Content analysis

Home Forums Default Forum Content analysis

Viewing 8 posts - 1 through 8 (of 8 total)
  • Author
  • #5648
    Ahmed Chouari

    I am working on a doctoral project on Communication and Conflict in Intercultural Mariiage, and I am using content analysis for qualitative analysis of the interview. Can anyone send me more details about the use of content analysis in dealing with interviews? Are there any suggestions about software programs that can be used in such analysis? Thank you so much for your help.

    SAGE Publishing

    Hi Ahmed, we’ve received a post from twitter in reply to your query (
    “Same drill as analyzing any other content: set strict & well-defined categories, train coders, code consistently. good luck!”

    Ahmed Chouari

    Hi John, thank you a lot for your reply. Indeed, this is what I found in the literature, but I want to trigger more discussion on the topic.


    if somebody could really explain to mewhat content analysis is all about i will really welcome hat.thanks

    Roger Gomm

    To mou ferdinand

    ‘Content analysis’ is a broad term which covers a range of different approaches to analysing written materials , or transcripts of speech. At its most general it simply refers to ‘analysing the content’ of whatever, irrespective of the method of analysis – in that usage to say ‘analysing’ would be sufficient. More specifically it refers to doing the analysis by coding items of content.’Coding’ really means indexing, which is classifying an item of content as being of a particular type, naming it as such, and giving it a location code so that you can find it again, or list all the places where the same thing crops up, or cross classify to identify when one thing does or does not crop up following another. Sometimes the unit of analysis is the word, and sometimes the concept or theme or trope (or whatever term you like to use for what is same but might be expressed in different words) (Thematic analysis might be an alternative term here). Usually but not always  the analysis includes analysis by frequency: for example what percentage of utterances referring to politicians in the set of interviews are perjorative (and what percentage in the interviews with sub-types of interviewee: eg males rather than females). There are acute problems with such counts however, first in deciding what diverse words to group together as meaning ‘the same’ (but that’s nearly always a problem with the analysis of qualitative data) , and secondly with hundred percents. If one type of interviewee, or one type of magazine article is much less verbose than another, are differences in counts due to differences in attitudes or differences in verbosity, or where the data are interviews differences in the way interviews twisted and turned – but this just a version of the familiar problem with analysing open-ended questions- which is, of course a kind of content analysis.And there is the problem that often frequency of mention will not be a good indicator of importance to the writer/speaker or of its consequentiality.


    Sometimes the term content analysis is restricted to a narrow word counting approach and an analysis which looks entirely quantitative, though actually below the surface there will have to be a great deal of qualitative decision-making about what words to group together as being instances of something similar (except of course in corpus linguistics which is entirely interested in word frequencies in a language, dialect or register).

    With the latter exception, content analysis of any kind depends heavily on a classification scheme which defines what kinds of ideas, concepts etc there might be to identify and index: thus the same material might be subjected to radically different content analysis if, the analysis was driven by a desire to find out the meanings people gave to their experience, or if it were driven by a desire to find out how people communicate with an interviewer in an interview, or how people use narrative conventions to tell stories about their lives. In that sense the results of content analysis are always predetermined by the interest of the researcher, via the classification system which is thought to be appropriate for the research question. The classification system might be established at the outset, or, as in a grounded theory approach it might be evolved during analysis (or during analysis and data collection, if data collection is driven stage by stage according to what was found previously)


    Neuendorf’s web site is a useful resource


    Avi Lambert

    Hello all,


    Some very intriguing points about the content-analysis (CA) method. I’ve started a group ( ) about CA for web based research. I am studying bottled water contestation efforts using social technology at the Masters level at the faculty of Environmental Studies in York University in Toronto.

    One of the main concerns with CA, as has been addressed, is that it is problematically constrained by the fact that it must reference seemingly objective, agreed open diacritic matters of interest to become valid and reliable from a statistical standpoint. In other words this deals with the tension of needing to stick to categories and schema that are more connotative than they are denotative; coding subjectivity prevents replicability and so on. However, what I would like to contribute here is how specific research points to a middle-ground method which holds promise for CA being used in mixed-methods research settings, ie providing qualitative and quantitative forms of knowledge.

    The research I am using for my inspiration is that which can be found in the book Social Communication in Advertising. The key proposition the book generates is that CA, as a form of methodological qualitative analysis is one capable of cohering with other existing methodological forms of research. In the case of the book in question, the CA form is combined with the semiotic method to analyze, among other things, historical trends in advertising. In my own research I am combing the WebCA method with the concept of analyzing the frames used in bottled water contestation. More clearly, I am of the opinion that CA in particular will enable me to statistically present and deliberate which frames are more dominant and subordinate.This is seen as critical to provide multiple forms of interpretation from my research about bottled water contestation efforts.


    For students and practitioners of the CA method The Content Analysis Reader, by Klaus Krippendorf and Mary Angela Block is invaluable. As well, the more recently published book by Krippendorf On Communicating: Otherness, Meaning and Information is useful for implicating the value of the CA method. 


    Here is Krippendorf’s home page at UPenn >




    Marina Kirac

    Hi, Ahmed,

    I think you’ve already received plenty of information on the use of content analysis. As to your question on software programs you could use, I found Atlas.ti  very useful, not to focus on linkers or repeated phrases -like fillers-  but to work on discourse content. The program allows you to write side comments, add memos, etc. You can also transfer that information into graphs or charts. Again, very useful for qualitative analysis.










    that was great and it’s very simple. i thaught it was a combersome method, i knew the basic ideas but the way people comment about it seems to me that it was really heavy you know. thanks very much i can now understand it better and it so simple a method.



Viewing 8 posts - 1 through 8 (of 8 total)
  • The forum ‘Default Forum’ is closed to new topics and replies.