Yes,

suppose you have 3 responses like  x,y   as  3,4   4,3,   and   2, 5

and you apply them over x,y  chart.  You get three different points.

with  more and more contributors say,,,  3,4     4,3, and 1,4..

                                         and        2, 5    4,3,  and  2, 2

You notice 2 separate urbanisation of points on graph where as the previous ones do get a crowding.

separately they may be called Factors. Hope your question is answered.

The statistical situation is more complicated in 4 quadrants of graph. 

Yes - the loadings are dependent on responses - however, if you have a stable factor structure, as you switch to different samples of respondents drawn from the same population the overall structure of the analysis should not change - the loadings will just moderate raw score differently.

Thanks Kimberly. Hope you will help me with these additional questions. What do you mean by factor structure stability? And how can factor structure stability be achieved? 

Thanks Priyavrat Thareja and sorry for late reply. Frankly, I did not understand your explanation. 

I suppose I should have said solution stability - meaning the factors that emerge and the items that they are comprised of can be replicated across the samples on which the instrument is implemented. You can improve this by having a good scale! Meaning the items in the scale are cohesive within the factors they align with.  Implement, refine, implement again

Thanks indeed. This is helpful.