Every research project, library survey, or policy decision rests on a single foundation: data. But data is not just a pile of numbers waiting to be crunched. It has a character, a structure, and a set of qualities that decide whether it will actually be useful or whether it will mislead you. For students of Library and Information Science, understanding what data is made of, and what makes it trustworthy, is the difference between drawing sound conclusions and building arguments on sand. This post breaks down the nature of data and the core properties that give it value in research and decision-making.
Table of Contents
What we mean by the nature of data
The word “data” comes from the Latin datum, meaning something given. According to the Oxford English Dictionary, data are known facts or things used as a basis for inference or reckoning. That definition sounds solid and permanent, but the reality is more slippery. Data may look like fixed truth, yet it is often elusive and even temporary. A figure that accurately describes a situation today may not hold tomorrow.
The nature of data refers to its fundamental qualities: the form it takes, how it behaves, and how it connects to wider knowledge. Researchers like Nicholas Walliman describe data as sitting within a hierarchy that moves from the general to the particular and from the abstract to the concrete. In simpler terms, raw facts become meaningful only when they are organised and interpreted. As the textbook framing puts it, data are raw facts that have not yet been processed to reveal their meaning, while information is the result of processing those facts. This is why the same dataset can be useless to one researcher and gold to another.
In both the natural sciences and the social sciences, data generally appears in three broad forms: numerical, descriptive, and graphic. Recognising which form you are dealing with is the first step in deciding how to collect, store, and analyse it.
Numerical data
Numerical data is expressed in numbers. In the sciences, nearly everything is captured through measurement and recorded as numerical values, which is why this is the most common form of data in quantitative work. Think of the number of books issued by a library in a month, the footfall in a reading room, or the average time a user spends searching an online catalogue. Numerical data is highly measurable and can be processed using statistical methods such as means, medians, and standard deviations. Because it can be added, averaged, and compared, it forms the backbone of quantitative research.
Numerical data itself splits into two useful categories. Discrete data is counted in whole numbers that cannot be broken into fractions, such as the number of students attending a workshop. Continuous data can take any value within a range, such as the exact time taken to retrieve a document. Harvard Library’s research guide notes that discrete data assigns finite whole-number values while continuous data can be subdivided within a range. Knowing the difference matters because it decides which charts and tests you can legitimately use.
Descriptive data
Descriptive data, also called qualitative data, is non-numerical. It captures qualities, opinions, experiences, and meanings rather than amounts. When a librarian interviews users about why they prefer e-books over print, the answers are descriptive data. This form is typically gathered through interviews, focus groups, observations, and open-ended surveys, and it usually arrives as words rather than figures.
Qualitative data has a lower level of measurability than numerical data, but it answers questions that numbers cannot. As researchers explain, qualitative research seeks to answer “why” and “how” by focusing on subjective experiences, while quantitative research answers “what” and “how often.” A reading-habits study that only counts borrowed books tells you the scale of usage; the descriptive feedback tells you the reasons behind it. The two together give a fuller picture than either alone.
Graphic data
Graphic data presents information visually through charts, diagrams, maps, photographs, and other images. A bar chart of monthly library visits, a pie chart of collection categories, or a heat map of shelf usage are all graphic representations. Graphic data is partly a way of communicating numerical or descriptive findings, but it can also be a primary form of evidence in its own right, such as a scanned manuscript or a digitised map in an archive.
The type of visual you choose is dictated by the underlying data. A pie chart or bar chart suits categorical data, while a histogram suits discrete data and a line graph suits continuous data. Picking the wrong chart can distort a finding even when the numbers behind it are perfectly correct, so graphic data demands as much care as the other forms.
How the nature of data shapes its usability
Why does sorting data into these forms matter so much? Because the nature of the data determines its usability. The form decides what methods you can apply, what tools you need, and what kind of conclusions you can defend. You cannot calculate an average from interview transcripts, and you cannot capture the emotional reasons behind a behaviour using a spreadsheet of counts.
Many strong research designs deliberately combine forms. A mixed-methods study on a new library system might track the number of logins, the duration of sessions, and user satisfaction comments all at once. Recognising what type of data you need before you start collecting saves enormous effort later. The nature of the research problem, not personal preference, should drive the choice of data, because the problem itself determines which approach will produce usable answers.
The essential properties of data
Identifying the form of your data is only half the job. The other half is judging its quality. Data quality refers to how well data is fit to serve its intended purpose, and it is the property set that gives data its real value in research and decision-making. Several properties stand out as essential: clarity, accuracy, amenability to use, and being the essence of the matter.
