Statistical Interference Flashcards
What is a distribution?
- Describes the frequency (or probability) of
occurrence for a given value - Describes the shape of the data
What are examples of probability distributions for continuous variables?
e.g. -Height, Age, - Normal, skewed
What are examples of frequency distributions for discrete variables?
e.g. GP visits - Poisson, Binomial
Why is knowing the distribution useful?
- We can use the distribution of a sample to make inferences about a wider population
- to generate confidence intervals ( assessing
variability of estimates) - test hypotheses
- calculate sample size
What is the normal distribution?
A probability distribution that describes data that is symmetric around a mean
The normal has two parameters:
- mean
- standard deviation (SD)
What is skewness?
a measure of the asymmetry of the distribution
What would a negative skew look like?
Elongated tail at the left. More data in the left tail than would be expected in a normal distribution.
What would a positive skew look like?
Elongated tail at the right. More data in the right tail than would be expected in a normal distribution.
How do you distinguish between outcome and exposure?
By formulating a research question using PICO
- Determining whether the intervention
influenced the size or occurrence of the
outcome.
Why are statistical tests used?
Ensure data on a sample can be represented on the overall population.
Statistical methods are needed when outcomes are unpredictable
What is the null hypotheses?
= Outcome in not associated with exposure
What is the hypotheses?
= Outcome is associated with exposure
What is a Type I Error False - Positive Alpha (α)
Occurs if an investigator rejects a null hypothesis that is actually true in the population.
What is the significance level?
- The probability that you will find an effect
that does NOT actually exist - Strength of evidence needed to reject NULL
hypothesis - Normally set to 5%
What is Standard Deviation and is it a summary or inferential statistic?
SD = a measurer of how variable individual
measures are
- how spread out the values are of this
variable
= a summary statistic
What is Standard Error and is it a summary or inferential statistic?
SE = an estimate of how variable a statistic would be if we repeated out study numerous times
- how precise the estimate of true mean is
based on the sample mean
- a statistic is a value calculated from our sample
Is Standard Error a summary or inferential statistic?
= a inferential statistic
What is Standard Error usually used for?
Used to create a confidence interval
e.g. We can be 95% confident that out true population mean lies in our 95% confidence interval
What is the equation for standard error?
SE = SD/ √n
What is a P-value ?
P-value tells us the strength of the evidence against the null hypotheses (that there is no association)
It is the probability that we observed an effect size as large as we did if the null hypotheses is true e.g. effect size is zero
What is a confidence interval?
A confidence interval gives us the range of values within which we are reasonably confident the true difference lies.
What are both p values and confidence intervals based on?
They are both based on standard errors.
The smaller the error, the smaller the p-value and narrower the confidence interval.
What happens to the null hypothesis as the p-value decreases?
As the p-value decreases the evidence against the null hypothesis increases.
What does a p-value of 0.1 mean?
Weak evidence against the null hypothesis
What does a p-value of 0.01 mean?
Increasing evidence against the null hypothesis with decreasing p value.
What does a p-value of 0.001 mean?
Strong evidence against the null hypothesis
What do the confidence intervals mean?
Confidence intervals show the range of values in which the true effect size is likely to lie.
What does a 95% confidence interval mean?
Tells us that in 95% of replicate experiments the true value will lie in the interval.