Then we sort the results

[kexej28769@nongnue]

We collect the search results. A based on various metrics such as the number of links. Finally, we compare the orders of the original search results with the orders generated by the various metrics. The closer they are, the higher the correlation between the two.

That being said, correlational studies are not entirely uruguay number data because they do not necessarily reveal causal relationships (i.e., the actual ranking factors). What correlational studies discover or confirm are interrelated .

Correlations are simply measurements that share some relationship with the independent variable (in this case, the order of search results on the page). For example, we know that backlink count is related to ranking order. We also know that social shares are a correlation of rank order.


Finally, correlation studies can help us rule out proposed ranking factors. This is often overlooked, but it is an incredibly important part of correlation studies. Research that provides negative results is often just as valuable as research that provides positive results. We have been able to rule out a variety of potential factors — such as keyword density and meta keyword tags — using correlation studies.