Term
| The whiskers of a boxplot tell you... |
|
Definition
|
|
Term
| The top of the box in a boxplot tells you... |
|
Definition
|
|
Term
| The bottom of the box in a boxplot tells you... |
|
Definition
|
|
Term
| The bar/line in a boxplot tells you... |
|
Definition
|
|
Term
| The formula to determine whether a point will be an outlier is... |
|
Definition
|
|
Term
|
Definition
a summary measure that is the property of the population. It is what a researcher is trying to measure/test in a study.
For example, the mean height of 20 year old women. |
|
|
Term
|
Definition
a summary or number that is a property of the sample. It is the answer to the parameter - the result that is reached.
For example, the statistic of the mean height of 20 year old women is 5 foot 4 inches. |
|
|
Term
|
Definition
| Not affected by any variable |
|
|
Term
|
Definition
| A type of independent variable that isn't independent for certain. Always plotted on the x-axis. |
|
|
Term
|
Definition
basically the same thing as dependent variable.
Important to find a good outcome variable for the respective study.
i.e. Post Office's outcome variable for the "bigness" of a package is weight in lbs and girth in inches. |
|
|
Term
| Standard Deviation (Definition) |
|
Definition
Typical deviation from the mean. Give or take; average distance to the average.
Smaller standard deviation desirable, demonstrates that data is more stable. |
|
|
Term
Standard Deviation (Formula)
and steps to find |
|
Definition
It is the SQUARE ROOT of the variance.[image]
Step 1: Find mean, n, and n-1.
Step 2: Subtract each value by the mean.
Step 3: Square every amount.
Step 4: Add every squared value.
Step 5: Divide by n-1.
Step 6: Square root the quotient. |
|
|
Term
|
Definition
| It does the same thing as the standard deviation. It just doesn't have interpretable units. |
|
|
Term
|
Definition
| the average amount that x and y deviate together. Used to calculate correlation. |
|
|
Term
|
Definition
The bounded version of covariance. Between -1 and 1. Having either -1 or 1 means a perfect linear relationship.
No units. |
|
|
Term
|
Definition
-Variables must be continuous
-Restricted to linear relationships
-No causation
-Sensitive to outliers |
|
|
Term
|
Definition
-Show distributions (i.e. bimodality)
-Shows patterns in data
-Organizes data into bins |
|
|
Term
|
Definition
-Hard to eyeball summary statistics like the mean and sd
-Easily manipulable
-Difficult to interpret when N is small |
|
|
Term
|
Definition
-Provides quick, 5 number summary of the data (minimum, Q1, median, Q3, maximum)
-Easily identify outliers
-Easy to do side-by-side boxplots to compare groups |
|
|
Term
|
Definition
-Doesn't show distribution
-Hides patterns like bimodality and clusters
-Skewed distributions make for an uninformative boxplot |
|
|
Term
|
Definition
Same thing as a bell curve. Describes lots of real world phenomena. The normal density function requires mean and standard deviation.
Features:
-Symmetric around the mean.
-Mean, median, and mode are equal.
-Area under the curve is equal to 1.
-68% of the area is within 1 SD of the mean
-95% of the area is within 2 SDs of the mean |
|
|
Term
|
Definition
68-95-99 rule
How to use the empirical rule:
Use with normal distribution.
Say the given mean is 30 and the sd is 5.
The area within 1 SD of the mean is 35 and 25, so 68% of the x values lie between 25 and 35.
The area within 2 SDs of the mean is 20 and 40, so 95% of the observations lie within 20 and 40. |
|
|
Term
|
Definition
mean > median. Income of the world is an example.
The histogram trails off to the right |
|
|
Term
|
Definition
| mean < median. The histogram trails off to the left. |
|
|
