Re: Tests and measurement

From: Chris Taylor (christ@ebi.ac.uk)
Date: Thu 11 Mar 2004 - 10:22:18 GMT

  • Next message: Chris Taylor: "Re: War Gods?"

    Francesca S. Alcorn wrote:

    > Chris said:
    >
    >> Thanks Frankie and Lawry. I must admit I'm surprised to hear that this
    >> is not a big area of study, although there is that 'survey survey'
    >> problem which is rather thorny.
    >
    >
    > I think you could find quite a bit of info out there. Try googling
    > "survey instrument design". Or just "survey design". "Survey design"
    > at Amazon gives you quite a few titles.

    Yeah. I'm gonna go for it :) I'll be a pro yet.

    >
    >> I found the idea of just discounting the answers of those who are
    >> suspected to be 'cheating' interesting -- kind of like the 'bad day to
    >> predict the weather' principle that comes out of complex short term
    >> climate models, cos that's all they can manage given the inherent
    >> difficulties.
    >
    >
    > Yeah, I thought it was pretty sneaky, probably why it stuck in my head
    > all these years.
    >
    >
    >> Just to put it in context; I'm predominantly a data modeller and user
    >> reqs gatherer these days, and getting people to (a) be completely
    >> honest about how they and others work, and (b) stop saying they're in
    >> favour of good things, and against bad things, is a bugger :)
    >
    >
    > Now why on earth would you want people to say that they are against good
    > things? :)
    >
    > Is this something you can fix by using value-neutral language? Or ask
    > them what would be the advantages/disadvantages of "A"? Might at least
    > help you reframe the questions. Or (trying desperately to bring this
    > back to memetics) transplant the idea from one memeplex to another.

    Good advice -- that's the kind of thing I do when dealing with eager developers who want to run with an idea, when I don't want them to :) I need to try that balanced thing in the wild though -- up till now I've been using the 'I am an idiot -- knock down my straw man' approach
    (which isn't too bad but it's a fine line between 'idiot' and idiot.

    > Can you guarantee the anonymity of the respondents?

    Almost never, which is a huge issue actually. Something I could try to address...

    Cheers, Chris.

    > frankie
    >
    > ===============================================================
    > This was distributed via the memetics list associated with the
    > Journal of Memetics - Evolutionary Models of Information Transmission
    > For information about the journal and the list (e.g. unsubscribing)
    > see: http://www.cpm.mmu.ac.uk/jom-emit
    >
    >

    -- 
    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
      Chris Taylor (christ@ebi.ac.uk)
      MIAPE Project -- psidev.sf.net
    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    ===============================================================
    This was distributed via the memetics list associated with the
    Journal of Memetics - Evolutionary Models of Information Transmission
    For information about the journal and the list (e.g. unsubscribing)
    see: http://www.cpm.mmu.ac.uk/jom-emit
    


    This archive was generated by hypermail 2.1.5 : Thu 11 Mar 2004 - 10:35:02 GMT