Generally IF is good for those journals which publishes a quality research work.The review criteria is stringent and every check for palgarism is made so as to assure that what ever reported in done and falsification of data is not there.So if your quality of work is good,Writing skills are good then it can be published in good journals with high IF. And as a researcher i and even would mostly trust these journals for reference since the data presented is true and no falsification has been done.so i guess it depends upon your quality of research work,and many good IF journals have open access too! so this should not be a hurdle or blow to science in fact it will filter a true and quality data for improvement of science on the contrary!
Well I totally agree with the correlation between impact factor and quality. But, lets say, if UGC makes no discrimination between social science, science and humanities in the point base academic appraisal and the number of indexed journal in the JCR databases indicates unequal distribution of journals, where do we draw a line of distinction?
I am a science student and i am also awestruck over the fact that we actually do not have a normalizing criterion!!. I find H index also very irrational. But I think, if we are in position of judging some one's work, we should be little deliberate and should have in our mind the thought of going beyond the brand of publication to quality of publication. One solution to your question may be, we can have a look on some top journals of that particular field at random and see, how many citations do authors from best journals of that field get generally, so we may have a relative idea of the quality of work of author in that field, which we have already normalized according to other fields by comparing it with the citations in the said field.
but, this is just a speculation, somehow I find it working.