number of patients that have to be treated to prevent one outcome
event occurring
define absolute risk reduction
The numerical difference between the risk of an occurrence in the control and treatment groups.
(Incidence in treatment group) - (Incidence
in control group)
define relative risk reduction
The ratio of the absolute risk reduction to the control group incidence (%)
(absolute risk reduction)/ (control incidence)
Relative risk
the ratio of risk of an occurrence in the treatment group to that in the control group:
incidence in treatment group/ incidence in control group)
If control incidence is low, this can lead to an overestimation of the treatment effect.
odds ratio
is the ratio of the odds of outcome in treatment group to the odds of outcome in the control group.
a/c over b/d (e.g smoker with ca./ non smoker with ca. OVER smoker without ca/ non smoker without ca.
1.what are unpaired tests?
2.what are paired tests
- unpaired tests: Different patients are studied in each of the intervention groups.
- Paired tests: The same patient is studied for each intervention, thereby acting as their own control. Matched patients can also be used.
what is students t-test
parametric test for comparison oof sample means
t= difference between sample means/ estimated SE of the difference.
once a value for t is found it is read from a table to see if it represents a statistically significant difference at the level of probability required, e.g p<0.05
draw table comparing statistical tests
what is sensitivity?
The ability of a test to correctly identify a positive outcome where one
exists. (%)
what is specificity?
The ability of a test to correctly identify a negative outcome where one
exists. (%)
what is PPV?
The certainty with which a positive test result correctly predicts a positive
value. (%)
what is NPV?
The certainty with which a negative test result correctly predicts a negative
value. (%)
what are the phases of clinical trial and number of participants.
- health volunteers: pharmacokintetics an dynamics (20-50)
- effect of freq and dose on pharmacokinetics and dynamics (50-300)
- RCT 250-1000+ (comparison and assess frequency of Side effecrs
- post-marketing survillence: rare SE (2000-10,000+)
