Research Websites 

 Getting Started

Quicklist of resources: Your First WebsiteDeveloping a Research Question (rough cut video, 5 mins) Duke’s What makes a Good Research Question?, SelfieCity, sample student projects, and various items on reviewing the literature: PDF, video diagram of sample lit review (from the PDF), and various Harvard videos and pages. 

At this point, as a result of your sandbox efforts in the lab sessions, you should be familiar with the basics of website creation. You should already be able to:

  • Create a landing page and links
  • Modify a WordPress theme
  • Manage fonts, images, navigation, credits and pull quotes
  • Make sharing decisions, such as by using a CC license
  • Create subdomains (
  • Install WP/Boldgrid and new themes on subdomains
  • Select appropriate templates (Gridblocks)
  • Build a coherent issue-oriented site of 6-8 pages


With this skill set, you’ll be able to build a research website to publish original analysis of data you have collected yourself. You may end up using additional tools, such as Piktochart, Excel or Powerpoint to organize or visualize your findings and the raw data. You may use interactive tools (eg, quizmakers and game engines) to create interactivity. Nonetheless in most cases you’ll embed the output of any such additional tools on a web page on your site. In other words, you’re ready to take the next step. Even if you need a tutorial in an individual tool, you already have the ability to frame your work for publication. 

Developing a Research Question

Learn more about my approach to teaching

Skills, check. What’s next? By far, the most important next step is figuring out your research question. The secret here is acknowledging that nobody does research writing except to answer real questions. Research writing is not a vehicle for argument. Sure, you will have opinions, and those opinions will inform the questions you ask and conclusions you draw. You may win an argument with research writing, but your opinion and the argument, and winning, are less important than the presumably high stakes of the research question itself. In the end, most real researchers publish to share data and their tentative, qualified inferences.

Unlike a newspaper editorial, which purports to solve a major question in a few paragraphs, research writing may consume many years and pages to make modest advances. (“Give my team $50 million and I promise you we won’t cure cancer. But we may able to identify changes in two proteins that could help detect the onset of metastasis.”)

Research questions come in two basic flavors:

  • Unasked questions: There are more of these than you might think. Most people, including researchers, have a tendency to enter existing discussions rather than start a new topic.
  • Unanswered questions: Important questions tend to have many, conflicting partial answers from different disciplines and different approaches within disciplines.

For this particular class, you will be researching the melodramatic dimensions of a single discourse genre other than film or television. This could be political speech, journalism, memoir, documentary, photography, poetry, music lyrics, advertising for weight loss; infomercials, websites targeting parents, subreddits,  music videos, recruitment videos, etc.

  • Any genre should be refined substantially. So rather than talk about political speech broadly, you might focus on speech by a particular person (Ronald Reagan), by a group (Hillary Clinton supporters), in relation to a particular event (closing US borders to some Muslim-majority countries), etc. 
  • Your research should involve discovering, discussing and archiving examples. You will have web pages devoted to exposition and commentary, and which answer an explicit research question. For instance, “Was Ronald Reagan’s use of melodramatic language received differently by journalists than similar rhetoric by Donald Trump?” or “Did Hillary Clinton supporters unfairly characterize Sanders supporters during the Democratic primaries of 2016?”
  • You should adopt a clear method and approach for your research. Your method might be to search listings of “100 great documentaries” for uses of the word “evil,”  “horror,” “abuse,” “terror,” and “victim” in an effort to discover which docs on the list might have adopted a melodramatic grammar; then to screen the docs to note examples of melodramatic conventions in language, mise en scene, or the like.
  • Your approach might be feminist because focussing on documentaries of women’s experience, or feminist because focussing on melodrama as a masculine framing device; or it might be comparative, looking at parallel discourses–such as advertising for a particular health product–in different languages or locations. It could be historical because it looks at parallel discourses at different times. Your approach could be a genre-studies approach because you look at different versions of melodramatic expression in a particular discourse–eg, “Which is a more common form of rhetoric in debates about public schooling in the US: Sirkian social/family melodrama, or traditional good vs evil action melodrama?” Your approach could be theoretical because it is driven by an academic school of thought. 

