Home    Basics Search Strategies EResources Things to Try Contact Linda Hansen Things to Try on Your Own: BA 3129 More Than Keywords and Google®: Search Skills and the Research Process

Related resources:
Public Access Full-Text

Related documents:
Evaluating Sources and Resources
Critical Thinking

What This Session is About ...

• The search process which forms part of the foundation of research
• technical skills
• understanding of data / information organization
• application of such skills and understanding
• Factors influencing the availability, accuracy, and nature of data / information -- what we search, where we search, how we search, when we search
• Selection of efficient and effective "paths" to data / information
• Selection and evaluation of data / information

Sources of Knowledge

• Faith / revelation
• Tradition / authority
• Intuition (feelings)
• Intuition (unexplainable grasp)
• Reason (rationalism)
• Experience (empiricism)
• Practice (pragmatism)

Diderot's Three Means

• Observation of nature
• collecting facts
• Reflection
• combining facts; analysis
• Experimentation
• attempting to verify our combination

Rule 1: Data (is not) Information

• Data
• "a representation of facts, concepts, or instructions presented in a formalized manner, suitable for communications, interpretation, or processing by humans by automatic means" [Information Warfare: Legal, Regulatory, Policy and Organization Considerations for Assurance, July 1996; ANSI X3.285]
• Information
• "meaning assigned to data by means of conventions applied to that data" [OECD Guidelines for the Security of Information Systems]

Information ...

• Provides a measurement of the degree of uncertainty in the recipient -- an indicator of possibilities
• supplies a finite number of solutions to any given problem
• Alters the perception of the recipient forever, causing a different reaction when confronted with the next information event
• essentially, the act of learning
• Is an activity, lifeform, relationship

Communicating Information

• Memory
• mnemonics
• loci (place) and imagines (images) -- two access points (location and look) to trigger the memory
• Written language

Rule 2

"The Analytical Engine has no pretensions whatever to originate anything. It can do whatever we know how to order it to perform. It can follow analysis; but it has no power of anticipating any analytical relations or truths. Its province is to assist us in making available what we are already acquainted with." [Ada Augusta, Countess of Lovelace, Note G, "Sketch of the Analytical Engine"]

Rule 3

In football, as you proceed down the field toward the goal, the shape of the field changes, getting shorter and wider. As you approach the end of the game, the rules change. Thus, your strategy for success must alter with the nature of the game, if you expect to win.

A good search works in a similar fashion.

Rule 4

It is called a SEARCH engine, NOT a RESEARCH engine ....
There is a significant difference between searching and researching.
Searching is what I do when I look for my wallet.
Find the wallet and the search is complete.
Search engines are designed for the completion of searches.
We know the search is complete because we already know the answer (what we are looking for).

When we don't already know (or anticipate) the answer, our search becomes research, and suddenly, the results from the search engine are not nearly as "good" ... as "useful" ... or as easy to accept.

A Google search on "canadians live within * kilometres of the" reveals just how bad the results might be ...

Searching Dilemmas

• Size / Completeness of the pool
• Link redundancy
• Speed
• Security / Privacy for the user
• Stability
• Identification of searcher
• Indexing controls
• Assistance
• Geography v Content
• Cost
• time, money, human resources
• Characteristics of the information
• age, format, accuracy
• Legal controls
• intellectual property considerations, personal privacy
• Government censorship
• sovereignty, cultural goals, public safety, national security

Research is all about asking questions, figuring out where to look, creating functional searches, finding a range of answers, analyzing those answers, asking new questions, deciding whether to keep looking in the original spot or looking somewhere else, creating new searches, finding a new range of answers, analysing those answers ... and so on.

There is a profound difference between finding something vaguely relating to your topic and finding material which allows you to support, critique, analyze, clarify, redesign and argue a predetermined and cogent hypothesis in a balanced and thoughtful manner -- and in so doing, use data and information to create original thought.

It begins with ... A Search

• Centred around: a theme, an idea, or a topic
• related to a course, a job, a project, an interest
• Reflecting: the over-arching purpose
• to find data / information
• to find "the answer", "an answer", or "answers"
• to create / assess / argue recommendations
• to create / assess / argue conclusions

Expectations

• Specific to requirements, needs, desires
• Specific to search / Specific to research
• gather facts, problems, solutions, data and speculation
• gather several (supported / supportable) alternative perspectives
• organize material gathered, analyze, respond to the analysis
• design new searches, gather more material OR make recommendations OR draw conclusions

The Expert Information User [Dillon and Schaap]

• Displays cognitive integration beyond the actual word, sentence or picture
• acquires (understands) the structural representation of layout and organization of information
• Follows embedded cues, rather than reading every word
• Displays (and utilizes) an ability to "sense" where s/he is in information space

Functional Search

• Requires technical skills / technical knowledge
• Deals with the query - retrieval problem
• Uses appropriate terms
• uses uncommon (discipline / topic specific) words
• eliminates noise
• uses boolean/positional operators; truncates or masks as appropriate
• Considers the appropriate information structure / container
• Uses the appropriate search engine

Influencing a Search ...

