6 LMM: Inputs

6.1 SPSS: Menu

Implementation in SPSS is similar to what we discussed earlier. Note that we are obtaining the estimation using Maximum Likelihood (ML) method. Generally, this helps to undertake model selection, particularly those of fixed effects.

  • Fixed

    • Include the interaction term of Group * Time into Fixed Effects dialog box (in Model section)

    • Select both main effects, choose the interaction term from drop-down menu, and add the item into the model clicking Add.

    • This means that the model includes both the main effects and interaction term of Group and Time.

  • Random

    • Select Variance Components, check Include intercept.

    • Selection ID and move it to Combinations in Subject Groupings

  • Estimation

    • Select Method as Maximum Likelihood (ML)

  • Statistics

    • Select appropriate Model Statistics required for outputs


6.2 SPSS: Script

MIXED SBP BY Sex Group WITH SBP0 Age BMI Time 
  /CRITERIA = DFMETHOD(SATTERTHWAITE) CIN(95) MXITER(100) MXSTEP(10) 
    SCORING(1) SINGULAR(0.000000000001) HCONVERGE(0, ABSOLUTE) 
    LCONVERGE(0, ABSOLUTE) PCONVERGE(0.000001, ABSOLUTE) 
  /FIXED = Sex Group SBP0 Age BMI Time Group*Time | SSTYPE(3) 
  /METHOD = ML 
  /PRINT = G R SOLUTION 
  /RANDOM = INTERCEPT | SUBJECT(ID) COVTYPE(VC)


6.3 Summary

The analysis of the data considering the patient as a random effect.