Statistics in Practice Flashcards
What is randomisation?
Random allocation into an experimental group
It’s avoids selection bias
What are the 4 types of blinding?
Unblinded/open label - all parties known treatment allocation
Single blind - participants are unaware of treatment
Double blind - participants and clinicians are unaware of treatment
Triple blind - participants, clinicians and data analysts are unaware
Unblinking is when disclosure to participants or study team is received
What is the intention to treat analysis?
It includes every subject who is randomiser according to randomised treatment assignment
It ignore noncompliance and withdrawals
It is an assessment of the people taking part in the trial, based on the group they were initially allocated to
What is a confidence interval?
A range of values that you are fairly sure the true value lies within
What is a 95% confidence interval?
It means if you take 100 samples and assume 95 will contain the true mean value
What is sampling error?
Different samples give different estimates which is sampling error
What is sampling distribution?
Sample estimates that are calculated from multiple samples from the same population will then have a distribution of differing values
What is standard error?
An indication of the extent of the sampling error
What are the assumptions when calculating a confidence interval?
Normal data
Sample chosen at random
Observations are independent of each other
Sample is not small
What is the proportion for 95% confidence intervals?
95% confident that sample proportion is +1.96 standard error
