For the semester so far, we have focused most on the rhetoric of the formation of data.
Rhetoric, again, for us, is just the intentional use of symbols to make meaning, influence others, and coordinate activity.
We have considered what consequences there are when something in the world is turned into data, into symbols. What does it mean to collect this kind of information to turn into data rather than another kind of information? What happens when you categorize something this way rather than that way? What is missing? What might these choices do to possibly benefit or harm people? How is the data set organized for use and how might that organization make something more or less accessible? Or suggestive of certain kinds of uses rather than others?
The list goes on.
Now, for the rest of the semester, we are going to shift to focusing on rhetoric once there have been some results. How do you communicate results? How do you do so effectively? And ethically?
Over the next few classes, we are going to go over a few specific techniques of writing about data. For this learning module, we already talked about communication in terms of talking about context/limitations which was a good middle-point between thinking about rhetoric in the data set vs. rhetoric on the page.
On these next two pages, we will talk about explicitly communicating your interpretation and using examples.
Interpretation and Communication
If there is a spectrum of kinds of writing you will do with data analysis, where one end of the spectrum is academic research for an audience of other experts and the other end of the spectrum is writing for a public audience where no expertise is assumed, the question is how much the numbers should “speak for themselves” anywhere on that spectrum.
D’Ignazio and Klein write that, especially for writing that is closer to the right side of that spectrum, offering no stated interpretation of your analysis risks numbers being misinterpreted.
But even academic writing, where experts write for other experts, the numbers really do not speak for themselves. Think of the conventional format of most academic papers in the physical and social sciences, the “IMRD” format. This acronym stands for “Introduction, Methods, Results, and Discussion.” Read any of these papers, and in both the “Results” and “Discussion” sections there will not just be a list of numbers without comment, but there will be a variety of interpretations explicitly stated.
In writing for public audiences, there will be less room and less assumed expertise, so writers need to work harder to explain why the findings are meaningful.
In chapter 6, D’Ignazio offer the example of the chart about disparities in mental health diagnoses in prisons. They offer three ways the chart could be framed:
- Mental Health in Jail: Rate of Mental Health Diagnosis of Inmates
- Racism in Jail: People of Color Less Likely to Get Mental Health Diagnosis
- Racism in Jail: White People Get More Health Diagnoses
None of these framings is inaccurate, it is just that two of them offer clear interpretations while the first offers no such interpretation.
If the researchers use a variety of evidence to come to similar interpretations of the latter two titles, why not just say that? What is your job as a public writer? Is it to say, “hey, you figure it out by doing the same work I did and go back to the research I looked at that isn’t presented here” or do you do some translation work to make sense of complicated information for folks who don’t have the tools or time to comprehend that information?
Communicating Interpretation in White Paper
In your white paper assignment, you will be in the middle of the spectrum above. You will be writing in a detailed way that experts could access but it will be something public-facing, and should be made accessible. It is a piece that is both for capable people of getting into depth about a subject, but perhaps not at the level of writing for super-specialized people.
You should make time to be clear one what your methods are for analysis, justify why you include certain information, be sure to include plenty of contextual information, etc.
You should still be clear on what your interpretation is of both any data analysis you do and your own understanding of other data analyses and various research you cite.
Task
Note the framing in the below image:
To see the image better, and to learn more about how this was calculated, go here: Quick Facts About the Gender Wage Gap – Center for American Progress
In a comment below, do one of the following:
- Create a new title for the bar graph and explain why you offered that new title. Think about how this is a graph for a public audience with varying levels of expertise.
- Explain why the title already in the image is appropriate for this graph. Think about how this is a graph for a public audience with varying levels of expertise.
After commenting below, click the button to continue.
I think a more appropriate title for this bar graph is “Gender Gap Wages Across Different Ethnicities.” When presenting a data set that usually establishes credibility in any argument, paper, news article, or writing in general, I don’t think its appropriate to opinionate it but better to leave it to interpretation and understanding of individuals accessing the graph or information pertaining to a data set. Just generalizing it with my title would create more credibility because I think it doesn’t desperately throw the opinion in your face.