• Technical elements
• defaults
• field and field limiters
• Boolean (logical) operators
• positional (proximity) operators
• truncation, wildcards, stop words
• indexes, thesauri, controlled vocabulary

The Query

• Queries are often viewed as a single word, word combination, or phrase
• Goal is unified retrieval
• unbounded by artificial means (technology, access method, or format)
• To achieve unified retrieval, must view the entire object as a query

Relations

• In controlled circumstances, the query can include the relationship between object and description
• subject content (major and minor)
• applicable / available aspects
• type of research presented / pursued
• geography relating to: content, authors, publishers, place of publication
• publication type (self-designation, adapted by user)
• container
• age of: content, publication forum

Relations ...

• In uncontrolled circumstances (the internet), the human searcher must attempt to create the relationship, expressing it in terms the computerized searcher is capable of understanding

In controlled databases, field tags facilitate:

• finding the object itself ...
• a search on Charles Dickens as an author, will find items written by Charles Dickens
• a search on Oliver Twist as a title, will find the novel Oliver Twist
• ... and things about the object
• a search on Charles Dickens as a subject, will find items written about Charles Dickens
• a search on Oliver Twist as a subject, will find items about the novel Oliver Twist and about the character Oliver Twist

... but only if we use them ...

Uncontrolled databases ...

• Make NO distinction between:
• the object and things about the object
• objects with the same spelling and differing meanings
• objects with the same meaning but described in differing terms
• Do not account for:
• hierarchically broader, narrower, and / or related objects
• Search only according to their algorithms
• may not search what the user tells them to search and may not conduct the search to user specifications

Saying What You Mean:

"Then you should say what you mean," the March Hare went on.
"I do," Alice hastily replied; "at least I mean what I say, that's the same thing, you know."

"Not the same thing a bit!" said the Hatter. "Why, you might just as well say that 'I see what I eat' is the same thing as 'I eat what I see!'"

Influencing the Search ...

• Language elements
• keyword choices
• controlled vocabulary (thesauri) availability
• terms which may be unsearchable / censored / influenced by political correctness
• Language choices skew the returns

Selecting Terms

• Uncontrolled database / no controlled vocabulary
• no term standardization
• no hierarchical control
• Uncomprehended / unknown / unusual / inadequate / unhelpful field searching / indexing
• Based on:
• experience; knowledge of subject; knowledge of presentation of subject; standard thesauri; guess

The Lessons of Amazon

• Statistically Improbable Phrases (SIPs):
• global social body, disciplinary markets, immaterial workers, horror statistics, political recomposition, conflicting value practices, reproduction loops, living labour capacity, disciplinary trade, organisational reach, endless rat race, struggling subjects, productive nodes, other value practices, labour commanded, boundless accumulation, immanent measure, different value practices, unwaged workers, neoliberal governance, shadowy projections, corresponding social relations, market value system, unwaged labour, money circuit

Lesson 1

• What we call a keyword will not necessarily produce the item we are searching for even if the word used is "key" to the idea of our search
• keywords are not concepts; keywords are not subjects; keywords are not inclusive
• keywords are simply words which (may) appear on the page being returned

Lesson 2

• The fact of the appearance of the word is no guarantee that the object returned will be useful or even relate to our search

Keyword ...

• May depend on cultural, societal, political norms
• May be descriptive but not prescriptive
• May be common
• May or may not be field limited or limitable
• May confuse the user with regard to what was searched
• Similar may not imply exact
• john smith; smith john; smith j a; smith john a

Keyword ...

• May be a homograph for one or more items
• seals/seals/seals/Seals; IRA/IRA/Ira; banks/banks/Banks; etc.
• In a particular document / discipline one word or phrase may be interchangable with or equal to another which, elsewhere, may carry a different meaning or may be conceptually different
• ob: obstetrics OR organizational behaviour (OR organisational behaviour OR organizational behavior etc.)

Keyword ...

• Will likely be non-inclusive, non-exclusive
• searching New Brunswick likely will not search for all items about Saint John or the Maritimes, although it might well produce New Brunswick, New Jersey, the University of New Brunswick, and Rutgers University
• searching America likely will not search United States or California, although it might well produce South America, Latin America, America the Beautiful, America Ferrera, and likely eliminate Americans

Mapping Information

• Unstructured
• form of string or pattern matching (how most people use search engines)
• where let's eat children is the same as let's eat, children and community college is the same as college community
• straight matching (find in page; search tools for files with date matching)

Mapping ...

