4.4: Data Commentary

Data Commentary

Data Commentary

Part of the research process includes making sense of the information you find, for yourself and especially for others. Data commentary is the process of explaining and contextualizing information. This section will discuss what data is (including some different types) and why commentary is important for understanding data.
What is Data?

Writ large, data can be defined as any information generated or collected through inquiry or observation. For example, in your class, you might observe individuals with curly, straight, or wavy hair. Observations of these categories of hair comprise one kind of data. Alternatively, you might measure how varying levels of carbon and nitrogen affect the growth of plant root structures. The measurements of plant roots represent another kind of data.

What are the Types of Data?

Both of the above scenarios involve the generation of data: information about observable or measurable phenomena in the world. In the scenario above where effects of carbon and nitrogen are associated with plants’ root structures, data would consist of measurements associated with the roots. For example, the length of roots produced by each plant, or the number of ‘shoots’ branching from each root are two different sets of quantitative data. Quantitative data is usually defined as numerical information, generated or collected through measurements conducted by humans or mediated through devices (computers, microscopes, scales, etc.)

Alternatively, the information collected in the first example—that people in a given group have different hair texture that can be described and grouped by category—is what we usually call qualitative data. Qualitative data is observable or measurable information that is not (typically) numerical in nature.
Writing and Talking about Data: What can we do with it and what should we do with it?

While the distinction between these two types of data can be valuable, it’s important to recognize that there are not hard and fast boundaries between the two. Qualitative judgments are often made about quantitative data, and quantitative judgments can be used to manipulate qualitative data:

“All quantitative data is based upon qualitative judgments; and all qualitative data can be described and manipulated numerically.” (http://www.socialresearchmethods.net/kb/datatype.php)

What does this mean? It means that when we collect quantitative data, our measurements are based on our qualitative judgments about what is important/valuable information to measure. So, in the root example above, the plant’s root length or complexity are potential measures that tell us something about how the plant has been affected by the variables, carbon and nitrogen levels. Through exposure to other scientific studies of similar phenomena, through which they have gained expert judgment and shared disciplinary knowledge, ecologists have come to expect that there will be a relationship between root length and complexity and these variables. This example demonstrates that what is important or valuable to measure is determined by a researcher’s questions, but also by cultural values and beliefs within that researcher’s discipline, as well as the larger culture.

How quantitative data is generated involves qualitative judgments that are influenced by larger disciplinary and cultural values and standards. In addition, analysis of quantitative data also involves interpretation and explanation influenced by what the interpreter already knows about the topic, what she wants to suggest with the information, and for whom she is presenting the information.

For example, the researcher studying the effects of carbon and nitrogen on plant roots might compose a paper for a scientific journal; the audience for this piece would be other researchers interested in plant ecology. In this paper, the researcher could focus on the data and results, emphasizing how varying levels of carbon and nitrogen differently influence plant root length and structure. Because of her audience (other ecologists) and purpose (communicating her results), the researcher doesn’t need to explain the significance of her findings in the same way she would if, for example, she was explaining her results to the public. The same study, written to appear in Popular Science, would need to explain why it matters that carbon and nitrogen differently influence plant root length and structure. The researcher might, for example, hypothesize that overuse of synthetic, nitrogen-based fertilizers on agricultural plants, such as carrots, could reduce the size or density of the crop yield. A conventional farming audience, or an audience interested in sustainable farming might differently interpret this representation of the research differently than the scientific audience reading the scientific article covering the same research and data. The difference lies in the purpose for which the writer is interpreting the data and how she must represent what is at issue, given the conventions and expectations of the genre of scientific article, versus popular news coverage of scientific research.
How Persuasion and Data Interpretation InteractAlthough it would be convenient for data to ‘speak for itself,’ this is rarely, if ever, possible. One individual looking at the data makes different sense of it than another, although through persuasion and discussion, we can come to agree what the data means. In the field of engineering, as well as many other scientific and technical fields, acceptance of your ideas and interpretations are based not only on your technical facility—coming up with the best answer or solution—but on your ability to persuade someone else that your interpretation or idea is the best answer solution. As such, technical writing and communication becomes a vital way of not only understanding what it is you know (e.g. writing an explanation of your results so you understand what it means), but also transmitting that knowledge to others. Until knowledge is shared, until you have effectively communicated what you know to another party who shares the same interpretation of the facts or data, what you know can’t really count as ‘knowledge’ (Winsor, 1996, p. 5).