Non-independence of observations

[Nikos Ntoumanis]

Dear all,

I would be grateful if someone could offer me advice of appropriate statistical analysis for the following problem. I am involved in a research proposal examining the choice of food in school canteens over a period of a few weeks. For various reasons, we are not  able to code/record the identity of kids; we can only code the meal choice. Obviously, we cannot assume that the repeated observations (i.e., meal choices) within a school are independent. How can we correct for this? I know how this problem can be addressed when student ID can be recorded, but what about when it cannot be recorded?

Thanks

Nikos

[Ryan]

Students are nested within lunch canteens. The pattern of meals
distributed in a day by a school canteen may be correlated with the
pattern of meals distributed during other days by the same lunch
canteen.

Ryan

On Sat, Aug 20, 2011 at 5:13 AM, Nikos Ntoumanis <[hidden email]> wrote:

> Dear all,
> I would be grateful if someone could offer me advice of appropriate
> statistical analysis for the following problem. I am involved in a research
> proposal examining the choice of food in school canteens over a period of a
> few weeks. For various reasons, we are not  able to code/record the identity
> of kids; we can only code the meal choice. Obviously, we cannot assume that
> the repeated observations (i.e., meal choices) within a school are
> independent. How can we correct for this? I know how this problem can be
> addressed when student ID can be recorded, but what about when it cannot be
> recorded?
> Thanks
> Nikos

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[John F Hall]

Is there no possibility of giving each kid a confidential pass number (on a scratch card if necessary) to be given as each meal is ordered?  That way there wouldn’t even be a list to tie names to numbers, but would the kids remember their numbers?

 

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Nikos Ntoumanis
Sent: 20 August 2011 11:13
To: [hidden email]
Subject: Non-independence of observations

 

Dear all,

I would be grateful if someone could offer me advice of appropriate statistical analysis for the following problem. I am involved in a research proposal examining the choice of food in school canteens over a period of a few weeks. For various reasons, we are not  able to code/record the identity of kids; we can only code the meal choice. Obviously, we cannot assume that the repeated observations (i.e., meal choices) within a school are independent. How can we correct for this? I know how this problem can be addressed when student ID can be recorded, but what about when it cannot be recorded?

Thanks

Nikos