I wouldn't call it observation, because you have a static set of text-based data that will not change. So it is the same as analysing historical documents, sets of emails, websites, books or letters. I'd call it document analysis. This is a separate question to what sort of analysis you might choose to use with this data, although you would obviously want to choose a methodology that is suitable for text-based data and related to your research question, such as discourse analysis, content analysis, or narrative analysis.

This webpage described an Observational Study:  https://explorable.com/observational-study

Based on this explanation, I think this is not you intend to do.

If it were me, I would simply call your methodology a Content Analysis.  Within Content Analysis, you can do a Conceptual analysis, which looks for themes and considers the ideas that appear in the text. When you do this kind of analysis, I think that the content in which the text or ideas appear is relevant.

Or you can do a Relational analysis, meaning an analysis of how certain words or ideas appear close to other words or ideas.  

So I think that you want to do a Conceptual Analysis.

Thank you soooooo much Michael

I considered it content analysis at the beginning, but then I felt in doubt because I thought that it won't include the ideas and general theme of the debate.

Thank you so much for explaining the conceptual analysis, it was super helpful..!

I think you need to differentiate between data collection and data analysis. In my view collecting data from a social media platform in the form of conversations, debates or otherwise interactions between people preserved in time certainly qualifies as observation. The question is, would it matter if it unfolded in real time before your eyes (provided you don't participate); I guess the answer would be no. So, it will qualify as a form of observation. That would describe your data collection. As for data analysis that's a different matter - content analysis, conceptual analysis, communication genre analysis, there is a number of possibilities and it really depends on what you want to achieve.

i am quite sure it is content analysis as you are going to collect the raw data and design a coding sheet etc.. 

content analysis is very good method although its exhausting process 

and by the way i will recommend you have a look over Dr. Adel M.Aladwani work's and luckily he already  answered your question

You can use the Nvivo software to have some qualitative analysis

I wouldn't call it observation, because you have a static set of text-based data that will not change. So it is the same as analysing historical documents, sets of emails, websites, books or letters. I'd call it document analysis. This is a separate question to what sort of analysis you might choose to use with this data, although you would obviously want to choose a methodology that is suitable for text-based data and related to your research question, such as discourse analysis, content analysis, or narrative analysis.

I would agree with Victoria that it is not observation. It's not a 'live' data set - and has passed. I agree with Kai that it depends on the context as to what you call it. He has correctly stated that it might be a 'concept analysis'; it could also be a 'discourse analysis'. It might be stretching it a bit, but depending on how old your data might be, it could be even be framed as 'historical research'. 

I would agree with Victoria that it's not observation. It are in some way recordings of naturally occuring talk. Discourse analysis or conversation analysis would seem to be appropiate analysis methods.

I agree with those above.  I would call this analysis of discourse from social media.  If you were to hang out in a chat room and participate in the discussion that might be an equivalent for "observation."

Again, I agree with others. This is secondary analysis. Which is fine. If you have access to the primary data set then you can do content analysis, but if not then I think you might be better to place in a meta-nalaysis framework with other studies.

I agree with all of the above. But I think most important is to look at the research question, what is your objective of looking into the past records of the social media. Based on the description of your question, I assumed you would like to observe 'what is going on here' sort of thing or looking at the trend perhaps?.

It is secondary data hence it is  based on document ( as data sources). In regards to the analysis it is very much in the content analysis type of data analys

However, if you like to see the experience of the people that engaged in the social media, then I suggest you do phenomenological type of analysis such as discourse or conversation analysis. Again most important is your research question? what is it that i want to know?

well good luck.

I suggest have a look at Ken Plummer's "Documents of Life 2". That will give you some great ideas on this sort of secondary documentary analysis.

I think this is an ethnographic descriptive study (qualitative)

I would even consider it suitable for a discourse analysis, depending on your research interest. But what you describe perfectly fits into a ethnographic study description

Note that whatever category it falls into, if your design is a "systematic investigation...[using] personally identifiable private information" from living individuals, your IRB will likely require you to submit and/or publishers will require IRB review. Internet research is still Human Subjects Research if the design and participants meet 45CFR46's definitions. 

http://www.hhs.gov/ohrp/humansubjects/guidance/45cfr46.html#46.102

I think this is primarily a matter of labeling so you need to think in terms of the terminology that exists within your own discipline, and find a label that will be familiar and comfortable for your own colleagues.

I myself do a rather interdisciplinary version of social science research, and the most common label I've seen for what you are describing is Qualitative Content Analysis. Here is a review article on that approach:

Vaismoradi, M. et al. (2013) Content analysis and thematic analysis. Nursing and Health Sciences, 15, 398–405.

Observation is of the process or conditions of live situations. What you are referring here is neither observation nor ethnography. It might be referred to an ethnography if you were the part of those discussions or platforms when these debates were being carried out. I would suggest you call it content analysis, which is an established research method (both in quantitative as well as qualitative). 

Dear Nouf, This would not be observation, as others have already said. As you mention, it is part of an ethnographic study, so it can be simply an analysis of....  with your own description at the start, of how you will analyze the information an what information you will use.

It can be content analysis only if indeed, you follow the technique and  method of content analysis.  In that respect, you can also conduct a thematic analysis.  Or, as part of your larger ethnographic study, I have to say again, it can simply be an analysis. Not every analysis has to be one single named method. Also, it sounds like you already have planned what/how of the analysis, rather than if you had decided on a specific method, such as content analysis, thematic analysis before thinking of what/how you will do it.

I agree with our colleagues. I wouldn’t call it observation because such practice involves a live situation in which you participate as an observer. I think that the proper way to call your analysis depends on what concrete aspects you are interested in, namely, if you are going to focus on discourse, dynamics, participants’ behaviors within the debate, etc. For example, If you are doing "discourse analysis", I would also remark that it has been done from a secondary documentary source. Maybe that could help to clarify your work's sort of analysis and the data procedence.

Hi,

Obviously it goes into the aspects of content analysis. Moreover, studying past events wouldn't provide you room for real-life observation (unless you have some pre-recorded artifacts). Perhaps you can frame all the elements that you want to study within the context of ethnographic studies?

Prof. Alhuzami:

I see a number of good observations here, notably the idea that this is mostly about an appropriate label for your work.  Observational anthropology, later relabeled as systematic ethnography is only partially appropriate, I think.  You are, in fact, talking about textual content analysis and have any number of sources to consult. You might begin with Webb, Campbell, Schwartz and Sechrest, Unobtrusive Measures. They define much of the field and offer time-tested, systematic approaches.

Krippendorf has produced a very helpful paperback on Content Analysis: An Introduction to Its Methodology that may prove helpful.  Margrit Schreier's text on Content Analysis will tell you just about all you need to know.  Base your methodologies on time-tested (and peer reviewed) approaches and you'll find much wider acceptance for your work.  

Best wishes,

J. S. O'Rourke, IV

Dear colleagues,

thank you soooo much

your comments were really helpful.

i thought it could be an archival-observation as it wont focus only on the content of the text but all the atmospher around the online debate. thanks for clarifying this.

and thank you so much for the references they were helpful.

best wishes

Nouf

I would rather think content analysis is more appropriate.

Good luck with your research!