sampling and survey designs Flashcards
population of interest
set of all individuals/units we seek to study
sample frame
sampling frame is a list of units from which the sample is chosen. ideally, it includes the whole population
unit
individual person, item or measurement
strata
- subset (part) of the population divides population into strata
statistic
- describes a sample
parameter
- describes an entire population
simple random sampling
Simple random sampling is used to make statistical inferences about a population. It helps ensure high internal validity: randomization is the best method to reduce the impact of potential confounding variables
each possible group of a chosen size from the population has an equal chance to be selected
from population -> (randomly selected) (unbiased sample)
(best method to represent a population from a sample)
stratified sampling
stratified sampling is appropriate when you want to ensure that specific characteristics are proportionally represented in the sample. You split your population into strata (for example, divided by gender or race), and then randomly select from each of these subgroups
population is subdivided into 2+ groups based on pre-existing characteristics, then only a few are selected from each grouping to be sampled
cluster sampling
cluster sampling is appropriate when you are unable to sample from the entire population. You divide the sample into clusters that approximately reflect the whole population, and then choose your sample from a random selection of these clusters and collect data from every unit in the sample (census)
could lead to (bias sample)
systematic sampling
systematic sampling involves choosing your sample based on a regular interval, rather than a fully random selection. It can also be used when you don’t have a complete list of the population
a unit is selected as the starting point and more units are chosen for sampling at a constant interval (k), from the starting point (ex, 1, 5, 9, 13)
convenience sampling
where the sample is “convenient to take”, like from friends or family, workplace (easy to reach)
ex) standing at a mall or a grocery store and asking people to answer questions
response variable
measures outcome of interest
observational study
individuals are observed or certain outcomes are measured, no attempt to affect response variable/outcome (ex, no treatment given, surveys)
experimental study
manipulates the explanatory variables with the intent to influence the response variable
treatment
any specific experimental condition(s) applied to individuals/objects
completely randomized design
units from sample are randomly assigned to treatments
randomized block design
units from sample are subdivided into ‘blocks’ based on pre-existing characteristic(s), then randomly assigned from blocks to a treatment
matched pair design
2 units are matched and paired based on similarity of characteristics, then randomly assigned treatments which the pair’s responses are compared
repeated measures design
each unit is assigned to all treatments usually in random order
confounding variable
- a confounder variable influences both the dependent variable and independent variable
ex) ice cream sales (dependent variable) and relationship of sun burns (independent variable)
sun exposure (confounding variable) is the reason why the dependent and independent variables show a relationship
replication
- have more than one individual/unit in each treatment group to reduce chance variation in the results
randomization
- random assignment of individuals/units to treatments, or the order of treatments
concerns with experimental studies
- confounding variables: use randomization to control
- interacting variables: measure and report variables that may interact
ex) condiment on food - placebo, hawthorne and experimenter effects: use double-blind design and placebo/control groups
- ecological validity and generalization: try to perform experiment in natural setting
experimenter effect
- observer-expectancy effect
- the observer expects or knows history of the experiment
