+ Reply to Thread
Results 1 to 2 of 2

Thread: Alternative for a multi-level analysis

  1. #1
    Points: 6, Level: 1
    Level completed: 11%, Points required for next Level: 44

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Alternative for a multi-level analysis




    Hi all,

    For my criminology thesis I am analysing police interviews and would like to research the effect (the presence of) an attorney has on whether or not a suspect gives a statement. The statement will be a dichotomous variable (0 = remaining silent 1 = giving a statement). My data is nested in the sense that I have a couple of variables on the level of the interrogation (1st, 2nd, 3rd interview and such) and on the level of the suspect himself (age, criminal career etc).

    The best method, as far as I know, is doing a multi-level logistisch regression analysis tot determine whether the attorney has any influence. The problem is that I don't have the correct knowledge or experience to perform such an analysis and I don't know anyone who can. Also, I am working with SPSS and I've understood that SPSS can only do linear multi-level regressions.

    Is there any alternative method I could use?

    Also, the problem with performing a 'normal' logistic regression without taking the hierarchy into account is that the variables from level 2 will be disaggregated to level 1. Will this problem be solved if I only take (for instance) the first interview from every case and use that as input for the analysis? Most subjects are interviewed multiple times.

    Thanks in advance for any advice. Should you have any questions, then please let me know.

  2. #2
    Omega Contributor
    Points: 38,303, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,993
    Thanks
    397
    Thanked 1,185 Times in 1,146 Posts

    Re: Alternative for a multi-level analysis


    How much time do you plan to take in doing this (months, years)?


    Don't use SPSS, then.


    Yeah, an option is to just look at time one data, but this will change your questions.
    Stop cowardice, ban guns!

+ Reply to Thread

           




Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats