Now when we begin to venture into the topic of the research and methods involved in gathering your data. Its important to briefly go over the three principles of tidy data. An also best methods for organizing research. An how you might implement both tidy data and organizational research methods into your own scholarly works. But I am not a tech person like a good majority of people, I will use some more common reference so I don’t lose anyone.
Well what is tidy data? Tidy data, without seasoning it up, is the standardization of research data. Mostly so other people can readily understand your data. Imagine it similarly to how a MLA or Chicago style paper is formatted for formal paper writing. So that way the mapping to the next reader can have a idea as to where things are or at least a “norm” for a layout. Now for tidy data, it will be comprised of Data Structure, Data Semantics and Tidy data (organization). Each variable forms a column and each observation forms a row along with each type of observation unit forms a table. So you can see now, you are really starting the footprint for a little database. Which is why we need a standard way of mapping the meaning of a dataset to its structure. Dataset equates to the tipping scale of balance between tidy and ideal on one end and messy on the other end. The tipping depends on how the rows, columns and tables are matched up with observations, variables and types.
So in my opinion the creation of a tidy database aesthetics aren’t too complex, as long as you don’t overthink it. So best methods and also how I would approach, would be don’t overtly complicate your database. When I create a dataset as a individual, I ask “would it pass scrutiny test”. But not the legal kind, I ask the Mom scrutiny test. Meaning if you, gave your mom a quick run down on the format and walked away. Would they be able to find that piece of information in your dataset with ease? An that is just it. Tidy data doesn’t need to be complex data, or overtly detailed in it information. But contain the appropriate amount of relevant data to support the goal of the research endeavor. So the opposite can happen if too much data is present and you can have a “ Death By Numbers” effect on the target audience, where their attention begins to tailor off. Especially with the Information Age, people are getting their information faster and faster and why should they fuss with a finding this buried information when I can just search engine it. So its important to hit the relevant information in your data. An thus also giving you malleability and palatability with your clean data to use, either in a condensed presentation form. Or more fully expanded version, like when incorporating your hard work into other databases. Which brings in another method for organizational research, Manipulation. Or consolidation and clean up of your data, adjust your database toward your audience , ethically of course. Ask do I need a 1 TB hard drive worth of data for a one hour class. Or filter out some of the irrelevant data points that may only serve to distract the audience. Which is why I saved visuals for last. As I believe it is the slippery slope of the three. As some visuals can be a delight and be a nice break from all the black and white numbers. Now mastering visuals truly is an art form. As too many can come off as unprofessional. You are going for palatable, not sexy (meaning don’t over do it). For example a bright multi-colored spreadsheet might be okay as a informal work schedule. But could be an eye sore and questionable in say, a peer review panel or for your later academic career if you ever need research funding.