Quantitative 2

Question 1

Regression:

Answer 1

DV = the outcome variable in predicting (y)

IV = using to predict (x)

Regression assumes linear relationship

Regression line acts to minimise the sum of the squared deviation around itself

Question 2

Regression VS Residual

Answer 2

Represents a gain in prediction due to regression the distance between the mean and the regression line good stuff

Represent our failure to predict - the residual or the distant between the regression line and actual score (bad stuff)

Question 3

Anova

Answer 3

Portiton variance attributed to regression and that attributed to residual variance

Divining the regression variance by the residual variance is used to determine significance

Question 4

Multiple Regression

Answer 4

Is a single interval level DV And multiple IV

called ordinary least squares regression because the technique fundamentally minimises squares deviations about the regression line

Question 5

Multiple Regression

Answer 5

Model is significant:

• shown in the Anova table, usually at the top of output.
• if it is significant, then the dig value ( or p - value) will less 0.05

-

Question 6

Trustworthiness:

Answer 6

Regression is robust to minor violation will require work around or different approaches to be employed.

Measure last are unreliable the analysis will be completely meaningless

Question 7

Correlations:

Answer 7

Summaries the degree of linear relationship between two variables.

• 1 = perfect negative relationship

0 = no relationship

+ 1 = perfect positive relationship

Correlations are presented with an effect size (i.e, the correlation coefficient) and a test of significance.