Ellis: Information Gathering Behaviour

Ellis (1987, 1989) carried out a study in which he used semi-structured interviews for data collection and Glaser and Strauss’s grounded theory for data analysis. His research resulted in a pattern of information-seeking behavior among social scientists that included six generic features:

  • Starting: comprising those activities characteristic of the initial search for information such as identifying references that could serve as starting points of the research cycle. These references often include sources that have been used before as well as sources that are expected to provide relevant information. Asking colleagues or consulting literature reviews, online catalogs, and indexes and abstracts often initiate starting activities.
  • Chaining: following chains of citations or other forms of referential connection between materials or sources identified during “starting” activities. Chaining can be backward or forward. Backward chaining takes place when references from an initial source are followed. In the reverse direction, forward chaining identifies, and follows up on, other sources that refer to an original source.
  • Browsing: casually looking for information in areas of potential interest. It not only includes scanning of published journals and tables of contents but also of references and abstracts of printouts from retrospective literature searches.
  • Differentiating: using known differences (e.g., author and journal hierarchies or nature and quality of information) between sources as a way of filtering the amount of information obtained.
  • Monitoring: keeping abreast of developments in an area by regularly following particular sources (e.g., core journals, newspapers, conferences, magazines, books, and catalogs).
  • Extracting: activities associated with going through a particular source or sources and selectively identifying relevant material from those sources (e.g., sets of journals, series of monographs, collections of indexes, abstracts or bibliographies, and computer databases).

Lokman I. Meho & Helen R. Tibbo: Modeling the Information-Seeking Behavior of Social Scientists: Ellis’s Study Revisited (PDF)


In other words … (Ellis 1993?):

  1. Starting: identifying sources of interest.
  2. Chaining: following leads from an initial source.
  3. Browsing: scanning documents or sources for interesting information.
  4. Differentiating: assessing and organising sources.
  5. Monitoring: keeping up-to-date on an area of interest by tracking new developments in known sources such as journals.
  6. Extracting: identifying (and using) material of interest in sources.
  7. Verifying: checking the accuracy and reliability of information.
  8. Ending: concluding activities.

Search as navigation

With the trend of minimalist and clutter-free design, sometimes it is assumed that users prefer to search above all else. But often that’s not the case. Typically, people use search if they know exactly what they’re looking for, or if they cannot find something through browsing.

Search should only be the primary navigation for a website if the site’s main function is to be a search engine. For example, Google, Bing, and job search boards can all use this approach. But in the context of an information-heavy site like a university, browsing is essential for increasing discoverability of content.

KATIE SHERWIN: http://www.nngroup.com/articles/breaking-web-conventions/

Search design patterns

10 major search design patterns:

  • Autocomplete/suggestions
  • Best (results) first: Relevance, Date, Popularity, Format, Personalisation, Diversity
  • Federated search (simultaneous search of different databases)
  • Faceted navigation
  • Advanced search
  • Personalisation
  • Pagination
  • Structured results
  • Actionable results
  • Unified discovery

Morville and Callender (2010). p: 81-130

Search Manifesto

  1. Search is a problem too big to ignore.
  2. Browsing doesn’t scale, even onm an iphone.
  3. Size matters. Linear growth compels a step change in decision.
  4. Simple, fast, and relevant are table stakes.
  5. One size won’t fit all. Search must adapt to context.
  6. Search is iterative, interactive, social, and multisensory.
  7. Increments aren’t enough. Even Google must innovate or die.
  8. It’s not just about findability. It’s not just about the Web.
  9. The challenge is radically multidisciplinary.
  10. We must engage engineers and executives in design.
  11. We can learn from the past. Library science is still relevant.
  12. We can learn from behavior. Interaction design affords actionable results.
  13. We can learn from one user. Analytics is enriched by ethnography.
  14. Some patterns, we should study and reuse.
  15. Some patterns, we should break like a bad habit.
  16. Search is a complex adaptive system.
  17. Emergence, cocreation, and self-organisation are in play.
  18. To discover the seeds of change, go outside.
  19. In science, fiction, and search, the map invents the territory (Note from somewhere elsein the book: maps hide more than they reveal…)
  20. The future isn’t just unwritten – it’s unsearched.

Morville and Callender (2010), p.20

David Ellis: Behavioural model of information seeking (8 primary behaviours)

Eight primary behaviour patterns in information seeking:

  1. Starting: identifying relevant sources of interest
  2. Chaining: following and connecting new leads in an initial source
  3. Browsing: scanning content of identified sourcves for subject affinity
  4. Differentiating: filtering and assessing sources for usefulness;
  5. Monitoring: keeping up to date of an area of developments in a given subject area
  6. Extracting: systematically working through a given source for material of interest
  7. Verifying: checking the accuracy and reliability of information.
  8. Ending: concluding activities

David Ellis, cited in Kalbach (2007, p.26)

Search results best practices

For easy reference, here is a list of some of the best practices that have been gleaned from different search results pages:

  • User should have easy access to the search box for follow-up searches
  • Where possible, search terms should be clearly indicated at the top, and in context in the results
  • Related sponsored links can be included below the search box, near the bottom, or on the right
  • Titles should be clickable and clearly differentiated from details
  • Visited links should be indicated
  • Pagination units should be visibly block-shaped and have a hover effect, to easily differentiate from one another
  • Related products, tags, or keywords should be displayed in a non-obtrusive section
  • E-Commerce sites should allow the “view” to be toggled between “list” and “grid”
  • Advanced search options should be easily accessible
  • Should allow re-sorting or filtering of results
  • Where possible, results pages should have RSS feeds or “subscribe” options
  • For complex interfaces, clear, easy-to-access search tips or instructions should be provided
  • Sorting and Filters should be JavaScript or Ajax-driven, where possible
  • Popularity or star-ratings should be shown for individual results
  • Include an option to increase the number of results per page
  • To monitor future improvements, request feedback from users after searches are conducted
  • If results span different sections of the website, indicate this by sub-headings or other dividers

Louis Lazaris: Search Results Design: Best Practices and Design Patterns

Faceted Search

  1. Decide on your filter value-selection paradigm—either drill-down or parallel selection.
  2. Provide an obvious and consistent way to undo filter selection.
  3. Always make all filters easily available.
  4. At every step in the search workflow, display only filter values that correspond to the available items, or inventory.
  5. Provide filter values that encompass all items, or the complete inventory.

G. Nudelman: Best Practices for Designing Faceted Search Filters

Confusion around sorting and filtering lists

“… many of the participants [in a usability study] said, “I am filtering by price,” while manipulating [the] Sort By control. After observing this phenomenon numerous times, it became clear to me that this was not merely a matter of a simple confusion of terms between filtering and sorting. Instead, it revealed a strong mental model of filtering by sorting that blurred the difference between these two modes of search results’ refinement.”

Greg Nudelman: The Mystery of Filtering by Sorting

site-search vs. web-search

    Why should site-search perform better than web-search

  1. smaller set of pages
  2. better handle on user’s intent
  3. prioritizing of documents is possible
  4. older documents can get a lower priority
  5. access to meta-data
  6. controlled vocabulary
  7. you can write summaries in your own words
  8. no spammers

From: Nielsen, J. and Loranger, H. (2006), Prioritizing Web Usability, Berkeley, CA.: New Riders, p.139-140