R help - Jul 2010 - Question about food sampling analysis

Greetings to all, and my apologies for a question that is mostly about
statistics and secondarily about R.  I have just started a new job that
(this week, apparently) requires statistical knowledge beyond my training
(as an epidemiologist).

The problem:

- We have 57 food production facilities in three categories
- Samples of 4-6 different foods were tested for listeria at each facility
- I need to describe the presence of listeria in food (1) overall and (2) by
facility category.

I know that samples within each facility cannot be treated as independent,
so I need an approach that accounts for (1) clustering within facilities and
(2) the different number of samples taken at each facility.  If someone
could kindly point me towards the right type of analysis for this and/or its
associated R functions/packages, I would greatly appreciate it.

Many thanks,

Sarah

	[[alternative HTML version deleted]]

Hi Sarah,

We regularly undertake work in the food sector and have developed many
custom built solutions. To be more specific, the statistics we employ is
that of sensory analysis and we regularly use the sensominer package in
R.

Regards,

Richard Weeks

Mangosolutions
data analysis that delivers


Mail: rweeks at mango-solutions.com
T: +44 (0)1249 767700
F: +44 (0)1249 767707
M: +44 (0)7500 040365

Unit 2 Greenways Business Park
Bellinger Close
Chippenham
Wilts
SN15 1BN
UK
 

-----Original Message-----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org]
On Behalf Of Sarah Henderson
Sent: 12 July 2010 22:42
To: R List
Subject: [R] Question about food sampling analysis

Greetings to all, and my apologies for a question that is mostly about
statistics and secondarily about R.  I have just started a new job that
(this week, apparently) requires statistical knowledge beyond my
training
(as an epidemiologist).

The problem:

- We have 57 food production facilities in three categories
- Samples of 4-6 different foods were tested for listeria at each
facility
- I need to describe the presence of listeria in food (1) overall and
(2) by
facility category.

I know that samples within each facility cannot be treated as
independent,
so I need an approach that accounts for (1) clustering within facilities
and
(2) the different number of samples taken at each facility.  If someone
could kindly point me towards the right type of analysis for this and/or
its
associated R functions/packages, I would greatly appreciate it.

Many thanks,

Sarah

	[[alternative HTML version deleted]]

______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
LEGAL NOTICE
This message is intended for the use o...{{dropped:9}}

Dear Sarah,

[snip...]
"I know that samples within each facility cannot be treated as independent,
so I need an approach that accounts for (1) clustering within facilities
and"

You could just use lm() & some planning. The data from within a specific
facility can be fit with a model to generate parameters that are compared
between facilities. Not to practical though - assuming the 57 production
facilities each have their own analytical lab, you'll have 57 different fits
to get your parameters from to use in your between test. Questions about
dependent data are fairly common, so it should be relatively straight
forward to get a solution and/or idea for a suitable package from the
archives.

"(2) the different number of samples taken at each facility."
It's a waste of time to worry about that. You'll be comparing aggregate
values between groups, and you'll have too few data-points within a group to
detect within effects... 

[snip...]

Sincerely,
KeithC.