stuck on regression

Home Forums Default Forum stuck on regression

Viewing 2 posts - 1 through 2 (of 2 total)
  • Author
  • #2522

    I have an assignment due next week, and don’t understand what I am supposed to do.


    A researcher wants to know how family income and IQ predict students’ frequency of delinquency. She collected data from 20 students and the data are recorded in “Assign6 Delinquency.savI attached the SPSS file.


    Part I. Assume that the relationships among delinquency, IQ, and income are linear. Regress delinquency on IQ, income (in thousands $), and the interaction between IQ and income, with both income and IQ centered around their means. Note that in the SPSS data file, IQ and Income given are the original scores. Therefore, some transformation is needed here to get the centered variables. 

    1. Conduct significance tests for the overall regression model and each predictor. Are they statistically significant at .05 level? How do you know? 
    2. Write down the regression function and interpret each regression coefficient. 
    3. Find the regression equation of delinquency on IQ at different income levels. (Note: If you are not sure how to solve this problem, see the example at the end of this document.)
    4. What is the regression equation if you add gender. should gender be added, why?

    The instructor responded to my question of (Do we need to transform the scores into Z scores and then run regression? :

    Before doing anything, you  need to transform the regression equation so that it accommodates original scores, as opposed to deviation scores. this can be done by replacing

    IQC = IQ – 109  (109 is the mean IQ)

    IncomeC = Income – 43 (43 is the mean income in thousands)

    [What is IQC? why did she subtract the mean/ is this to be done for all scores and made into a new predictor variable? What is the “C” designate after the new variable? (The DV is student delinquency rate)then once new variables are made, put those in the regression SPSS as IVs? Am I supposed to do the raw score method using r, Sx and Sy? I am not sure of the multiple regression formula, just simple.

    Regression formula is:

    My thoughts:

    Y’ = intercept + b1X1 + b2X2

    Her example that was attached to the assignment was from a different data set with deviation scores, and there is an extra group on the end which I don;t understand:

    Y’ = 25.89 – 0.262AgeC + 0.973ExerC +0.047AgeC*ExerC

    I assume I should make my regression formula:

    Y’ = Intercept + x.xxxIQC+ x.xxxIncomeC + x.xxxIQC*Income C

    If so, where do I get the b values from SPSS-If I transform all IQ and Income variables by subtracting their means, do I use the b values that  SPSS generates?


    Never mind I got my answer

Viewing 2 posts - 1 through 2 (of 2 total)
  • The forum ‘Default Forum’ is closed to new topics and replies.