Factorial bs anova

13 important questions on Factorial bs anova

What is added in the sum of squares of the model?

Besides the two variables also the interaction between the two variables

What are important values in the spss tables when producing these anovas?

The df of the variable and the error
The f value
Significance and the effect sizes

What are two ways of interpreting interactions?

Graphically
Simple main effects
  • Higher grades + faster learning
  • Never study anything twice
  • 100% sure, 100% understanding
Discover Study Smart

When are there no main effects and no interaction in interaction graphs?

When they are horizontal right next to eachother and do not cross

When is there no main effect for one variable but there is for the other and no interaction?

When the lines are apart, horizontal and parallel and do not cross

When are there no main effects, but there is interaction?

When they switch places and cross

Why do you need to transform data?

To deal with outliers
Correcting problems with normality (skewness for example)
Correcting problems with the assumption of homogeneity of variance (reduce heteroscedasticity)

On what is data transformation based?

It does the same to all the scores, so no change of relations between variables
It does change the absolute values or differences (units are changed)

Why use a log transformation?

Make a skewed distribution nearer to normal
Converts value x to power y, which is used for the analysis

How do you perform log transformations?

x_new=10log(X)
x_new- ln(x)

Correct for skew and unequal variances, then the original values are close to 1, zero or a negative a constand needs to be added:
x_new=ln(x+15)

What are square root transformations when do you use them?

x_new=sqrt(x)
Can correct for positive skew and unequal variances
Add a constant in case of negative values in dataset

What is reciprocal data transformation?

x_new=1/x
Correct for positive skew and unequal variances
Rerverses the scores large becomes small, small becomes large, solution:
x_new=1/(xmax-x)

What are reverse score transformations?

x_new=xmax-x
x_new= x(max+1)-x
Can correct for negative skew, if followed up by e.g. log transformation and reverse scores back again

The question on the page originate from the summary of the following study material:

  • A unique study and practice tool
  • Never study anything twice again
  • Get the grades you hope for
  • 100% sure, 100% understanding
Remember faster, study better. Scientifically proven.
Trustpilot Logo