Term
|
Definition
| Descriptive statistics is the study of data to develop an understanding of the underlying phenomenon driving observed processes including mean, median, variance, range etc. |
|
|
Term
|
Definition
| This is the use of statistical data and descriptive statistics to forecast future events or provide probabilities of their occurrence. |
|
|
Term
|
Definition
| A population is the largest collection of objects belonging to a group, defined by a common characteristic. |
|
|
Term
|
Definition
| A sample is a sub-set or sub-group of a population. |
|
|
Term
|
Definition
| A parameter is a measure that describes some feature of a population. |
|
|
Term
|
Definition
| A sample statistic is a measure that describes some feature of a sample. |
|
|
Term
| 4 Types of Measurement scales |
|
Definition
1. Nominal
2. Ordinal
3. Interval
4. Ratio |
|
|
Term
|
Definition
| This is the weakest and represents groups of objects that share a common characteristic, such as, number 1 representing boys between the ages of 5 and 10; number 2 representing girls between the ages of 7 and 12 and so on. There is no particular order to the numbering system, and it does not represent a ranking, |
|
|
Term
|
Definition
| This is the next weakest scale and rank orders objects. For example, a dinner at a restaurant may be ranked by its patrons in one of five ways: very poor, poor, fair, good, and excellent |
|
|
Term
|
Definition
| This is stronger than the ordinal scale and in addition to creating a ranking system like the ordinal scale, it also ensures that the difference between any two adjacent ranks is the same across all members. In other words, the difference between a poor and fair dinner would be the same as the difference between a good and an excellent dinner |
|
|
Term
|
Definition
| A ratio scale is the strongest. In addition to having all the properties of the other three scales, it also has a natural zero. If any value is doubled, the characteristic being measured is also doubled. Meaningful addition, subtraction, multiplication and division (ratios) operations can be conducted on this scale. |
|
|
Term
|
Definition
| A representation of data in a tabular format such that possible outcomes are divided into convenient intervals of equal lengths, and are arranged sequentially in ascending order with the number of observations corresponding to each outcome denoted against each outcome. |
|
|
Term
| Relative Frequency Distribution |
|
Definition
| A representation of observations in each interval as a fraction of the total (%), computed by dividing the observations in each interval by the total number of observations. |
|
|
Term
|
Definition
| A graphical representation of a frequency distribution for a quick visual display and impact. The y-axis represents the observations or their frequency and the x-axis represents different intervals. |
|
|
Term
|
Definition
| Generated by connecting the midpoints of each vertical bar in a histogram with a straight line to create an impression of continuity. The solid line connecting all midpoints is called the frequency polygon. |
|
|
Term
|
Definition
| Weighted mean is different from arithmetic mean in that instead of simply adding all the observations and dividing the sum by the number of observations, the observations are first assembled into a relative frequency distribution and then multiplied by their respective relative frequencies (or weights) before being added. |
|
|
Term
|
Definition
| Harmonic mean is the inverse of the arithmetic average of the inverse of all observations. |
|
|
Term
|
Definition
| Mean absolute deviation measures dispersion by taking into account all observations. It measures the deviation of each observation from the distribution mean, and then computes the arithmetic average of all such deviations, after taking their absolute (|Y|) values, i.e., ignoring their signs. |
|
|
