Term
|
Definition
| science of collection and organizing data to then draw conclusions based on data |
|
|
Term
|
Definition
| descriptive, multivariate, inferential |
|
|
Term
|
Definition
| summary of large data sets to be put into an easy to read manner |
|
|
Term
|
Definition
| allows comparisons of factors by isolating the effect of one variable from others |
|
|
Term
|
Definition
| statements of a population based on a sample |
|
|
Term
|
Definition
| nominal, ordinal, interval and ratio |
|
|
Term
| Characteristics of nominal data |
|
Definition
| lowest level, must be exhaustive, mutually exclusive, only classified one way, provides names or labels (race, gender), can only be one of something |
|
|
Term
| Characteristics of ordinal data |
|
Definition
| exhaustive, mutually exclusive, exhibits a degree of difference, the degree of difference indicates order/rank, ordering distance has no meaning (good, better, best/agree, disagree, neutral) |
|
|
Term
| Characteristics of interval |
|
Definition
| assumes all items on scale have equal units of measurement, distance between categories have meaning, logical distances (temp readings, IQ score) |
|
|
Term
|
Definition
| has all the characteristics of the other levels plus a true "zero" point. highest level of measurement. (income, weight, time, age) |
|
|
Term
|
Definition
| lists the number or frequency of scores/labels for each case |
|
|
Term
|
Definition
| summary statistic with limited information. Highest score-lowest score= range |
|
|
Term
|
Definition
i= range/"n" of desired intervals
example:i=12-1/4=11
11/4= 2.75
i=2.75 |
|
|
Term
|
Definition
|
|
Term
|
Definition
|
|
Term
|
Definition
| most frequent score/label |
|
|
Term
|
Definition
| sum squared (S^2)deviations of each score from the mean divided by the total number |
|
|
Term
|
Definition
| average distance each data item is away from the mean of the all the data items. Provides comparison insight between two distributions. |
|
|
Term
|
Definition
| the spread of scores weighted to one side of the mean. |
|
|
Term
|
Definition
| bell shape, the curve doesn't touch the x-axis, curve is symmetrical. Mean, median, mode are equal. |
|
|
Term
|
Definition
| unimodial scores weighted to the left. Hump to the left, tail to the right. Mode is the largest, followed by the median, and then mean. |
|
|
Term
|
Definition
| unimodial weighted to the right (hump to the right, tail to the left). Mean largest value, then median, then mode. |
|
|
Term
|
Definition
| used to standardize measure for comparison between two different items of size. (compare car theft rates between two cities with significant difference in population size). |
|
|
Term
|
Definition
example:
60 officers / 15 sgts
(15/75)= .20 portion of sgts
(60/75) = .80 portion of officers |
|
|
Term
|
Definition
| subtract old value from new value then divide by old value then multiply by 100. |
|
|
Term
|
Definition
| measures the relative fit or degree of association between two variable or 3 or more variables to determine the quality of regression line or reliability of prediction. |
|
|
Term
| Pearson's product moment coefficient |
|
Definition
=r
operates within range 1.0, the closer to one the stronger the value. Zero = no correlation. |
|
|
Term
|
Definition
| seeks general pattern evident in observation relationships through pattern recognition and placement on a trendline. |
|
|
Term
| Regression analysis and correlation analysis are similar except regression |
|
Definition
| locates the best line through the points. |
|
|
Term
|
Definition
| is one in which data points, x & y coordinates, make more a less straight line. |
|
|
Term
|
Definition
| is one which the data points, x & y data, make some type of curve shape that can be expressed by an algebraic formula. |
|
|
Term
|
Definition
| to find the best straight line through the points and has the smallest variance, reflecting gaps between the actual placement of the data points on the graph. |
|
|
Term
|
Definition
| is multivariate in nature when two or more variables (usually of ratio or interval type) are examined together with the focus on observable relationships or patterns of association, between or among the variables. |
|
|
