Analyses per outcome types

Analyses per outcome types

This is the landing page for topics related to analyzing different types of outcomes in your ILD study.

The type of outcome determines what modeling approach is appropriate. The outcome may be a continuous-valued variable (e.g., using a slide-ruler), a discrete-valued variable (e.g., using a Likert scale), or a categorical variable (e.g., using distinct, mutually exclusive answer categories). In addition, the distribution of your construct can influence which models are suitable and how results should be interpreted.

Below we have specified multiple articles where you can read more about types of outcome.

Think more about your outcome type

Different types of outcome

The type of outcome variable influences assumptions about distribution, link functions, and estimation methods.

[Discrete outcomes]
[ARMA models versus discrete variables]
[Categorical outcomes]
[Markov models for categorial variables]

Number of outcomes

You may focus on a single outcome variables using a univariate model (possibly with multiple predictors), or on multiple outcome variables simultaneously using a multivariate model.

[ARMA versus VARMA]
[Network models]
[SVAR]

Think more about the distribution of your construct and measurements

The distributions of variables imply what scores you should expect for the process you are studying.

[Discrete or continuous constructs]
[Types of distributions]
[Distributional characteristics]
[Measurement levels]
Floor and ceiling effects