Difference between revisions of "An initial path towards statistical analysis"
From Sustainability Methods
| Line 30: | Line 30: | ||
Dependent variable not normally distributed | Dependent variable not normally distributed | ||
| − | + | Type III Anova | |
Dependent variable is count data | Dependent variable is count data | ||
| Line 38: | Line 38: | ||
Random factors | Random factors | ||
| − | + | Generalized linear mixed effect models | |
Only categorical variables | Only categorical variables | ||
| − | + | Chi-Square test | |
Only continuous variables | Only continuous variables | ||
| Line 49: | Line 49: | ||
Non dependent relations | Non dependent relations | ||
| − | + | Correlations | |
Clear dependent relations | Clear dependent relations | ||
| Line 57: | Line 57: | ||
Dependent variable normally distributed | Dependent variable normally distributed | ||
| − | + | Linear Regression | |
Dependent variable not normally distributed | Dependent variable not normally distributed | ||
| − | + | GLM | |
Dependent variable is count data | Dependent variable is count data | ||
| − | + | Poisson GLM | |
Dependent variable is 0/1 or proportions | Dependent variable is 0/1 or proportions | ||
| − | + | Binomial GLM | |
Revision as of 22:56, 10 January 2021
Start here with your data
Do you have several continuous variables without clear dependencies?
Yes-> Multivariate statistics
No-> Univariate statistics
Univariate statistics
At least one categorical independent variable
Categorical and continuous data
Analysis of Variance
Dependent variable normally distributed
More than 2 groups
Type II Anova
Random factors
Linear mixed effect model
Dependent variable not normally distributed
Type III Anova
Dependent variable is count data
Dependent variable is 0/1 or proportions
Random factors
Generalized linear mixed effect models
Only categorical variables
Chi-Square test
Only continuous variables
Non dependent relations
Correlations
Clear dependent relations
Regression models
Dependent variable normally distributed
Linear Regression
Dependent variable not normally distributed
GLM
Dependent variable is count data
Poisson GLM
Dependent variable is 0/1 or proportions
Binomial GLM
Multivariate statistics
