Key Concepts in Statistics and Sampling Methods

Analytics

the use of data to drive decisions and actions

Continuous Data

can assume any value within a specified range; results from measurement.

Descriptive Statistics

organized, summarized, and presented data for informational purposes.

Inferential Statistics

data used to make predictions, assumptions, forecasts.

Deduction

The logical process of using the general knowledge of a population to draw a conclusion about a specific element or sample of that population.

Induction

the logical process of using specific knowledge about a sample to infer about the population.

Discrete Data

can assume only certain values; there are gaps between values; results from counting.

Qualitative Variable

a non-numeric characteristic to be studied about each observation.

Quantitative Variable

a numeric characteristic to be studied about each observation.

Statistical Thinking

a philosophy of learning and action based on the principles that (1) work consists of process, (2) variability exists in process, and (3) addressing variability is key to improving process.

Mutually Exclusive

when a unit is included in one category, no overlap.

Exhaustive

when each unit (individual or object) available must appear in one category, all possible units have a chance to be counted/included.

Purpose Sampling Method

Nonprobability: Selection based on a specific interest or focus. Ex: Interested in video game satisfaction of only gamers playing 10+ hours per week

Quota Sampling Method

Nonprobability: Want certain characteristics from subjects to make up a % of sample. Ex: ensuring that 50% of the sample contains gamers playing 10+ hours per week.

Convenience Sampling Method

Nonprobability: selecting subjects based on ease of access. Ex: standing outside a door and catching people for their opinion.

Snowball Sampling Method

Nonprobability: Rare characteristics found by using on/some subjects to find more and so on. Ex: determining satisfaction of Spokane residents driving a Bentley. Find one owner and then ask them to refer you to more owners.

Simple Random Sampling Method

Probability: creating an easy representation of each item so that the sample is selected with each item having the same chance of being included. Ex: slips of paper for all classmates placed into a hat, then selecting 5 from that.

Systematic Random Sampling Method

Probability: arranging subjects in some manner using a random starting point and then selecting every k th of the population. Ex: among a long, alphabetized list of charge accounts, selecting every 33rd listing.

Stratified Sampling Method

Probability: dividing population into subgroup and selecting a given portion from each subgroup. Ex: breaking the population into socioeconomic groups (wealthy, middle class, working class) and then selecting a % from those groups more reflective of the population.

Cluster Sampling Method

Probability: grouping population among a large geographic area using data that relates the groups, then selecting from those groups or regions. Ex: textbook cost study across U.S. by grouping into West, Southwest, South, Midwest, East. and then selecting from a large main university in each region.

Sample Error

The difference between a sample statistic and its corresponding population parameter.