Factorial Designs

in this instruction video we will discuss the factorial design in this design two or more independent variables also known as factors are investigated simultaneously as with other randomized experimental designs this design can be represented in a scheme in this scheme x and y are two factors or independent variables the entire design contains every possible combination of the vectors factorial designs like these are also called 2x2 designs because each vector has two levels let's consider an example about investigating if a new drug reduces migraine attacks we could create three conditions that differ in the dosage of the drug low medium and high and now we are able to investigate the effect of the vector dosage the independent variable on the number of migraine attacks the dependent variable for example does a higher dosage of the new drug reduce the number of migraine attacks the line in this graph simply reflects the average migraine attacks among the participants with the same condition of dosage so these dots represent the effort just per dosage and are being connected to visualize the effective dosage on migraine attacks to see if the average migraine attacks decreases or increases or stays the same across conditions in this example there is only one independent variable so one vector namely dosage but this simple design could be extended by adding a second vector for example gender the effect of two or more factors is being investigated we refer to the design as factorial design we could assign men and women to each of three dosages we would end up with six Commission's men you receive a low medium or high dosage and women who receive the low medium or high dosage in addition to the effective dosage we are also able to investigate the effect the second sector gender we could for instance investigate our hypothesis that women suffer more from migraine attacks than men not only can we investigate the effect of dosage and gender but we can also investigate the combined effect of the two vectors we can see if a higher dosage of the drug is more effective in reducing the number of migraine attacks for women as compared to men the effects of the fixer separately are referred to as main effects the combined effect of the vectors is called the interaction effect in this example we are dealing with a two-way interaction because the effect combines two vectors dosage and gender the combined effect of more than two vectors is called a higher-order interaction effect suppose we add diet as a factor with two conditions a normal diet and a no-fun diet that eliminates all chocolate and red wine so now we have three factors dosage gender and diet this requires that each of the six groups is split into half with the participants are being assigned to the normal diet and have to the no-fun diet you can now investigate the main effect of diet is a no-fun diet effective at reducing migraine attacks but we can also look at the two-way interaction between diet and gender maybe the no-fun diet is effective in reducing migraine attacks for men but not for women we could also look at the two-way interaction between diet and dosage maybe a higher dosage is more effective when participants follow a no-fun diet as compared to a normal diet finally we can look at the three-way interaction the combined effects of dosage gender and diet it could be that a higher dosage is effective for women regardless of the diet but maybe for men a higher dosage is effective only when they follow an alpha diet to summarize in a factorial design two or more independent variables are investigated simultaneously factorial designs are very useful because they allow us to investigate not only the main effects of each vector - also the interaction effects

Channel: Onderzoeksmethoden UvA

Views: 63,842

Keywords:

Id: 6o7wyyh8Lu8

Length: 5min 20sec (320 seconds)

Published: Fri Jul 07 2017