Clarity
Clarity means the data is unambiguous and easy to understand. Data that is muddled, vaguely labelled, or open to several interpretations creates confusion and invites error. If a survey question can be read in two ways, the responses will mix two different meanings, and no amount of clever analysis can untangle them afterwards.
Clarity also covers how data is represented and communicated. The way information is laid out affects how quickly and correctly people can act on it. Research on decision-making found that representational consistency affects the time it takes to reach a decision. Clear labelling, consistent formats, and well-chosen visuals all reduce the mental effort needed to use the data, which is why clarity is treated as a foundational property rather than a finishing touch.
Accuracy
Accuracy is the degree to which data correctly represents the real-world entities or events it is meant to describe. It is widely regarded as the cornerstone of data quality. Inaccurate data does not just weaken a study; it can quietly reverse its conclusions. A library that records the wrong circulation figures might cancel a popular subscription or over-invest in a neglected collection.
Accuracy has measurable consequences. Studies show that data accuracy has a direct effect on decision-making performance. Early data-quality researchers Ballou and Pazer identified accuracy as one of the first essential dimensions, alongside timeliness, completeness, and consistency, and later work confirmed that accuracy and completeness are the factors most important to decision-making. To protect accuracy, researchers cross-check evidence against multiple sources and review how the data was originally produced before trusting it.
Amenability to use
Data may be clear and accurate yet still be hard to work with. Amenability to use describes how easily data can be accessed, processed, and applied to the task in hand. Data locked in an unreadable format, scattered across incompatible systems, or buried without proper organisation has limited value no matter how correct it is.
This property overlaps with what data professionals call accessibility and relevance. Data should be easily accessible to those who need it yet secure enough to prevent unauthorised use, and it must be relevant to the problem at hand. Irrelevant data, even when accurate, can crowd out meaningful signals and push researchers toward wrong conclusions. Amenability to use is therefore about practical fitness: can this data actually be put to work without disproportionate effort?
Being the essence of the matter
The fourth property is more conceptual but equally important. Data should capture the essence of the matter, meaning it should genuinely reflect the core of what is being studied rather than peripheral noise. There is always a temptation to collect everything that is easy to measure, but volume is not the same as relevance. Good data isolates the variables that truly matter to the research question.
This is also where the distinction between data and noise becomes critical. Not every recorded figure carries meaning. Discussions on the nature of data stress the difference between genuine evidence and data versus noise, and the need to treat data with objectivity. Capturing the essence means filtering out distractions so that the dataset speaks directly to the problem, which keeps analysis focused and conclusions defensible.
Why these properties matter in research and decision-making
Together, these properties decide whether data deserves trust. High-quality data is essential for making informed decisions and ensuring the reliability of any process that relies on it, while poor-quality data leads to flawed decisions and unreliable systems. The stakes are real: weak data can waste scarce resources, damage credibility, and in regulated sectors even create legal trouble.
For LIS professionals, this is not an abstract concern. Libraries and information centres constantly make choices about collections, services, staffing, and budgets, and these choices are only as sound as the data behind them. In research, the same logic applies on a larger scale. High-quality data ensures that research outcomes are valid, actionable, and reliable, whereas low-quality data produces erroneous conclusions and misinformed strategies. Understanding both the nature and the properties of data is what allows a researcher to move confidently from raw facts to credible knowledge.
What do you think? When you look back at a recent assignment or project, did you give more attention to collecting a large amount of data or to ensuring its accuracy and relevance? And if you had to drop one property – clarity, accuracy, amenability to use, or capturing the essence – which would damage your research the least, and why?
References
- https://methods.sagepub.com/book/social-research-methods/n5.xml
- https://www.researchgate.net/publication/332818793_THE_NATURE_AND_TYPES_OF_DATA
- https://www.datamation.com/big-data/what-is-quantitative-data/
- https://guides.library.harvard.edu/c.php?g=1324910&p=9889267
- https://www.fullstory.com/blog/qualitative-vs-quantitative-data/
- https://uen.pressbooks.pub/uvumqr/chapter/4-3-data-organization-and-representations/
- https://www.researchgate.net/publication/283781996_Data_Quality_and_its_Impacts_on_Decision-Making_How_Managers_can_benefit_from_Good_Data
- https://pmc.ncbi.nlm.nih.gov/articles/PMC9912223/
- https://www.marshall.edu/irp/2023/10/04/importanceofdataquality/
- https://academic-publishing.org/index.php/ejbrm/article/download/2423/2092/5598
- https://www.teradata.com/insights/data-platform/data-quality-for-informed-decision-making
- https://www.philomathresearch.com/blog/2024/12/11/understanding-data-quality-in-research-why-it-matters-for-accurate-insights/
Leave a Reply