You calculate the mean and SD, You can read our paper and see how we did:
Depression, anxiety, hostility and hysterectomy. https://www.ncbi.nlm.nih.gov/pubmed/16295517
Depression, anxiety, hostility and hysterectomy. Ewalds-Kvist SB(1), Hirvonen T, Kvist M, Lertola K, Niemelä P. Author information: (1)Department of Psychology, University of Turku, Finland.
I agree with the former comment about needing to know what items you are using.
However, the severity of depression can be viewed in different ways. For example, with the DSM-5 there are nine basic criteria to define major depressive disorder. In this case severity can be defined by the number of positive criteria. If you are more concerned with the level of dysfunction caused by depression then items elated to functioning might be weighted.
In the case of milder forms of depression, such as depressive feelings within the "normal" range, other strategies for defining severity, such as the statistical one mentioned, might be appropriate.
To provide a more specific response one would need to know the purpose for determining severity and the item set used, or proposed.
It depends on the scale you are using in your questionnaire; for example, if you are using a likert 5 point scale, the range of your questionnaire is 30*5=150, then you may take .
0-30 as none,
31- 30 as mild
61- 90 as moderate
91-120 as severe
120-150 as very severe
for an example if anyones score is 145 then they are very severe.