Answers to the checkpoints of chapter 7
- Hypotheses about relationships (correlations) between continuous variables.
- The correlation is moderately strong.
- -1 < r < 1.
- The more you cycle, the fitter you are. So there is a strong positive correlation.
- A correlation is two way; an effect has a direction and it is one way.
- You would use regression analysis if you want to test the effect of one or more variables on a dependent variable and the variables used are continuous.
- The population formula has a residual (referred to as “error” in SPSS) to indicate that your prediction can never be 100% correct.
- The constant in the regression equation is the starting point of the regression line. The point on Y where X = 0.
- The regression coefficient gives the direction of the regression line.
- The amount of variance in a dependent variable that is explained by the independent variables in the model.
- You can use explained variance to check how well your model “fits.”
- If R2 = 0.60, the model is strong.
- Observed (y) value minus expected (y) value.
- Linear effects.
- The test statistic for this is t.
- Two-tailed, because there is no direction to the assumption.
- The bèta coefficient (β coefficient).
- The difference between the b coefficient and the β coefficient: the b coefficient is unstandardized, each is based on its own scale. This means that b coefficients cannot be compared in terms of strength. β coefficients, on the other hand, can be compared. Standardization allows for this.