a parameter is a numerical summary of a variable's distribution in a given population
a census is an attempt to generate the population distribution (thereby producing the parameters) by measuring the variable of interest on every individual in the given population
a sample is a portion of the population chosen in such a way so as to reliably represent the population
simple random sample (SRS) of given size = n
is a sample chosen in a way such that: a) every individual in the population has an equal probability of making it into the chosen sample b) every sample of a given size = n has an equal probability of actually being the sample chosen
a statistic is a numerical summary of a variable's distribution in a given sample
available data is data that has been produced in the past. Some pros include easily accessible and inexpensive. Cons include no way to verify possible bias and might not answer the question of interest
an observational study is a study in which no treatment is imposed. Such studies include research of available data or sample surveying.
An experiment is a study in which an SRS of a given size = n is selected from a population of interest, and allocated into treatment groups (control and experimental) using randomization. Treatments are then imposed upon each unit (placebo and drug) in each group, and the responses summarized and compared using a chosen statistic. Levels of statistical significance are then determined.
randomization is the process by which treatment groups are created from the SRS using a table of random digits. Such as process creates treatment groups that should be similar to one another except for slight variations due to chance
experimental units, subject treatments
the individuals upon which the experiment is performed are called experimental units. When the units are human beings, they are called subjects. A specific condition applied to the units is called a treatment
causation in an experiment is said to have occurred when changes in a given explanatory variable singularly cause changes in a response variable