Table of Contents
As a researcher, you surely come across the words “research data” countless times during a normal working day. They are so familiar to you that it would be rather impossible to even explain the meaning of “science” or “research” without using them.
But sometimes the things that surround us, and that form our own reality, are the ones that make us stutter the moment we are asked to define them. This is definitely the case with the phrase “research data.” We live with it, work with it, but do we really know what it means?
What is research data?
Roughly said, research data is information coming from observations and experiments which validate research findings. Although the term “research data” is commonly thought to relate exclusively to science, that is not entirely true. Other disciplines, like arts or philosophy, also analyze data in order to produce their own kind of results.
Another common mistake is to think that EVERYTHING that ends up in the hands of a scientist is considered research data, but this is also untrue. Not just any type of information is considered research data – and furthermore, not every research data is sharable for the readers. This is especially important, since most journals and funders require data availability statements.
Although you can always state that the data underlining your work is not available for sharing due to reasons outside your control, we recommend using shareable material whenever possible. You can de-identify personal records in order to render data closed to the public into shareable files, for example.
The table below shows some examples of data types and what is generally considered “research data,” or not, by the majority of institutions and/or funders. Note that these may differ from country to country or institution to institution. We recommend that you carry out your own research, and contact your partner university, funder or organization for exact information concerning your situation.
Research Data Formats | Not Considered Research Data | Not Shareable | |
Documents/spreadsheets |
X |
||
Notebooks/diaries |
X |
||
Questionnaires/surveys |
X |
||
Codebooks |
X |
||
Experimental Data |
X |
||
Audio-visual files |
X |
||
Image files |
X |
||
Sensor readings |
X |
||
Test responses |
X |
||
Physical samples |
X |
||
Models/algorithms |
X |
||
Content analysis |
X |
||
Interviews/focus group recordings |
X |
||
Methodologies/workflows |
X |
||
Preliminary analysis |
X |
X |
|
Drafts of scientific work |
X |
X |
|
Plans for future research |
X |
X |
|
Peer reviews | X |
X |
|
Communications with colleagues | X |
X |
|
Information protected by law |
X |
X |
|
Trade secrets/commercial information |
X |
X |
|
Information that would constitute invasion of personal privacy (personnel and medical) |
X |
X |
|
Unpublished confidential material held by a researcher |
X |
X |
Even for the sake of progress, confidentiality and protection of individuals must be preserved. Even if you use this kind of protected data to support your findings, they are not properly considered as research data and therefore cannot be used or shared to validate research results.
What is a data analysis in research?
Data analysis is actually as important and mind-consuming as writing a manuscript itself. This is the phase where all of the collected data is examined, filtered, summarized and, most of all, interpreted to translate into actual findings. By interpreting and analyzing data, researchers want to find patterns and relationships by using analytical and logical reasoning.
Knowing exactly what to extract from data – and often very large quantities of it – definitely determines the outcome of the research. It is normal that newcomers to research feel overwhelmed while managing research data. It may be difficult at first to identify what really matters and what is secondary and/or distracting. Answering a few simple questions may actually help the process:
- Can I interpret this data in a way that helps me answer my research question?
- Is this data within my theoretical framework?
- Do context and circumstances in which the data was collected fit in the scope of the research?
- Are there biases that could compromise findings and results?
You may have heard more experienced researchers say that “data speaks to them” and you can’t help thinking “How does she/he do this?,” and “What exactly does this mean?” It actually means that, with experience, it will get easier and faster for you to establish relationships between data, along with meaningful patterns and common themes. Until then, there is no other way than going through piles of data over and over again. As time-consuming as it is, don’t delegate this task to anyone else. It is important that the researcher knows every detail and nuance of his or her data to help in future decision-making.
What is data reduction in research?
Of course, large quantities of scattered research data become impossible to analyze. Data reduction is the process of organizing, selecting and transforming this data into intelligible formats for further examination and display – that is, suitable for appearing in field notes or transcriptions, for example. Of course, this demands a lot of decision-making, focus and a huge capacity to simplify uncategorized material. A good example of data reduction may be rendering field guides or diaries in tables and schemes, where data is highly summarized to highlight similarities and differences between patterns.
Preparing a paper to publish is a huge task. Time and self-management are essential to deliver high-quality work. With Elsevier’s review and text editing services, you will have more time to manage data for your next project, with the guarantee of a top-quality text for publishing. If you’re not 100% satisfied, we’ll give you your money back. All this, plus a personalized cover letter and as many rounds of language review that you might need. Use the simulator below to check the price of your manuscript; simply enter the number of words.