<BLOCKQUOTE><font size="1" face="Verdana, Arial, Helvetica">quote:</font><HR>Originally posted by anoopt: Can any one give some suggestions on how to form a procedure for creating metrices <HR></BLOCKQUOTE>
As far as "creating metrics", what you really do is create a basis for measurement. A metric is just a standard of measurement. So you really want to ask how you go about getting good measurements. Keep in mind that measurement is the process by which numbers or symbols are assigned to attributes of entities in the real world in such a way as to describe them according to clearly defined and hopefully unambiguous rules.
Some measures are direct, like lines of code in a given program source listing. Some are more indirect, like "defect density", which is basically the number of defects in a software product related to the total size of product. There are also prediction measures, such as predicting the effort required to develop software from measure of functionality – such as via function point counts or use case points. These tie very closely into estimation measures, since estimation is just one way to predict. It is these measures that give you your metrics.
You might want to check out A Software Metrics Primer for a few extra details. Also definitely do as the other poster requested: search in these forums for other discussions on metrics. Various types of metrics have been discussed in a wide variety of contexts.