Stat Exam 1

Simple random sample (SRS)

every individual in the population has an equal chance of being selected for the sample

Systematic sampling

subjects are selected at a fixed interval determined by the total number of individuals and the sample size desired (ex. every 5th person is chosen)

Stratified random sample

where a population is divided into groups of similar characteristics and the number of individuals randomly chosen from each group is determined by the proportions of the true sizes of these groups in the larger population

Cluster random sample

where pre-existing small groups are selected at random from within the population and use all individuals in the selected clusters

Multi-stage sampling

uses stages to randomly sample increasingly more specific groups, typically using two or more random sampling methods

Probability sampling methods

SRS, systematic sample, stratified random sample, cluster random sample, and multi-stage sample

Non-probability sampling methods

convenience sample, volunteer sample, and haphazard sample

Convenience sample

sampled individuals are found at a time or in a place that is handy for researchers

Volunteer sample

include only individuals who have taken the initiative to participate, as opposed to having been recruited by researchers (typically participants have strong opinions that aren’t representative of an entire population)

Haphazard sample

when individuals of a sample are selected without a scientific plan, according to the whim of whoever is drawing the sample (not reliable because without a plan, the study cannot be repeated)

Two sample independent

study design where the initial sample is split into two independent groups using a categorical variable that each receive different treatments so that group results can be compared

Two sample paired

study design where the initial sample is split into before and after treatments for the same group of people so that individual differences between measurements can be compared

Randomized block design

study design where experimental units are divided into homogeneous blocks based on potential confounding variables and within each block, experimental units are randomly assigned treatments

Placebo effect

occurs when subjects know that they are receiving treatment

Experimenter effect

occurs when researchers know who is receiving treatment

Hawthorne effect

occurs when subjects change behavior because they think they are being watched

Nonrealism

occurs when the methods used in statistical analysis do not match the reality of the data or the population being studied

Non-compliance

occurs when after agreeing to be in a study, subjects do not comply with treatment assigned

Confounding variables

other variables present that may influence explanatory and/or response variables

1 variable categorical summaries

count (X), proportion (p=X/N), and percent

1 variable categorical visualizations

bar graphs and pie charts

1 variable quantitative summaries

mean, standard deviation, Z-score, median, and IQR (technically also five number summary)

1 variable quantitative visualization

histogram, stem-plot, and box plot

Empirical (68-95-99.7) rule

important because it tells us whether a value is common or unusual for a particular data set