• Controlled structure
• controls relationship of one piece of information to another (much as grammar does in a sentence)
• governs relationship of one piece of information to another (as in an outline)
• illustrates relationship of one piece of information to another (as in a site map)

Mapping ...

• Controlled vocabulary (thesauri)
• Vocabulary browse (alphabetic, numeric list)
• often called an index

Subject Access

• Assignment of subjects / accompanying aspects derived from subject indexing
  • conceptual analysis -- "aboutness" of item; why it has been included; how users might be interested in it
  • translation -- converts conceptual analysis into a set of terms often called a controlled vocabulary

Controlled Vocabulary

• Set words or phrases used by indexer to describe the contents and typology of documents represented by the citation
• May be esoteric and require subject knowledge to use accurately, effectively
• Will vary from database to database even when the subject covered remains the same
• May be a lag time before new concepts / words / phrases appear in controlled vocabulary
• eg "aids" for human immuno-deficiency syndrome

Purpose

• To control synonyms through the use of see and see also references
• To distinguish among homographs
• To link terms hierarchically and/or associatively

A (butterfly) rose by any other name would be Rosa chindensis mutabilis

Skewing the Returns

• The language of the knowledgeable
• professional / scientific v lay terms: opac v library catalogue; closed reduction v put the broken bone in a cast; adolescents v teenagers
• The language of the form (container)
• instructions for authors; abstract (periodical literature)
• coordinates (maps)

Skewing the Returns

• The language of the pre-determined response
• terrorist / guerilla fighter / freedom fighter
• traitor / patriot
• religious fanatic / martyr / saint
• lunatic asylum / mental institution

We have to create the appropriate relationship between the object and the type / format / nature of the material we desire, based on experience, knowledge, skills and intuition.
We have to choose the information infrastructure (container) and the search engine to meet the parameters of that relationship.

Possible Approaches

• By subject -- I'm looking for business journal articles
• By format -- I'm looking for journal articles in business
• By name -- I'm looking for ABI Inform; Business Monitor International
• By accessibility -- I'm looking for full-text journal articles ... or ... I'm looking for material I can get from home ...

Influences

• Whether you have "named" what you are looking for
• Whether a single element is present
• Whether the source / resource seems accurate and current
• Whether a single element is present
• Whether an item "strikes your fancy"
• Whether it is convenient
• Whether it costs money

Information Infrastructure (Containers)

• What type of information are you looking for -- and how is it likely to be loaded?
• books, journals, newspapers ...
• maps
• raw data
• web sites (blogs, rss, individual, portal ...)
• ftp sites
• files within databases
• graphics

Choosing the Search Engine

• Based largely on experience
• Based on knowledge of searcher operation
• ie indexing abilities/disabilities of chosen searcher limiters, operators
• Based on (likely) structure of the information sought
• Based on information form and content sought

The Retrieval Problem -- Dealing with vast amounts of information means:

• Filtering
• user limited returns (decision rules and universes create deliberate choice / unintentioned results)
• technology limited returns
• Customization
• technology customized returns (intelligent agents, bots, etc.)
• user customized returns (deliberate, default, unintentional)

Precision v Recall Ratios

• Precision ratio: the ratio of useful items to the total items retrieved
• Recall ratio: the extent to which all useful items are found
• Precision is the ability to avoid useless items while recall is the ability to find useful items
• Measure of success of a search would be a high level of precision combined with a high level of recall

Decision Rules

• Assign attributes (values)
• Cost-benefit analysis
• Choice behaviour (interpretations)
• Reasoned action (reactions to interpretations)
• Duration (time)
• Used for:
• classification
• subject/aspect determination
• context
• content

Universes

• Govern decision rules
• Are created by:
• experience
• knowledge
• skills
• intuition

Value of Data / Information Searched / Retrieved

• True
• False
• Meaningless
• Combination

depending upon /affected by the rules and the universes

Retrieval

• Relevance
• content relevance
• term relevance
• Magnitude (Postings threshold analysis)
• Tactical review
• NOT a strategic review
• Terminology review

Evaluation

• Consider: authority, currency, relevance, scope, audience, validity, degree of doubt
• Identify: missing elements
• too much of the same thing
• non-responsive to hypothesis
• Extrapolate, identify areas of follow-up as retrieval occurs
• Maintain: search history, order of the search, notes for analysis

Evaluation: Peer Review

• Valuing the method whereby content was created
• was the experiment properly constructed?
• was the data properly gathered and manipulated?
• does the experiment appear to be replicable?

