EXAM REVISION Flashcards
(28 cards)
Categorical Nominal
a variable that can be measured in more than two categories that arent in any particular order. eg something can be green, blue, or purple.
What are the types of variables?
Continuous - continuous scale in which every value is possible eg. 5.14
Categorical - restricted to one set of categories eg. blue
Discrete - counted on counting/whole numbers (number of children in a family)
Continuous variable
measured in a continuous scale in which every value is possible eg. 5.14
Categorical variable
restricted to one set of categories eg. green or heads
Categorical binary
a variable that can only be one of the two categories that it is measured in. eg. you can only be a smoker or non smoker.
Categorical Ordinal
a variable that can be measured in more than two categories that ARE in a particular order. eg. pain tolerance from 1-10
what is the difference between categorial ordinal and categorical nominal?
ordinal is in a particular ranking, whereas nominal in just any order of variable categories that the thing can be. eg. level of pain is ordinal, blood type is nominal.
discrete variable
a variable that can be counted on counting/whole numbers (number of children in a family)
difference between raito and proportion
ratio is one number compared to another
proportion is one number compared to the whole population
Rate
Like a ratio for things with different units eg new cases per day.
two main types of studies
descriptive or analytic
descriptive study
surverys which aim to describe things eg. characteristics, attitiudes.
uses sample from pop. of interest
analytic study
aims to test a hypothesis eg. causes, effects, methods of prevention
experimental study
a type of analytic study in which the conditions are manipulated by the researcher (interveneing into the natural process).
Aims to control all other factors to isolate the effects of the intervention
observational study
a type of analytic study in which the researcher does not intervene. they just observe what is happening and collect data
cohort study
type of analytic observational study in which participants are selected for the demographic they’re in BEFORE they’ve gotten the disease and then oberved over time.
case control study
type of analytic observational study in which participants are selected BECAUSE they have the disease
A group without the disease is used as the control.
stratified sampling
when population is comprised of several groups of similar individuals. find the stratums and take an equal (or proportional) amount from each stratum (via random sample)
probability sampling
every member is selected at random. each member of the population has an equal chance of being selected.
eg. cluster, stratified, simple random sampling
how to find the sd of probability distribution
∑ x2 times P(x) - m2
- make x2 column
- times the x2 column by the p(x) column
- add all those numbers together
- get that total and minus it by m2
- find the square root of that number
Cluster sampling
what the pop. is grouped into natural clusters (households) and an amout of hosueholds are selected randomly
two stage - households are randomly selected, and then from within those households, individuals are randomly selected.
Sensitivity
the ability of a test to correctly identify patients WITH a disease
Specificity
the ability of a test to correctly identify people WITHOUT the disease
what are the four factors of binomial distribution?
- fixed number of trials
- each trial must have the same two possible outcomes
- the trials must be independant
- the propabability of success must be the same in each trial
