Re: how to measure the amount of stain left in carpet?Where to find device?

     It is interesting that your teacher believes that observations are 
not scientific evidence. The American Association of Textile Chemists and 
Colorists, which is the organization that usually establishes the 
standards for testing like determining staining, uses exactly that method. 
There is a lot of information on their site. One establishes a control, 
that is, a piece of fabric that wasn't stained or cleaned, and compares it 
to the stained and then treated pieces to determine the effectiveness of 
each stain remover. A key point is the training; what is usually done is a 
chenist or colorist (like me) makes a set of standards for technicians to 
which to compare the samples, so that the person who determines what's a 
"good" stain removal and what's not good enough has a clear idea of what 
they're examining.

     I wouldn't use a colorimeter; I'd do exactly what you did -- look at 
the samples, as long as you have a control to which to compare them. That 
would show you which cleaner works best, and if (as sometimes happens) one 
or more of the stain removers also takes some color from the carpet as 
well as the stain. Using a colorimeter would have the "advantage" of 
generating numerical values that you can compare; but unless you and your 
teacher know what the numbers mean and how to determine what's a "good" 
result, it's actually a better test to simply look at the pieces.

The AATCC site may be useful to you: http://www.aatcc.org

---------------

Admin note:

We received the following additional response from Gareth Evans:

Emily,  your teacher is probably right in saying that you need something a
little more accurate than people's opinions about which of your stain
removal treatments give the best results.  This is because we all see the
world a little differently and you may need quite a few volunteer judges to
make the results reliable enough to draw firm conclusions.  The issue of
what makes a trustworthy result is very important in science.  However I
will make some suggestions which may convince you and your teacher that
you have a method of measurement which will make your conclusions
reliable.  Of course you could try to get hold of an optical in
strument like a colorimeter but let's try to find a cheaper solution.  The
first difficulty we run into is that I'm not sure what the nature of the
stain in the carpet is, so we will have to try to find a way which to
measure the degree of staining which applies to most stains.  I am going
to assume that the stain is not very coloured, that is, it is fairly
neutral or near grey. If this is not the case we will think
again later how we can improve our measurements if the colour of the stain
is a long way from neutral.

For a neutral coloured stain the problem of measurement is simpler.
"Neutral" colours are those which absorb light approximately equally over
the range of light to which the human eye is sensitive. This visible light
ranges between the extremes of the rainbow colours, from violet to red.
When we look at stains, the amount of light to which we are sensitive, visible 
light which we see coming from a stained part of a carpe is less than the
amount of light which we would have seen if the carpet had not been
stained.  This difference in the amount of light which we see
is a measure of the degree of staining and as long as we don't need to
know what colour the stain is we only need to know how dark it is.  For
this purpose we often describe the degree of darkness as the "optical
density" and we use what is called a densitometer to measure it.
Although this is a lot simpler than a colorimeter, you may not be able to
get hold of one of these either.

I suggest we build on what you were going to do instead of using
densitometer or colorimeter.  Instead of simply asking people which
treatment they think has worked best we will make them make a judgement
which will give us a number to use.  You will need to produce a set of
areas of different densities going from light to dark.  Try making the
lightest one about the same as the un-stained carpet and the darkest one
darker than the worst stain you expect to get.  Then you will need to
produce a number of areas with densities between the two extremes.
 You will need to have several levels between the best and worst results.
There are several computer programs which will enable you to make these
areas.  Photoshop is one of the best and will certainly do what you need
but it's quite complicated.  Powerpoint may be good enough.

Make each area about the size of the stain patches and arrange them in a
line on a strip of paper in order of increasing density.  Give each patch
a number.  This comparative set of patches can now be used to compare with
the stains.  Make a mask consisting of a piece of card with a circular
hole which you can place over the stain.  Place your strip of comparative 
patches so that half the hole is over the strip and half over the stain.
You can move the strip up and down to vary the density of the comparative
patch until it matches the density of the stain.  Tell your
observers to try to ignore any colour differences and concentrate on
density or darkness.  Note which patch gives the best match.

You may find this difficult if the stain patches are very different in
colour or a long way from neutral.  Assuming the stains are all the same
colour you could try to make your comparative strip match the colour, if
your computer program is good enough.  An alternative is to use a
coloured filter over the observers' eyes.  Use a filter which makes the
stain look darker rather than lighter.

There are several ways to make the experiment more "scientific".  One
useful, almost essential way is to include what we often call "replicates"
that is repeats of what should give the same result.  By doing so we can
get a much better idea how reliable theresults are.  The more replicates
the better! but they do take time and you have to use your judgement
how many to include.  I'd ask your teacher for advice about this.

I hope you will be allowed to go ahead with your experiment.  This way of
making comparisons is not unusual.  For example pH sensitive paper for
measuring acidity can produce different colours which are then compared
with charts of colours.

The next level to go to would be to try to make a simple densitometer.  If
you want some advice about that let me know.