https://blogs.helsinki.fi/quantitative-communication/methods/content-analysis/

 

Research 1: Speaking ‘unspeakable things’: documenting digital feminist responses to rape culture

Summary– Research conducted in order to examine how occurrences of sexual harassment are being spoken about in digital spaces-the digital spaces examined are both a website as well as Twitter.  Using an ethnographic, multi-methods approach, the researchers seek to better understand how conversations around rape and sexual harassment are being communicated as well as benefits of doing so in online communities.  Data for content analysis  was retrieved from an anti-harassment site known as Hollaback.  General findings from the Hollaback content analysis have contributed to understanding how anonymous posts regarding sexual harassment provide avenues for resisting rape culture by increasing conversations and visibility of these occurrences (Keller, Mendes, and Ringrose, 2016).  In this research, the authors show supportive evidence that digital spaces serve as network capabilities for strengthening feminism,  specifically a stronger voice in combating rape culture using not only traditional forms of communication but also the benefits of range of scope of having conversations online.

Content Analysis– Data collection and analysis done using the blogging site Hollaback included content analysis, specifically frequency analysis, as well as discursive textual analysis to support the frequency analysis which is limited to descriptive results ((Keller, Mendes, and Ringrose, 2016).  Samples from Hollaback included posts from 2006-2015, there were a total of 159 posts, all coded to specifically target posts that included content related to harassment.  Conclusions drawn from the content analysis use an inductive approach to reasoning, using specific examples of a relatively small population to make more generalized conclusions about a phenomena.

Method(s)-The methodology employed for content analysis on the blogging site Hollaback was appropriate based upon the exploratory nature of the research and source of data.  Hollaback was appropriately chosen as a data source when reflecting on the purpose of the research- it provide a good data source for online conversations regarding sexual harassment.  Coding used to examine specific posts from a time range was the source of data, with a relatively small data sample considering the estimated users that post on Hollaback.  I believe the researchers do a great job selecting their data source but a couple of changes I would use would be to increase my sample size, being more specific about the number of posts I collect from each year, and as well increasing my sample size.  One benefit of using content analysis is the relative ease of capturing samples and with Hollaback having representation in almost 100 cities and used by persons in over 30 countries it is not unreasonable to capture a larger selection of posts.

 

Research 2: Keeping It In “The Family”: How Gender Norms Shape U.S. Marriage Migration Policies 

Summary- Research conducted in order to examine how gender intersects with the treatment of couples seeking to marry- U.S. citizens marrying foreign born nationals.  The focus of the research was on the processes used in order to determine the validity of marriages (avoidance of marriage fraud), specifically how the use of a “red flag” system can be biased depending on one’s gender.  Under current systems, there may be unfairness in how families and sexuality are defined and further used to determine levels of genuineness and “red flags” (Longo, 2018).  Data for content analysis consisted of conversation threads from various sources that included samples from varying demographics- nationality and sex.  Findings from the research show that gender and sexuality play a part in influencing how those seeking marriage and citizenship are treated, with distinct disadvantages associated with women-increased red flags and greater sexual privilege afforded  to male relationships between accepted (Longo, 2018).  In this research, the author argues for greater equality among those seeking marriages between U.S. citizens seeking those outside of U.S. citizenship, specifically, how sexuality should not influence the processes towards these marriages leaving females often at a disadvantage.

Content Analysis- This research study was based upon an online digital ethnography from a digital space- “Immigration Pathways” which is a self-help site for those seeking assistance in immigration processes (Longo, 2018).  In order to more easily extract and use data, a content analysis was appropriate for the research.  Data was specifically extracted from “Immigration Pathways” using the programming language Python, searching for the key phrase “red flags” against over 48,000 conversation threads (Longo, 2018).  The term “red flags” was chosen because of the likelihood that the term would apply to conditions where issues of validity and fraud could me matched up to gender and/or citizenship.  Conclusions drawn from the content analysis use an inductive approach to reasoning, although there was a large sample size used and conversations surrounding the use legitimacy of red flags and issues related to gender and citizen to non-citizen marriage occur in relatively few legitimate spaces.  The research results where not randomly sampled should provide valid supporting evidence to show that females may be at a disadvantage when seeking marriage to non-citizens as compared to their male counterparts.

Method(s)- The research, like all others has both advantages and disadvantages.  A disadvantage is that the population sampled was not done so randomly.  This challenges the representativeness of the sample and cannot provide evidence to greater populations, both within the U.S. system as well as make statements to any other like systems in other countries.  A key advantage of this research is the effective use of content analysis.  Digital spaces can be made readily available for content analysis, and being able to pull samples from across over 48,000 conversations is something not feasible with other methodologies.  The access to such a large sample size should be able to produce significant evidence for findings within the population sampled and some effective data analysis and potential for identifying strong associations should be able to be made with the sample population.  If I were to borrow off of this research I would most likely want to include some face to face interviews, yes, there is the greater potential for reactivity bias but face to face interviews are very effective at getting details that even online environments cannot, considering in online environments users feel less inhibited but also potentially less likely to report accurately.

 

Reference(s):

Keller, J., Mendes, K., & Ringrose, J. (2016). Speaking ‘unspeakable things’: Documenting digital feminist responses to rape culture.27(1), 22-23-36. doi:10.1080/09589236.2016.1211511

Longo, G. M. (2018). Keeping it in “the family”, how gender norms shape U.S. marriage migration policies.32(No. 4), 469-470-492. doi:10.1177/0891243218777201

Shorts in Psychology. (Oct., 2018). Content analysis. Retrieved from https://www.youtube.com/watch?v=iZZABFd8cSA

Found this useful along with the week’s resource on content analysis.

 

 

 

 

 

 

5 comments on “Content Analysis

  • I agree that adding face to face interviews would be an effective way to get get details that online environments could not but do you think that this could be time consuming and expensive?

  • Thanks for your comment. Yes, it would be both time consuming and expensive, and that is a good point about the benefit of doing digital data extraction and analysis- it is typically less expensive and resource intensive because you have a captive audience of Tweets, posts, etc. that can be pulled simply from programming. I always consider the value of all methods but yes, as you mentioned in this case it would limit the feasibility of doing this research. Thanks for your feedback.

  • Hi Benj,

    Great summary of both studies! Increasing sample size is really important to having a sample that is more representative of the population. Face interviews is a good qualitative approach because questions can be explored in depth and in detail during interviews by using probes, asking them to elaborate. You can also pick up on subtle cues like mannerisms of your subjects.

    Best,
    Alice

  • Thanks for your comments Alice, great meeting you formally in the meeting today. Yes, I agree, face to face interviews provide nuanced differences from other forms of obtaining participant data. Yes, they are time consuming and require additional resources but they can also be instrumental in getting a deeper understanding than what some digital data represents. I am learning this with my own research- I did data analysis of my first semester surveys and if it had a been research larger in scope I could have used big data analysis techniques but it does not replace the value that focus groups interviews are providing for me in determining the “WHY”.

Leave a Reply

Your email address will not be published. Required fields are marked *