Lecture 2- variables

population

a collection of people who share a common observable characteristics

sample

a subgroup that has been selected from the population (for a study, using one of several selection methods)

parameter

numerical descriptor of a population

statistic

numerical descriptor of a sample

representativeness

the degree to which characteristic of the sample correspond to the characteristics of the population from which the sample was chosen

estimation

using sample-based data to infer conclusions about the population

why sample sizes?

impossible to study, inferences are drawn about populations, possible cost savings

simple random sampling

every member of the population has an equal chance of being selected, most useful for small populations because it requires a complete enumeration of the population as a first step

systematic sampling

sample size and base population, uses a systematic procedure to select a sample of a fixed size, every 11th person gets chosen, feasible when a sampling frame, like a list; need to be careful of hidden patterns in lists

stratified random sampling

population split into non-overlapping groups, following which, simple random sampling applied in each group; guarantees that members from each group will be represented in the sample

cluster sampling

population divided into groups, following which clusters are chosen (instead of individuals)

convince sampling

includes the individuals who happen to be most easily accessible to the researcher, representativeness of the sample to the population is questionable

voluntary response sampling

individuals volunteer themselves to participate in the study by responding to advertisements, people who respond may be overly health conscious or may not represent general populations

categorical/qualitative variables

contain a finite number of categories or distinct groups, they usually don’t have logical order or hierarchy; ie eye color (green is not greater than blue)

binary variables

contain only 2 categories (yes/no)

nominal variables

contain more than 2 categories, categories not ordered; type of property (apartment, condo, house)

ordinal variables

contain 2 or more categories, the categories are ordered or ranked, difference between 2 categories not necessarily equal, loose hierarchy can’t place a value

continuous/quantitative variables

numeric variables that have an infinite number of values between any 2 values; height, weight

interval variables/scale

can be measured along a continuum, have a numerical value measured with equal difference between consecutive values; difference between 20 and 30 is the same as 30 to 40

ratio variables/scale

these are interval variables, but with the condition that 0 of the measurement indicates that there is none of that variable, blood lead levels being 0 mens no blood lead

discrete variables

have a finite number of values in whole numbers, can be categorized, how many children someone has is 2 not 2.6