POLS 209 2nd Quarter

68–95–99 rule

A handy guideline for bell-shaped curves showing that 68%, 95%, and 99% of your data will fall within one, two, and three standard deviations from the middle, respectively.

Census

A complete gathering of data from absolutely every member of the group you are studying.

Central limit theorem

A statistical rule proving that if you plot the averages of countless random samples, they will always form a perfectly predictable bell curve centered around the true overall average.

Confidence interval

A mathematically calculated range of numbers where we can reasonably expect the true population average to live based on our sample data.

Frequency distribution

A graph or tally showing the actual, real-world data points and how often each one occurred in a single sample.

Independent outcomes

Random events where the result of one event has absolutely no effect on what happens in the next event.

Normal distribution

A perfectly symmetrical, bell-shaped curve where the average, median, and most common values all sit exactly in the middle.

Outcome

The specific result you get when you observe a random event.

Population

The complete, total set of every possible person or thing you want to draw a scientific conclusion about.

Random sample

A smaller group chosen in a way that gives every single member of the entire population an equal chance of being picked.

Sample

A smaller, manageable subset of data drawn from a larger, underlying population.

Sampling distribution

A theoretical graph showing what the averages would look like if you somehow took an infinite number of samples from a population.

Standard error of the mean

A measure of our uncertainty that shows how much the averages of different random samples are expected to vary from the true population average.

Statistical inference

The scientific process of making an educated, probability-based guess about a massive group by looking only at data from a smaller sample.