I would change the title of this bar graph to “Women of color earn at least 38% less than white men”. The reason I would choose this title is to appeal to a public audience. As we discussed above, by simply doing some back-of-napkin math for the reader and changing the title to something that explains the data, we are able to help the reader better understand it. “do some translation work to make sense of complicated information for folks”. I like this strategy better when I am trying to prove a point to an audience with a broad level over expertise, or public audience.
I would change the title to “Gender Wage Gap is steep for women of Color” because this title sensationalizes the topic more. Learning from social media and such, clickbait rules media at the moment. People will not even read the article but from the headline and subheading readers will get the gist. The title already is great because it highlights large groups of people and using “significant” shows readers that this is a big deal.
I probably change the title to ” Ethnicity Plays Important Role in Gender Wage Gap” because there are a lot people on the internet (included myself) that we read the article based on the title; the title is the attraction in which the readers are motivated either a portion or the full article. The article does not talk just about women of color, it also talks about other ethnicity, the title should be broad and not just focused on one specific thing.
The new title I would offer for the bar graph is “The Gender Wage Gap in Different Ethnics”. I would choose this title because the graph shows the wage gap between different ethnicities. I offered this new title because I feel like it’s more appropriate and help readers understand what they are looking at. Just by reading the title you get a sense of what it is about and what’s the graph showing. The original title didn’t really make sense to me. With my title, it doesn’t really tell readers that i’m trying to be biased, or just showing one sided graph.
I would change the title to “Minority women take the cap in gender wage gap” because it looks like a huge difference in the statistic shown. The graph is very generally speaking of women and men but clearly there was a 10K or more difference between women who are minorities and white women, as an example. This might give a more clear viewpoint for the reader.
I would change the tittle to “White men still earn more than women of different ethnics”. The reason I want to tittle the graph as this is because I feel that having the phrase “white men” is more specific and embodies the idea of men as well as white supremacy. I added “different ethnics” instead of “women of color” because I want to be more inclusive and state that this is a problem that all ethics groups take part of rather than just of “color”.
I would replace the title to “Women of color’s earnings are the lowest.” The reason why I want to change the title is because I want to shorten it, making it as simple as possible but still carry out the important content, keep the main meaning of the title to the audience when they first look at the title. Also, I think the replaced title tells us everything from the graph, bringing out a brief and concise summary about what the graph is about.
I would change the title to “Wage Gap Amongst Different Ethnicities of Women.” I chose this title because if the original title already told you a fact about the graph which might make people less inclined to look over the graph because they already read that information. The title that is already there is good because if someone didn’t know how to read a bar graph, they could get the information from the title.
I would edit the title of this graph to “Gender wage gap emphasizes inequality and women of color face it worst.” I chose to go with this title because the gender wage gap is a major social issue women are facing the disadvantages of but particularly with women of color their is a major increase I believe this needs to be addressed/maximized within the title for the reader to understand immediately the differences. The title that is provided I believe is a good fitting description for the information/topic it addresses the issues occurring in the gender pay gap.
A new title for the bar graph would be ‘Racial and Gender Discrimination Through Pay.’ I proposed this title to emphasize remuneration inequalities against women, especially those of color.
A new title for the bar graph would be ” The Gender Pay Gap Within Different Ethnicity In Women, I choose this title because occupational segregation can be influence by gender and racial stereotypes. When I look at the bar graph, I saw different ethnicities rather than gender being male vs. female. The title in the image makes sense because it clearly shows women of color with a significant lead.
The title is appropriate used in the graph since it gets to the point, isn’t clickbait, and actually shows the gender wage gap between different ethnicities of women. A new possible title could be “gender wage gap between different ethnicities of women compared to white men.” I did this because the graph is showing different pay gaps between different ethnicities of women and it’s also being compared to white men.
The title of the graph could be changed to read “The gender wage gap is also racially correlated..” It would be a reasonable titled because the data measures women’s median earnings for every dollar a white man earns, based on their race. The original title is also appropriate but it differs from my suggestion in that it puts an emphasis on comparing white women’s earnings vs white men.
I think a better title for this bar graph would be “The difference in wages across race/ethnicity in the US”. This title would be better because it would explain more of what this graph actually is showing compared to a inference made from the data on the graph. The title on this graph may already be appropriate because it is explaining the reason the graph exists in the first place, to show how women of color make less money on average compared to White and Asian women.