3 things an experiment must include
Levels
Each variable can have levels (=values). For example if a variable is the groups of participants the levels can be the experiment group and the control group.
Conditions
The amount of levels to each variable gives us the amount of conditions by multiplying the levels of the different variables
Internal validity
When the value of the dependent variable can surely be attributed to the independent variable. i.e. a measure of how well the experiment was executed. An experiment without confoundings has internal validity.
External validity
How well the findings are applicable on a larger scale.
Three conditions (villkor) for causal inference
Confoundings
If another independent variable is affecting the independent variable you are looking at.
Control techniques to eliminate confounding
-Keeping things constant
-Balancing