Term
|
Definition
| the collection of items we want to better understand, and which we wan to draw a conclusion |
|
|
Term
|
Definition
| summary calculations of the entire population set of data |
|
|
Term
|
Definition
| a portion of the population |
|
|
Term
|
Definition
| a variable, since the value it takes on will depend on the sampling chance of which items are chosen for the sample. |
|
|
Term
|
Definition
| the extent to which a sample statistic mis-estimates a population perameter, due only to "luck of the draw" (the random chance of sample selection). error caused by taking a sample that does not accurately portray the population as a whole |
|
|
Term
|
Definition
| Actual human errors, encompassing all sources other than random sampling chance, that lead a statistic to mis-estimate the corresponding population perameter. i.e. innaccurate measurements, ambiguous terms, falsified responses, non-responses, lost data, not using genuine random selection from the population, computational errors |
|
|
Term
|
Definition
| serves only to distinguish one outcome from another (we can only tabulate "how many" fall in each catagory) Ex. color of shirt, brand of car |
|
|
Term
|
Definition
| also conveys meaningful "rank order" (we not only tabulate frequencies, but also find percentiles) Ex. top 20 college football rankings, Consumer Reports product ratings |
|
|
Term
|
Definition
| also allows us to interpret differences along the scale, i.e., a particular difference has the same meaning anywhere along the scale (we can tabulate frequencies, find percentiles, find a mean) Ex. # of children you have, current credit card balance |
|
|
Term
|
Definition
a method of selecting a sample such that: 1) every item in the population has the same chance of being selected 2) every combination of items from the population has the same chance of being selected |
|
|
Term
|
Definition
| The list/collection of items from which the sample is actualy chosen; i.e. the collection of items that could possibly appear in the sample. |
|
|
Term
|
Definition
| values representing the amount of data that lies beneath 25%, 50% and 75% of a sampled population (1st, 2nd, 3rd quartile, respectively) These numbers are medians within the data, therefore the 2nd quartile may represent the point in the data at which ten numbers lie above and below that point in the set |
|
|
Term
|
Definition
| A data point which is rather far removed from the bulk of the rest of the data. |
|
|
Term
|
Definition
| takes the numeric value of every data point into consideration; hence can be inluenced by extremes |
|
|
Term
|
Definition
| the middle data item's value, after the items are put in increasing or decreasing order |
|
|
Term
|
Definition
| the square root of the variance; hence will always be in the original units of the data measurements |
|
|
Term
| Statistically Independent Variables |
|
Definition
| pattern in which outcomes for one variable is the same at every level of the other variable |
|
|
Term
| Statistically Related Variables |
|
Definition
| statistical relationship in which values have comparative probability patterns (not necessarily one variable "affecting" the other for a statistical relationship to exist) |
|
|
