NUMBAT OER - Open Educational Resources

How to do the test:

  1. State your Null Hypothesis in the form:
    "The dependent variable is not related to the independent variable in a linear fashion"
  2. Choose a Critical Significance Level (α: alpha):
    This is typically α = 0.05.
  3. Calculate the test statistic:
    Use a spreadsheet to calculate your results, or access a helpsheet for SPSS and check here for data entry into SPSS.
  4. Reject or accept your Null Hypothesis.
    Either
    1. Compare the calculated statistic with critical values (Critical values table):
      • If F ≥ Fcritical → reject H0 → significant result.
      • If F < Fcritical → accept H0 → non-significant result.
    2. Look at the calculated probability (P value):
      • If P ≤ α → reject H0 → significant result.
      • If P > α → accept H0 → non-significant result.
  5. Report your results:
    Plot your data using the appropriate graphs.
    (Regression test: Fdf regression, df error = …, P = …)
  6. If your regression test is significant, you can use your model for prediction, using a linear model of the form:

    y = bx + c

    (where y is the dependent variable, x is the independent variable, b is the slope of the line and c is the constant or intercept)

    Use the coefficient of determination, R2, to indicate how much of the variability in y is explained by the regression.

    View an example report.