|
Title:
|
|
Statistical analysis for solar cell research
|
|
|
|
Author(s):
|
|
|
|
|
|
Published by:
|
Publication date:
|
|
ECN
|
1996
|
|
|
|
ECN report number:
|
Document type:
|
|
ECN-RX--96-021
|
Other
|
|
|
|
Number of pages:
|
Full text:
|
|
4
|
Download PDF
|
Abstract:
The use of statistical blocks is important for larger experiments withmulticrystalline silicon wafers and in cases where the cell parameter depends
on the material quality (such as VOC). Neighbour wafers have to be used for
these kind of experiments and each neighbour type has to be distributed over
all the treatments. Each neighbour type represents one block and each block
contains all the treatments. Cells that had a certain treatment represent an
experimental group. With the use of the statistical blocks one can separate
the two sources of variance within one group: variance due to the difference
of different neighbour types and variance due to the experimental random
error. This will result in a reduction of the error. The use of statistical
blocks is not important for cell parameters that are not or only slightly
dependent on the material quality (such as FF). 2 figs., 4 tabs., 1 ref.
Back to List