Imagine that you and a group of friends are trying to decide what restaurant to dine at tonight. What factors do you take into consideration when making this choice? Do you look at the menu? Maybe you read some reviews? Or perhaps someone in the group has eaten there before and offers feedback? With each of these exchanges, you are gathering information that will inform your choice.
We use information every day to make decisions.
Data is "information, especially facts or numbers, collected to be examined and considered and used to help decision making" (Cambridge Dictionary).
In STEM, data is especially important, as research projects are grounded in collecting, analyzing, and writing about data collected. Because of data's central role, it is crucial that data is meticulously recorded, ethically analyzed, and transparently reported.
Guiding questions:
- Who collected the data and how was the data collected?
- Research teams often create their own research methodology; their methodology should outline their data collection practices. These details are usually found in the "methods" section of a scholarly journal.
- Data is also collected by agencies such as the U.S. Government, international organizations, state agencies, schools districts, among many other examples (eg., U.S. Census, The World Bank).
- How is data provided and interpreted?
- Similarly to "material type," look to see what format the data is presented in. Data can be reported in many different formats, so make sure to choose a material type that aligns with your needs. For example, you may find a fact sheet that provides a general overview of health data, but it might be beneficial to find the original report that first presented the data.
- When reading a source that relies on data, probe into the methods section. Researchers should transparently describe how they analyzed their data.
- Does the data provided support the author's claims?
- Data should be collected and interpreted without bias. Data can be manipulated to support an author's claims, or data can be construed in graphics or other visual representations.
Ethics & Data
There are many ethical considerations that must be taken into account when collecting, processing, analyzing, or interpreting data. The considerations below offer a short list of ways to be vigilant throughout the data life cycle. Whether you are collecting your own data or reviewing another researcher's work, these are key areas to consider:
Data collection and preprocessing:
- Researchers should describe the data collection method employed and/ or refer to data sources used during the project.
- Ensure that data was collected using ethical practices such as informed consent and data privacy.
- Detailed information should be provided in regard to sampling and collection methods.
- If the data underwent any cleaning, limiting, refining, or filtering, the steps for this process should be transparent and well documented.
Data visualization:
- Data should be appropriately transformed into charts, graphs, and tables that align with the data type.
- Data presented visually should have a detailed legend that explain what data is included in the image and how the data is being interpreted. All axes should be labeled, and data should include a unit of measurement.
Data analysis and interpretation:
- Researchers should describe the statistical methods employed, and authors should outline why this specific statistical method was chosen for the project. This information should be provided in the article's "methods" section.
- Alongside the chosen statistical method, researchers should discuss the reliability and validity of their data analysis method.
- The results from the statistical analysis can then be used to interpret the data collected. Data can be interpreted in light of the research question or hypothesis being asked.
If you are consulting a work that does not appropriately address the considerations above, consider finding another resource for your project. If data is the foundation of the work, data should be grounded in an ethical research approach. To learn more about ethical data standards, consult the FAIR and CARE Principles (Nature, 2021).
Comic made by XKCD, and is available under CC-BY-NC 2.5