blog post 5

blog post 5

Representing primary sources as data can help observers comprehend the information associated with the source. For example, for our most recent project, we were instructed to log historical data of donations made in the late 18th century and early 19th century. Organizing the information as data would help readers comprehend the text more easily. Some people may have difficulty reading the original cursive handwriting. In addition, having the data logged into a spreadsheet helps people notice trends and make associations within the dataset. For example, multiple donations listed by the same donor could be split into multiple pages in the primary source. However, when placing data in a spreadsheet, related content can be output on the same page. The donor data could be displayed in graphs that focus on given variables. For example, graphs could be used to visualize the location of origin of the donors. They could also be used to measure the quantity of donations from a given donor. The donations could have also been categorized into different types. For example, some items were related to food, while others were related to construction. Having different formats of the data gives readers a better understanding of the given information. 

Having multiple sources of information, such as in the form of data, helps to spread awareness of the information. The data could also be used to support an argument or observation related to the primary source. One side effect however, is that data could be skewed to distort or exaggerate information. If someone wanted to exaggerate or minimize a contrast between different variables in a bar graph, they could change the width of the bars, or change the scale of the y axis. Fortunately, these decisions are merely possibilities. 

Based on the tidy dataset research of Hadley Wickham, the data could be more neatly organized when modifying the structure of the information from the primary source. Thorough organization of data can drastically improve the legibility of the data. According to Wickham’s principles of tiny data, each variable forms a column, each observation forms a row, and each type of observational unity forms a table. He considers these methodologies to be ideal for organizing tidy data. In relation to tidy data, like Hadley Wickham, different people may have different interpretations of tidy data. For example, when my group and I, for the folio tidy data project, transcribed the donor data, we would discuss how to organize the data. At one point, we discussed whether or not to store multiple donations from the same donor into the same row. After the collaboration activity was over, at the end of the lecture, we continued to work independently. When I submitted my work, I felt that the data was neatly organized. After my submission, someone else had further rearranged the data. So Henry Wickham’s principles of tidy data are a nice template. However, I feel that the ideal organizational design may vary, based on the creator of the dataset.

Leave a Reply

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