Stats Definitions

Population

The set of all objects or persons eligible to be included in the study

Sample

a subset of the population of interest

Variable

a trait of an object under study

Statistics

The science of data

Population parameter

A number that describes a trait’s performance

Sample statistic

A number that describes a sample’s variable performance

Descriptive statistics

Use of numbers to describe data and help identify trends

Inferential statistics

Use of numbers to make general conclusions about a population

Quantitative data

Data that can be recorded on a numeric scale in a meaningful way

Qualitative data (categorical data)

Data that cannot be recorded on a numeric scale in a meaningful way (the data can only be categorized)

The two most common misuses of statistics

Biased sampling in order to obtain a desired result and making claims about proving causal relationships

Random sampling

A method of choosing a sample so that every sample of the same size is equally likely. Each person or object has the same chance (probability) of being chosen.

Stratified sampling

Choosing the sample in such a way that significant subgroups of the population are represented in the sample in the same percentage that they exist in the population. The researcher must know something about the composition of the population.

Cluster sampling

Preexisting groups rather than individuals are randomly selected for participation in a study

Systematic Sampling

Choosing every k^th subject in a population to be in the study until the desired sample size is obtained

Biased sample

A sample that over represents some traits that influence the results of the study

Sampling bias

Systematic differences in performance between the sample and the population

Convenience sampling

Use of whomever happens to be easily accessible to obtain a sample

Quota sampling

Use of quotas for participants with given traits to determine who is eligible for participation. The ratios in which these traits are quota-ed does not necessarily reflect the population.

Nominal measurement scale

Organizes data into categories. Usually done with qualitative data. Does not provide information about the extent to which one member of a group differ from another.

Ordinal measurement scale

Breaks people, actions, or responses into categories and ranks them by the extent to which they possess some trait

Interval measurement scale

Can be broken into categories, ranked by possession of some taint or score on some variable, and equal differences in rank represent equal differences in possession of the trait or variable

Ratio measurement scale

Can be broken into categories, ranked by possession of some trai or score on some variable, equal differences in rank represent equal diffrences in possession of the trait or variable, and there is a true zero point.