Leveraging Observation for Research in Libraries and Information Centers

When a student walks into a library, hesitates at the catalogue terminal, gives up after two failed searches, and quietly heads for the nearest familiar shelf, no questionnaire will ever capture that moment. The student will not report it, because they may not even remember it. This is exactly where the observation method earns its place in library research. Instead of asking people what they do, researchers watch what they actually do. For libraries and information centers, this shift from self-reported opinion to recorded behaviour has become one of the most reliable ways to understand users and improve services.

Why observation works so well in libraries 🔗

Observation is the systematic watching and recording of behaviour in its natural setting. A library is almost an ideal natural laboratory for this. People come in with real tasks, real deadlines, and real frustrations, and they behave as they normally would rather than as they think a researcher wants them to.

The big advantage is that observation collects data in a real-world environment, not a controlled one. Surveys and interviews depend on what users are willing and able to articulate. But many library behaviours are non-verbal or happen too quickly to describe later, such as where someone chooses to sit, how long they scan a shelf, or whether they ask for help. Watching directly captures these unspoken actions that other methods miss entirely.

There are two broad approaches a researcher can take. In participant observation, the researcher takes part in the activity being studied, perhaps working at the reference desk while noting interactions. In non-participant observation, the researcher stays an outsider, simply watching and recording. Libraries also lend themselves to structured observation, where predefined categories are used to record behaviour systematically, and unobtrusive observation, where users are studied without any intervention.

Observing user behaviour with catalogues and resources 🔗

One of the most practical applications is studying how users interact with the library catalogue and physical collections. The catalogue, often the OPAC (Online Public Access Catalogue), is the gateway to the collection, yet it is frequently a point of failure. Observation reveals where that failure happens.

Watching how people search the catalogue 🔗

A classic qualitative study of public library users combined questionnaires, interviews, and direct observation to understand catalogue interaction. It found that search behaviour and the user’s confidence varied sharply across three search types: known-item searches were characterised by simplicity, area searches by a need for speed and convenience, and unknown-item searches depended heavily on the user’s ability to generate the right terms. None of this nuance would surface from a simple satisfaction rating.

Observation paired with usability testing is even sharper. In OPAC usability studies, researchers watch users attempt set tasks and record where they stumble. One evaluation of a school library catalogue using observation and the think-aloud method recorded a task success rate of just 49.3%, with problems traced to library jargon and poor error prevention. That number is a direct call to redesign.

Tracking the use of physical books 🔗

Observation is not limited to screens. A simple but clever technique for studying physical book use is the Dot-on-the-Spine method, where small coloured dots are placed on book spines and the accumulation of dots over time indicates how often each item is handled or borrowed. Books with many dots show high engagement, while those with few reveal neglected sections. This turns a quiet, invisible behaviour into visible, countable data.

How the library environment shapes behaviour 🔗

The physical environment of a library, its layout, comfort, lighting, and accessibility, has a measurable effect on how users behave and how satisfied they feel. Observation is the method best suited to mapping this relationship, because environment and behaviour reveal themselves through where people go and what they avoid.

Seating sweeps and space utilisation 🔗

The most established technique here is the seating sweep. A researcher walks a fixed route through the library at set intervals, recording who is sitting where and what they are doing. This ethnographic approach was used to map the social activity space of central public libraries in two large Canadian cities, producing evidence directly useful for facilities redesign and planning around real usage patterns.

The findings from such studies are often surprising. A multisite observational study across five campuses used seating sweeps and questionnaires together and found remarkably similar usage patterns everywhere, with academic study remaining the most common activity despite the popular image of the modern library as a social hub. Evidence like this helps libraries justify their role in student learning to their parent institutions.

What makes some seats more popular than others 🔗

Direct observation consistently shows that environmental quality drives seat choice. A space utilisation study at one academic library found that seating near windows and in well-lit areas was strongly preferred, while isolated sections were barely used at all. A study of an academic library’s reading room went further, drawing on personal-space theory and the human tendency toward “wall-hugging” to explain why study carrels with individual outlets and lighting were highly preferred over open tables.

Modern technology has extended observation far beyond the clipboard. At the UCL Bartlett Library, researchers gathered over 1.3 million passive infrared sensor readings across a year and used machine learning to model seat occupancy, finding that seating behaviour is shaped mainly by environmental controllability, such as personal lighting and fresh air, and by distraction management, such as quiet and visual privacy. Sensor-based observation like this is unobtrusive and continuous, removing much of the bias that comes with a visible human observer.

Understanding information-seeking behaviour 🔗

Information-seeking behaviour is the heart of library and information science, and observation gives researchers a window into how people genuinely search, evaluate, and use information. The value lies in catching the gap between what users say they do and what they actually do.

Observation reveals the unspoken parts of the search process, such as how a user scans a list of search results or how long they spend reading a single document before moving on. These micro-behaviours signal whether a system is working. A user who scans the first three results and then abandons the search is telling you something important, even if they never say a word.

Combining observation with digital traces 🔗

Digital systems quietly record behaviour, and this data complements direct observation beautifully. OPAC search logs reveal how users formulate their queries, and database usage statistics show which resources are accessed most often. Transaction log analysis is valued precisely because it is an unobtrusive method that captures the actual activity of online users at a large scale.

The strongest research designs combine these. Watching a handful of users closely explains why a behaviour happens, while logs show how often it happens across thousands of sessions. Together they answer questions that neither could answer alone.

What observation cannot do, and how to manage it 🔗

Observation is powerful but not flawless, and good researchers plan around its limits. The most discussed problem is the Hawthorne Effect, where users change their behaviour simply because they know they are being watched. Unobtrusive and sensor-based methods reduce this risk.

The method is also time-consuming and carries a real risk of researcher bias, since the observer must interpret what an action means and may misread it. There are ethical considerations too. Before observing, a researcher must confirm the site is open to observation and that any privacy concerns or permissions are settled. Running a small pilot study first helps test the recording tools and confirm the approach is feasible within the available time and budget.

This is why observation is rarely used alone. The most reliable picture comes from combining qualitative observation with quantitative tools so that deep insight into motivation meets large-scale data about frequency.

What do you think? If you ran a seating sweep in your own college library this week, which behaviour do you suspect would surprise you most, and would you trust what people told you about their study habits more than what you actually saw them do?

References

1. https://www.sciencedirect.com/science/article/abs/pii/S0740818803000495
2. https://eric.ed.gov/?id=EJ1073902
3. https://ideas.repec.org/a/bla/jamest/v51y2000i8p757-773.html
4. https://www.researchgate.net/publication/401696718_Usability_Evaluation_of_a_School_Library_OPAC_Using_Heuristic_Evaluation_and_User_Testing
5. https://www.academia.edu/2799174/Academic_library_seating_a_survey_of_usage_with_implications_for_space_utilisation
6. https://www.sciencedirect.com/science/article/abs/pii/S0099133320300343
7. https://pmc.ncbi.nlm.nih.gov/articles/PMC12319776/
8. https://www.researchgate.net/publication/222433291_The_Use_of_OPAC_in_a_Large_Academic_Library_A_Transactional_Log_Analysis_Study_of_Subject_Searching