Suppose that a researcher has used a short form of a life satisfaction survey wi

Question

Suppose that a researcher has used a short form of a life satisfaction survey with 10 items and uses that to predict effective parenting. (Let's assume that a generally positive relationship is present in the data.) She calculates a regression to test this hypothesis. Suppose, though, that the researcher is interested in estimating what the unstandardized regression weight might look like if the long form of 20 items was used and simply multiplies each life satisfaction survey by 2 and reanalyzes the data using this new scaling of the short form. Which of the following will be different in the two regressions? (Check any which will probably be different)

- Unstandardized regression weight
- Intercept
- Standardized regression weight
- F test of statistical significance

Explanation / Answer

Only Unstandardised regression weights will be different in two cases because we are multiplying them with 2 .

An Unstandardised regression weight tells, if i change the independent variable by one unit, keeping other variables constant, what is the change in dependent variable.

Let us take a simple example :

Credit Score= Intercept + Unstandardised Regression weight * Salary

Now, Weight says , if I increase Salary by one unit , Credit score should increase by Weight units.

But, if I multiply weight by 2, and then increase Salary by one unit, Credit Score will increase by 2 weight units.

Standardised Regression weights remain unchanged, beacuse these do not depend upon units of variables.

Intercept says, what is the value of dependent variable, when all independent variables are equal to zero, which remains independent of weights.

F test, tells the overall significance of predictors in model, which also wont change if all the variables are multiplied by 2.

Related Questions

Navigate

• Browse (All)
• Browse S
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.