Data
Information that is collected for a statistical analysis. Just the facts, ma'am
Estimate
A huge fan of the population parameter, so it tries its best to imitate it. Measured from a sample.
Expected Value
The long-term average from repeating an event over and over again. Shmoop takes no responsibility if you run out of money before reaching that point.
Experiment
A process for formally testing an idea under controlled conditions. Has led to such scientific advances as the baking soda volcano and the potato-powered light bulb. Also, baking soda and light bulbs.
Margin Of Error
95% of the possible sample estimates fall within ±(margin of error) of the estimate. It isn't used in very many recipes, unlike the margarine of error.
Normal Distribution
A hump-shaped, symmetric distribution of continuous data. Always tops statisticians' lists of best distributions.
Null Hypothesis
An assumption about the population parameter that allows us to create a sampling distribution. We love to reject it when the data lets us. We can be cruel at times.
Observational Study
A study of an already-created group, sampled without randomization. Be sure to tell people that you're conducting an observational study, or they may just think you're a creeper for staring at them and taking notes.
Parameter
Some aspect of the whole population. It's kind of creeped out by the estimate always following it around.
Population
Every item that a study could potentially be interested in. If you are studying pancakes, then "every pancake ever" is the population. That is a good population to study, especially with syrup.
P-Value
The probability of getting a result as likely as the test statistic, or even more extreme. It's always getting compared to the significance level and is starting to resent the comparison.
Random Sampling
The process of randomly selecting individuals from the population to be in a sample. This is why we can say that a sample is representative of the population, not luck or magic.
Sample
A subset of the population. If you are studying pancakes, then "these pancakes Mom made this morning" is a sample. We're going to need a bigger sample.
Sampling Distribution
The probability distribution of possible values of the estimate if you repeatedly took the same size of sample from the population.
Sampling Survey
A random sample used to estimate a parameter of the population. Once the sample runs out, you have to pay $19.99 plus processing and handling to get the full survey.
Significance Level
The cutoff for rejecting the null hypothesis of the sampling distribution. It's often 0.05, but sometimes moonlights as 0.1 or 0.0.1.
Simulation
What would the sample estimate look like if the population followed the null hypothesis? We use simulation to find out. And computers. Lots of computers.
Standard Deviation
Measures the spread of the data around the mean. Has the scary-looking formula
, but is really a big softie at heart.
Standard Normal Distribution
A normal distribution with a mean of 0 and standard deviation of 1. You can't get more standard, more normal, or more distribution than that.
Standard Normal Table
A table of values giving the probabilities for different Z-scores. You can't get more standard, more normal, or more table than that.
Statistics
the branch of math that deals with collecting and analyzing data; you can justify almost anything with the right stats
Summary Statistics
Single numbers that sum up a lot of information about a dataset. The mean, median, mode, standard deviation—they've all got something to say, hey hey.
Test Statistic
Something calculated from the data in order to see if the data are consistent with the null hypothesis. Students are tested on statistics, and test statistics are tested by students. It's the circle of math.
Z-Score
The number of standard deviations away from the mean a particular value is in a normal distribution. It will almost never be over 9000.