Critical Thinking

• Recognize leading proponents and theories
• identify controversial aspects of standard theories
• Understand statistical / other methodology
• Summarize accurately, analyze fairly, present ethically
• in support of, or in opposition to, own work and that of others
• Affects
• learning, reasoning
• analysis, evaluation
• decision-making, problem solving
• output, presentation
• quantity, timing
• search, research, professional, personal choices

Things We Can't Control

• Whether the information ever existed
• Whether the information is extant, available and accessible
• Whether the search engine has indexed it, and when
• Priorities assigned by the algorithms
• Weight and distribution of information which survives (Boorstin)
• the unread and ignored [to be discovered later]
• the durable -- unable to be removed or displaced easily
• the collected and protected
• the precious -- high intrinsic value
• the academically classifiable -- the dignified, the taught, the politically correct
• the controversial -- especially relating to religious controversy
• the self-serving -- or self-aggrandizing
• the victorious -- the success bias -- those who win get to write the history
• the rosy and the comfortable -- as opposed to the truth of daily existence

Identifying and Acquiring Information

• Consider:
• ... the original topic / thesis statement / research questions / hypotheses
• ... any secondary topics or refinements to the original topic / thesis statement which may result from your searches
• ... aspects of the topic which may narrow or broaden the search significantly
• ... balance / bias

The "Topic"

• Physical requirements
• product required (essay, report, case study ...)
• length
• elements: particular sources, types of sources; bibliography; footnotes ...
• format
• time considerations (due date)
• Intellectual requirements
• topic: subject / aspects (geographic, historic, gender, economic, legal, political, social, ethnic, psychological, cultural, technological ...)
• topic: course (tourism, marketing, human resource management ...)
• topic: theoretical / practical

Consider ...

• Closed questions
• usually have a known, standard, and / or politically correct, answer
• who, what, when, where
• sometimes how many, how much
• Open questions
• often result in or reflect controversy, opinion, and force you to think critically
• how, why

Thesis Statement

• An interpretive statement about the topic, preferably one which will allow you to produce an argument, and place your topic appropriately in a wider context
• Central assertion which bounds those ideas, questions, and statements you intend to discuss and argue

Purpose

• To guide the development of individual research questions and hypotheses
• To provide the framework for your agenda
• you are attempting to persuade your reader of something

Statement

• May include:
• an accepted / established truth or two (the basis or foundation of the topic)
• where the differing perspective or potential controversies appear (either obviously or not)
• path(s) you intend to pursue to support (or not) your opinions on the topic

An Argument

• In academic terms: controversial aspect(s) of, or statements about, a topic which a scholar (student or faculty) might attempt to support or refute through written persuasion backed by evidence
• Type of argument presented, the nature of the research undertaken, and the acceptability of the evidence often depends upon the intended audience

Context

• The purpose of pursuing the topic and the argument
• remember, we are not trying to "prove" something -- we are offering an argument, which we will present in a balanced and accurate fashion, backed by balanced and accurate information and valid, significant data

Research Questions

• Focused inquiries
• specific questions the project / paper intends to answer
• Guide:
• research design, research methodology
• Influence:
• choice of searcher, search methodology, resources and sources
• Not every question is a research question
• lack of probative value
• factual only, lacking novel associations and causal relations

Hypothesis

• Proposed, untested, possible answer to a research question

Original Research

• Adding to a body of knowledge by discovery of new facts and exposition
• Qualitative
• observation; interviews (structured, in-depth); focus groups (interviews and observation)
• document studies; case studies; key informants
• action research; behavioural research; field studies
• Quantitative / Statistical
• sampling, measurement, data analysis, validity
• Surveys

Experiment Design

• Hypothesis
• attempt to predict causal relationship between / among variables
• Causal relationships
• how much cause results in how much effect
• dependent and independent variables
• Validity
• internal (in the study design)
• external (extent to which results may be generalized)

Bibliographic / Literature Review

• Confirming previously expostulated opinion
• Postulating new conclusions through re-interpretation of previously presented data/info (data mining?)
• Reviewing current / historical opinion / research
• Requires the ability to perform functionally correct searches
• Requires the ability to think critically

Where do we go from here?

• What you know
• What you need
• Resources (time and money, in particular)
• Expectations

Top of Page

This page copyright, created and maintained by Linda Hansen.
Comments and suggestions to: lhansen16@gmail.com
Created: 2013/10/16 Last updated: 2017/09/13
Terms of Copyright
This document: ...tutor/ba3129.htm