Yes it is. However, to be perfectly sure it would be necessary to know about your research goals/questions.

Antonio M. Oller-Marcén Many thanks for the response, my research is about the impact of social networking platforms on changing communication culture in higher education and the aim of the study is to explore the factors that influence the use of social networking among academic staff. for this purpose, I have collected posts shared on the social networking platform and I am using qualitative content analysis to explore meanings underlying physical messages and to identify the intentions of an individual, or group, and describe attitudinally and behavioral responses to communications. Thus, for sampling, I am thinking about random sampling but I still have doubt if I can use random sampling for qualitative research.

Nothing wrong you employ random sampling for your qualitative study. But in your case I guess based on secondary information like posts from social media, cannot recruit informant real-time as they already post their opinions.

Therefore, I recommend if you could use purposive sampling in this study, would be more benefited. As you can weight the quality of the information, not representation of the populations.

On the other hand, random sampling will be more helpful in quantitative research where you expect to generalize your findings in the population.

If you could more explain your research situation, I would be happy to comment more. Good luck...

Roshan Panditharathna Many thanks for your reply, my research is about the impact of social networking platforms on changing communication culture in higher education and the aim of the study is to explore the factors that influence the use of social networking among academic staff. for this purpose, I have collected posts shared on the social networking platform and I am using qualitative content analysis to explore meanings underlying physical messages and to identify the intentions of an individual, or group, and describe attitudinally and behavioral responses to communications. in my case, there are around 1000 posts uploaded into 2 years and there are a high engagement and collaboration among employees, the messages are varied in different topics. and the trend of posting is actually higher in the second year. Thus for sampling, I want to pick up posts randomly and then doing qualitative content analysis, but i still have a doubt if I can use random sampling or not.

In such a situation, I recommend you to go for non-probability sampling methods (purposive) as the content is much important than filtering methods. You may end up with no meaningful findings if you choose random sampling as the valuable and relevant content may not capture in random sampling.

Alternatively, if you are desperate on sampling techniques, you can use other probability sampling methods such as stratified sampling or clustered sampling in which reduce the sample bais.

The most relevant issue is the research question you're trying to answer. If you want to make claims about the frequency or proportion of different types of posts (for example 33% of post were about sports, 20% about clothes, etc.), then you'd want to do something like random sampling. If you're more interested in how different sorts of posts are responded to, then purposive sampling would make more sense (because then you'd focus on identifying different kinds of posts and comparing responses. So it really comes down to your research question and the types of claims you plan to make. Based on the description of your project, you might want to be doing a little of both. If you use random sampling and identify the kinds of posts, you could then go back and purposively sample specific kinds of posts to investigate responses to them.

I think the systematic random sample is the best choice for the content analysis especially that you have a large Population.

Firstly you should think what is important. Information or individual? I think it is determined by your objective. I would choose the source where more information available in qualitative research. I do not recommend random sampling for qualitative study and content analysis.

Thank you all for the responses, they were really useful and i will consider all into my sampling. Devaraj Acharya Fuad Altahat Mike Kohut Roshan Panditharathna

Mike Kohut thanks for your response, actually based on my research model which is using enterprise social networking (Yammer) among academic staff facilitate both professional ( building academic identity, developing common ground, feeling connectedness) & personal benefits, i think using purposive sampling is much more suitable than random sampling.

since based on my theme ( developing common ground, building academic identity) , i think its better if i consider the posts whom has a response and conversation.

The difficulty in implementing purposive sampling in your situation is that you need to be able to determine whether or not each message mets some criterion related to your purpose. So, the question is how easy it for you to determine whether or not a given message meets your purposive sampling goals. If you have to read a message to make that decision, then you will have to review the whole set of messages in order to do your sampling.

If that is the case, then I definitely recommend random sampling.

Maybe you should think of some relevant sampling criteria like the time when the message was posted: first year or second year. And follow them until saturation.

Purposive sampling on 2 years of data is 'doable', but a lot of work---though in its favour, it uses all the data judged relevant; however, resources might dictate random sampling as the pragmatic way to go.

Random sampling is used in probability sampling technique and is more compatable with qualitatitive research whereas qualitative research should be biased with purposive sampling technigque which is non-probability sampling technique.

Additional considerations, though secondary to several mentioned above, might be your discipline, your supervisor's opinion/advice (if you're a student), and the intended audience for the research (which journals, discipline/s, other targeted outcomes [policymakers, etc.]). For example, one might encounter resistance from some (though not all) qualitative journals if one applies (only) random sampling--though you can select another target journal! More difficult to switch a supervisor, perhaps.... But I give a vote for 'disrupting' methodological orthodoxies if supported by a sound rationale (and some support from folks with relevant expertise).

You can if your sample size is not small as we do not use large sample for qualitative research. However, the difficult issue here is how to identify sample frame to apply sample random.

Wouldn't the sampling frame be the accrued 2 years of data?

Dear Antonio M. Oller-Marcén

Hello, can you explain your valuable reply please?

Dear Florentina Scârneci-Domnișoru

could you explain more about saturation in qualitative research, please?

The following book is useful in your topic

Research Methods for Everyday Life: Blending Qualitative and Quantitative Approaches (Research Methods for the Social Sciences)

Scott W. VanderStoep, Deidre D. JohnsonYear:2009

Please Go To This Link:

https://pdfs.semanticscholar.org/063d/8f0ba824da42f1a55136bdd7aac0f37cc8df.pdf

Please check the following link:

file:///C:/Users/amal/Downloads/90214-Article%20Text-226365-1-10-20130703.pdf