Qualitative Research

1
Qualitative Research

• Interviews
• Observation
• Focus Groups
• Diaries/Journals
• Triangulation of Multiple Methods
• draw on strengths of individual methods to
  examine phenomena from a variety of perspectives
  (e.g. interviews diaries participant
  observation)
• Case Study Methodology commonly used

2
Focus Group Interviews

• A small group of people (or several small groups)
  who have been chosen based on a particular
  characteristic
• The interviewer asks a small number of questions
  on a topic in order to explore the participants
  perceptions/ideas
• Heavily used in market research to test peoples
  reactions to a product/service
• More useful than questionnaires for in-depth
  responses
• Can be used with other techniques

3
Case Study Methodology

• May focus on individuals, organizations,
  departments/teams
• Purpose to understand phenomena within a
  particular context
• Cases to be studied constitute purposive sampling
• May use a wide variety of techniques (interviews,
  questionnaires, observation)

4
Observation a Survey Methodology

• Descriptive, quantitative research
• To examine/document the world as it is (What is
  happening? How often does it happen?)
• Typically uses checklists, log analysis, etc.
• Exploratory, qualitative research
• Ethnographic approach typically with
  interviews, etc. to document the natural setting
  where behaviours occur as part of larger
  investigation of individuals perceptions
• Typically uses video/photos, etc.
• Or a bit of both!

5
Where do Observational Methods Fit?

• First step what do you want to know?
• To map the social activity space of the library
• To watch patrons in the moment as they engage
  in tasks
• Focus here typically on what is actually
  happening in a particular space
• To document what/where/how research questions
  related to patron/staff activities
• Context for quantitative data (e.g.,
  questionnaires)
• Context for qualitative (why) questions (e.g.,
  interview)
• Original data for GIS mapping or other
  visualization techniques
• To document taboo or other activities (e.g.,
  eating defacing books) that are difficult to
  gather

6
Observation Different Approaches

• Overt Methods
• Individuals are aware that you are gathering data
• Signs posted at door alerting individuals to
  study under way
• Individuals asked to engage in an activity (e.g.,
  computer search) which will be videotaped
• Researcher may be engaged in activity under study
• Covert Methods
• Individuals are not aware that you are gathering
  data
• Researcher as disengaged observer who documents
  and reflects on activities going on in the
  library
• Researcher as engaged observer, but participants
  unaware they are being studied

7
Data Collection Tools of the Trade

• Observation Checklists Journals (paper vs.
  digital)
• Created in advance to guide observation
• Created during field notes of what is
  happening in the space
• Digital video/photographs (is audio required?)
• Of spaces, alone or of patrons
• Of computer screens, storytime groups, etc.
• Maps (paper vs. digital)
• Hand-drawn vs. blueprints
• Analysis software
• Quantitative Excel, SPSS, etc.
• Qualitative (including images) Atlas.ti
• GIS visualization ArcGIS

8
Toronto Reference Library
9
Toronto Reference Library
5th Floor
10
TRL Information Desk Foyer
11
Vancouver Public Library
12
Vancouver Public Library
13
VPL Indoor Street Library
14
Seating Sweeps Method

• Observational walks (or sweeps)
• 3 times per day over a one week period
  (1015-1130am 2-330pm 6-730pm)
• Covered all library spaces separate floors,
  café area, circulation desk, etc.
• Systematic detailed observations of 60
  different variables
• Who is present in specific locations? What are
  they doing? What possessions do they have with
  them?

15
Seating Sweeps contd

• First step scour library space to develop
  locational acronyms
• CW computer workstation
• BT book truck
• Second count seats in/outside library space
• Benches, sofas, carrels, worktables, etc.
• Photos/maps of spaces (with/without patrons)
• Third develop coding worksheet
• Academic library settings used PDA for this
  task, so also needed to pilot test equipment
• Benefit much less obtrusive and easier to prep
  for analysis
• Downside more time-consuming, up-front

16
Sweeps Method Issues

• Daily debriefing re codes
• E.g. searching vs. physical searching
• Differences between sites
• E.g. VPL cellphone-friendly zone
• Unobtrusive?
• Us watching them vs. them watching us
• Individuals in motion
• People leave, pick up new books, start to talk to
  the person next to them, etc

17
Individuals Activities in Library
18
How Might we Extend Sweeps?

• Use furniture counts (quantitative) to determine
• How much of the library seating space is actually
  being used?
• How are patrons using that space sleeping?
  Staying all day? Do these trends vary by location
  in library?
• Observational shadowing following patrons
  throughout their visits (qualitative)
• Mapping data with GIS to compare patterns
  across floors, across time periods, etc.
• Active use of photos to paint a full picture
  of the spaces and activities under study

19
Bringing Data to Life

• Alex 2nd year, Business/Commerce
• When Im really trying to study, I like the
  reading room in Rutherford, that big room with
  paintings on the wall, at the top. I think for me
  its almost like an historical sense and I almost
  have like a duty to study in there, you know?
  Its just kind of that atmosphere in there so
  thats where I go to really studyI think its
  pretty open and its quiet in there, theres
  space for you to go at the tables, spread all
  your work out, youre not like at a little,
  little desk thing

