Terminology

Dichotomous variable

A dichotomous variable is one that has only two options, one meaning true and the other false (these can be denoted as 1/0, “”/”yes”, “true”/”false” and so on).

An example of a non-dichotomous variable is one that has the likert scale, “

Likert scale

A typical Likert scale is Strongly disagree, Disagree, Neither agree nor disagree, Agree, Strongly agree.

NETs

The combination of several answers into one, such as “Strongly agree” and “Agree”, is often called a NET. In Dejuice, NETs can be created by editing categorical variables.

NPS score

A scale from 0-10 rating how likely a user would be to recommend a service or business. NPS scores are stored as numerical variables and single-choice variables (sometimes the latter has to be created with Dejuice). The single-choice variable will have three categories, detractors (0-6), passives (7-8) and promoters (9-10), created from the numberical variable.

RIM weighting

RIM weighting (Random Iterative Method) is an algorithm commonly used to make sure a survey data sample accurately represents a given population. Dejuice supports RIM weighting with complex weighting schemas

Weighting schema

A weighting schema is a set of instructions on how to weight sample data to better represent a population. Common examples are schemas for weighting populations in specific countries, where the schema indicates what the true demographic looks like, e.g. that women and men are 51% and 49% of the population.