Part 3. Collecting Data with the Stress Meter
Version 1.0, April 1998
Your Measurements
Whether or not you succeeded in generating a C
programme during Part 2 of the project, you have a QBASIC programme which works,
so that you can take some measurements, and can then read the data into Excel
for analysis. You have the tools to complete the project.
We want you to try resolve two linked questions :
- whether the stress meter will work as intended
- what are the best conditions for the measurements
The two
electrodes need to be placed on adjacent fingers, and must be taped firmly in
position with a piece of ribbon, just gripping the electrodes between the
fingers is certainly not going to be much use. You need good electrical contact
with the skin, obviously it should not be wet since you are trying to measure
the subject's perspiration!
- Find out how stable the system is, that is how much 'noise' (variation in
the signal) there is, and whether the range of readings you can get is
adequate. Is the number you see on the screen a realistic measurement of the
person's skin resistance? Does adjustment of the 1Mohm potentiometer, VR1 on
the pcb have any effect? Because of the variability of the electrical
components that make up your circuit, and the way it has been constructed by
you, there may well be problems for both the stability and the sensitivity of
the meter.
- How rapidly are the readings changing if the subject just sits passively
looking at the screen? The amount of time between the readings may be
significant, that is should you take measurements every second, every ten
seconds, every minute? Would it be more effective to take a number of
measurements and average them?
- Remember that an aim of the stress meter is to see if the person using it
can influence the measurements by relaxing more effectively. This may be
difficult in the conditions of the Tiger laboratory with a lot of other people
around, and where the atmosphere is rather humid and hot. (Turn the air
conditioning on!) We leave it to your ingenuity to generate a relaxing
situation for whomsoever has the meter attached.
The aim here is to collect enough data to show whether or not
the meter works, that is you will need the data largely to enable you to comment
on the method and design of the experiment. The same would be true
with any experiment, you need to run tests to see how well the apparatus works,
to calibrate it and to find out its precision. Thus, collect sufficient data
from your apparatus, using either the QBASIC Programme 3, or your own C version
of Programme 3, to enable you to determine how effectively the apparatus
'works', to test it, and to allow you to find any obvious problems. Move your
data into Excel so that you can use the spreadsheet to examine the results.
Excel has sufficient graphical, mathematical and statistical facilities to
enable you easily to do this.