Please enable JavaScript.
Coggle requires JavaScript to display documents.
BIVARIATE DATA - Coggle Diagram
BIVARIATE DATA
DATA
MULTIVARIATE
Multivariate data refers to a set of data where multiple (more than two) pieces of data are analysed. For EG: Height, age, weight, and gender. The purpose of this is to determine correlations between multiple pieces of data, clarifying relationships, and for more accurate representations (since there are more I.V that affects the D.V.
-
BIVARIATE
When a pair of variables from a data set has at least two different variables from the same subject. The purpose of bivariate data is to be able to analyse, or compare, the relationship between the two pieces of data. Often, bivariate data is visualized via a scatterplot graph.
-
-
REGRESSION
-
Regression refers to the statistical method used to find a model for bivariate data. The purpose of regression is to allow us to see how one variable affects another, or how variables contribute to a particular outcome.
PREDICTIONS
EXTRAPOLATION
Extrapolation refers to a type of estimation using existing data and to go beyond that, to make an assumption that the trends will continue, or the current methods remain applicable. In other words, similar to an educated guess or a hypothesis.
-
INTERPOLATION
Interpolation is the statistical process in which unknown values are estimated, that fall between the known values. For example, when 2 points are plotted on a graph, a line can be drawn between the two points. This allows for an estimation of the unknown values in between, as it lies midway.
-
VARIABLES
RESPONSE VARIABLES
The variable in an investigation which changes, or 'responds' by changing due to the explanatory variable. It is measured and influenced. Normally it is represented on the y-axis, like a dependent variable.
-
EXPLANATORY VARIABLES
The variable in a investigation which is controlled, or doesnt change. It is 'fixed', and therefore is able to explain possible changes seen in the response variable. The Explanatory variable is represented on the x-axis of a graph, like an independent variable.
-
CORRELATION
The statistic measure which is able to express two variables that may change together at a constant rate (linearly). The purpose of this is to analyse simple relationships without having to explain 'cause and affect'.
-
TYPES OF VARIABLES
CATEGORICAL VARIABLES
A categorical variable refers to a variable which has a limited or fixed number of possible values, assigning each individual to a corresponding particular group. For EG, categorial values include sex, race, age, educational level, etc.
NUMERICAL VARIABLES
CONTINUOUS VARIABLES
Continuous variables are values obtained by counting. They have infinite numbers of values, between 2 values, therefore, counting continuous values would quite literally take forever. It is continuous as the name suggests. An example of continuous variables include date and time.
-
DISCRETE VARIABLES
Discrete variables are values obtained by counting (in whole numbers). Discrete values are countable in a finite amount of time, for example: counting money, or how many pebbles there are in a jar.
-
CAUSATION
Causation in statistics refers to a correlation between variables. For example, it is a casual relationship between variables, indicating that one event is the result of another.