There’s a catch to all this: How do you know if a research question is unasked or unanswered?

Reviewing the Literature

The answer is that you don’t know–until you read the most current knowledge on the question in scholarly books or journals. These are chapters and articles just like the ones you have been assigned for this class. This is an unpleasant, inescapable truth. If you don’t review the existing literature, it is a near certainty that all of the effort you put into this project will be spinning your wheels. 

The easiest way to review the scholarly literature is to focus in advance. Knowing your interest as narrowly, clearly–and honestly–as possible is the difference between miserable confusion and feeling like you’re a genius. 

I will work with you to refine your research question, and media specialists will discuss the role of annotated bibliographies and a formal literature review in lab. Here’s my favorite guide to doing a literature review, complete with good and bad examples, and another guide from Harvard, featuring video lectures. I have made a short explainer video (5 mins) diagramming this sample lit review.

One thing to keep in mind: A truly comprehensive literature review can take months. You are not being asked to create a comprehensive lit review. Instead you’ll do a representative lit review. A representative lit review is like a thumbnail of a mega-pixel photo: It is recognizably the same photo, with much less detail. Your lit reviews might mean scanning fifty articles or chapters. You may actually read parts of fifteen or so, and your representative lit review should have 6-8 scholarly items. Non scholarly items may be included, but they do not count against the total. 

Required Elements

There are many ways to do an academic website, ranging from guided tours of archives to crowdsourcing data with gameplay. The example I discuss on this page is the fantastic SelfieCity by Lev Manovich et al. It has many of the elements required for your midterm project. You will: 

  • develop a research question and a plan for data gathering  
  • do a literature review
  • create a site of at least 8-10 web pages featuring original analysis of your data (1200 words excluding the lit review and other elements)
  • share your raw data in some kind of gallery, table, spreadsheet, album, or series of charts 
  • create at least one infographic displaying your findings 
  • adapt from your web pages and visuals a slide deck or prezi suitable for teaching your research question, the existing literature on the subject and your findings 
  • compose a downloadable white paper (which will make use of the other writing you did for the site, including the presentation)

If appropriate, you might include a call to action, create memes, quizzes, or other interactive elements.


SelfieCity by Lev Manovich et al uses rich digital publication to share the research.

The investigators share their findings, raw data, and additional visualizations.

They also share their methods and rationale, and created a tool, the selfiexploratory, empowering site visitors to easily interact with their dataset. 

Of course the site hosts white papers–several of them, in fact. This one by Liz Losh critiques the project’s approach from a feminist perspective.

The site includes powerpoints, video, a blog, and press coverage, among other resources.



There are many, many examples of student research projects in my archived courses. Assignments change every term, but you can see recent student work from this course.  Joey Benevento Jenny Zhou Chang MengLaura Flint Natalie SterrettSaher Fatte Meg Airey Vivie Lee Yuchen ZhangPaloma Bloch Philip Maghen Ean Kitchens Grace Kim Katrina Peed Izzy KornmanVirginia Spinks Sheena DesaiHelen Zehan HouSam NichaminNick LalJoseph North Cody PerezChelsea Walton Gideon Weiss Ajay Harish

Some other examples: Kelli RyanClaire BattyAldo Atienza Vanessa Casalegno Judith MartinezCharlotte West

More examples: Lauren BalotinCayla BambergerUrmi ChatterjeeScott ChoiMax CornwellSumera DangRosie DitrePhoebe EdwardsEmily GoldnerCourtney Greer Lauren GuliasiHannah HigginsAlex KassJustin LeeRusha NaikNina PatricofTameka Pierre-JeanEskew RobertsSamah SadigCecilia SchrammAreeta ShahToby TeitelLuis TrujilloCharles Zakkour