20
A Photo tells the Story
21
Why Use Observational Methods?

• A snapshot
• Who are our patrons? What are they doing in the
  library?
• A chance to see what patrons really do
• How do staff interact with patrons (or one
  another) in the library space?
• This information is invaluable for
• Allocating staff within the library
• Determining financial support
• Organizing furniture (re)designing library space

22
Sampling Access

• Always start by thinking about the population you
  wish to study then make decisions about sample
  selection
• Will you study the population or a sample?
• How will you find participants/documents and
  can you gain access?
• Target population the population to which you
  would like to generalize or transfer results
• All librarians or only those in academic
  settings
• All teenagers or only those with diabetes

23
Sampling Frame

• The set of all cases from which the sample is
  drawn
• 2 ways to construct this frame
• (1) listing all cases
• e.g. directory of academic librarians
• (2) providing a rule to define membership
• e.g. teenagers diagnosed with diabetes during
  early teen (13-15) years
• How many participants?

24
Non-probability Sampling

• Convenience Sampling
• Individuals chosen who are available or easy to
  find
• Give questionnaire as people enter the library
• Purposive Sampling
• Each sample case is selected for a purpose
  because of its unique position
• A typical rural library a typical urban
  library
• Quotas set so sample represents certain
  characteristics in population
• LIS programs 80 female, 20 male
• Snowball Sampling
• Identify one member of the population and get
  that individual to identify others

25
Probability Sampling

• Simple Random Sampling
• Each case in the population has an equal chance
  of being included in the sample
• Systematic Random Sampling
• First case is randomly selected, then every nth
  case
• Stratified Random Sampling
• Population is first subdivided into 2 or more
  strata based on mutually exclusive categories
  of one or more relevant variables (e.g.,
  male/female)
• Simple random samples are then drawn from each
  stratum, and these sub-samples are joined to form
  the complete stratified sample

26
Probability Sampling (2)

• Cluster sampling
• The population is broken down into groups of
  cases (called clusters), and a sample of
  clusters is selected at random
• The clusters generally consist of natural
  groupings (e.g. geographic - city blocks
  organizational library branches in one city)

27
Quality Research - Criteria

• Quantitative
• Validity
• Generalizability
• Reliability
• Objectivity

• Qualitative
• Credibility
• Transferability
• Dependability
• Confirmability

28
Quantitative - Validity

• The degree to which you are truly measuring what
  you intend to measure
• E.g. Hawthorne Effect (Study of productivity at
  an Electric company)
• Looking for evidence that a particular
  independent variable (e.g. bright lights) has
  caused a change in an observed dependent variable
  (e.g. productivity)
• Use control groups triangulation of methods
  focus on operational definitions

29
Qualitative - Credibility

• Does the description developed through the
  inquiry ring true for members of the group
  under study?
• Assessed through - prolonged engagement,
  persistent observation, triangulation, assessment
  of contextual materials, peer debriefing, member
  checks

30
Quantitative Generalizability

• The ability to generalize findings across
  different settings or populations
• Asks whether the results of the study apply to
  cases not included in the study
• E.g. study of MLIS students use of OPACs not
  generalizable to the general population, as
  library students have special knowledge
• Select sample carefully repeat measure
  (different setting/time of day/etc.)

31
Qualitative Transferability

• Can the findings be applied in other contexts or
  with other respondents?
• Key to focus on those elements which will not
  shift with context/time (and recognize
  limitations of context/time to your findings)
• E.g. will Canadian studys findings transfer to
  U.S. context?

32
Quantitative - Reliability

• Will the research yield stable, consistent
  results when applied repeatedly? Can the study
  be replicated?
• E.g. How many books have you borrowed this
  year? may be unreliable they may not recall
  the exact (so they guess) they may feel that
  they should have borrowed more books (so they
  inflate )
• Use pilot studies interviewer training apply
  measure consistently etc.

33
Qualitative - Dependability

• If the study were replicated with the same (or
  similar) respondents in the same (or similar)
  context, would the findings be repeated?
• Key not looking for invariance but for
  trackable variance - that if there are
  inconsistent findings, these are attributable to
  particular sources (e.g. reality shifts, better
  insights)

34
Quantitative - Objectivity

• Quantitative measures are value-free and
  therefore objective
• Key here idea that subjectivity leads to
  results that are both unreliable and invalid
• Eliminate sources of bias and error that may
  distort results
• E.g. use several, independent observers withhold
  information from participants to not taint
  opinions tape record events etc.

35
Qualitative - Confirmability

• Researcher as Instrument
• The degree to which the findings are the product
  of the focus of the inquiry and not of the biases
  of the researcher (i.e. key neutrality)
• Not to ensure that data is free from
  contamination (as objectivity is an illusion),
  but to show that the data (assertions, facts,
  etc.) can be tracked to their sources
• E.g. inter-coder checks on interview data