Time Series – A set of observations (data that are classified according to a unit of time. For example: monthly sales. See also: Time series Analysis
Time Series Analysis – Techniques that are employed in an attempt to comprehend alterations in time series. This may result in an improvement of predictive techniques. See also: Time Series
Trichotomy – Classification into three parts, three classes: “yes/no/don’t know;” expensive/cheap/ right price.”
Two-by-two Frequency Table – A tabular representation of data that occur in a double dichotomy. For example: each sample element is A or not-A and contains X or not -X. See also: Double dichotomy/Table
Vital Statistics – The gathering, analysis and interpretation of numerical data pertaining to people, specifically figures for birth and death.
Asymmetrical Distribution – A distribution in which no central value exists. Expressed as a formula this means that: f(x-a)=f(a-x) f(x) is the frequency function.
Bernoulli Distribution – Syn: Binomial Distribution See: Binomial Distribution
Bimodal – A frequency distribution with two modes. The bimodal distribution in a graph has two peaks. For example: the birth figures since 1900 indicate that there have been two birth peaks following both of the world wars. See also: Unimodal/Frequency distribution/Mode
Binomial Distribution – Syn: Bernoulli Distribution A probability distribution. The probability (P) that ® appears in (N) independent trials is: (N/R)Q-RpR, where Q=I-P In this distribution of probability something is either P or not-P. There are no other possibilities. For example: head or tails, pregnant or not. See also: Hypothetical population/Trial/Probability Distribution
Bivariate Distribution – A distribution consisting of two variables, either composed as such or coincidental.
Categorical Distribution – The classification of data into categories according to a qualitative description and not according to a numerical variable. For example: sex (male/female) See also: Category
Frequency Distribution – involves the collection of data from a large group (individuals, objects) followed by the classification f these data in sequence. The purpose of the frequency distribution is to render a large amount of data accessible and comprehensible. For example: in a sample, 70 persons own 5 radios, 90 own 4, 200 own 3,400 own 2, and 50 persons each own 1 radio. When such a distribution is expressed in table form, it is termed a frequency table. See also: Histogram/Table?Frequency?Frequency table
J-shaped Distribution – An extreme form of asymmetrical frequency distribution. The highest frequency occurs at the beginning (or end) of the frequency group, and a decreasing or increasing frequency is found elsewhere. The shape of this distribution corresponds, approximately, to the letter “J” or an inverted “J”. A frequency distribution of traffic accidents is J-shaped. By far the greatest number of victims are young. See also; Skewed distribution/Asymmetrical Distribution
Multivariate Distribution – The simultaneous distribution of a number of P variables (P<1) or equivalent; the probability of P variables.
Non-parametric Statistics – Form of statistics that makes no assumptions concerning population distribution or concerning any constant in the population. See also: Parameter
Norm Setting – Establishing terms of reference to which subsequent data or decisions can be compared. For example, in a large-scale survey, the percentage of TV viewers who are able to recall a particular commercial is determined. That percentage is taken as the norm. In later small-scale surveys, the extent to which certain commercials are remembered is compared to the original figure. See also: Norm
Normal – 1. According with a certain norm
- Falling within a certain category, e.g., if 99% of the population has a telephone, then the remaining 1% is not normal.
- Accepted behavior pattern (free from mental disorder). See also: Norm
Normal Distribution – Syn: Normal Probability Cure A statistical distribution that possesses special characteristics. A normal distribution states that there are approximately an equal number of people, data or objects positioned on both sides of the average. For example: there are as many obese people as there are very thin people. While the number of ordinary “average people” is very large, the number of exceptional (obese or thin)people is considerably less. Graphically represented, the normal distribution is bell shaped. A great many human characteristics appear to be distributed normally.