Textbook Notes (270,000)
CA (160,000)
UTSG (10,000)
CSC (50)
Chapter 8

# CSC318H1 Chapter Notes - Chapter 8: Web Analytics, Jargon, Preempt

Department
Computer Science
Course Code
CSC318H1
Professor
Olivier St- Cyr
Chapter
8

This preview shows half of the first page. to view the full 3 pages of the document.
CH 7 – DATA GATHERING
Five Key Issues of Data Gathering
1. Setting Goals
oNeed to ID specic goals for study before gathering data
Goals inuence nature of data gathering sessions,
techniques used, anaylsis to be performed
oGoals expression can be informal/formal
Interaction Design, usually express data gathering
goals informally
2. Identifying Participants
oGoals indiciate the specic people to gather data from
(population)
Sampling – Population too large, need to choose
subset
Techniques: Simple Random Sampling, Stratied
Sampling
Saturation Sampling – all members of population
accessible
Techniques: Convenience Sampling, Volunteer
Panels
oParticipant Selection Techniques
Porbabaility methods = statistical tests, generalize to
population
Random Sampling – use random number
generator, or choosing nth person in a list
Strati!ed Sampling – divide population into
groups (ex. by attributes) then apply random
sampling
Non-probability methods = cannot reliably generalize
to population
Conveniencee Sampling, Volunteer Panel
selection rely more on participant availability
3. Relationship with Participants
oRelation btwn participants and researcher = clear and
professional
oMay need Informed Consent from participant to ensure
they're aware
1. Purpose of data gathering
2. How data will be used
3. Can withdraw at anytime
4. Triangulation
oInvestigation of a phenomenon from (at least) 2
perspective, 4 types:
1. Triangulation of data  data drawn from di9erent
sources
at di9erent
times
, in di9erence
places
, from
di9erent
people
Possibly with di9erent
sampling techniques
2. Investigtor Triangulation  di9erent researchers
(observers, interviewers) collected/interpreted data
3. Triangulation of Theories  di9erent theoretical
frameworks to view the data/ndings
Frameworks must share similar philosophical
underpinnings
4. Methodological Trangulation  use di9. data
gathering techniq.
oLmitations: validation w/ triangulation hard, data may be
uncomparable
5. Pilot Study
oSmall trial run of the main study
Make sure methods is viable before embarking on real
study
oEnsures:
Interview questions, questionaires checked for clarity
Experimental procedures conrmed as viable
Potential problems identied in advance, corrected
oCan involve colleagues/peers, not allowed for main study
Data Recording Approches
Note & Still Camera
oAdvantage: handwritten notes (pen paper) less intrusive
than keyboard
More exible, can be transcribed later
oDisadvantage: di?cult/tiring to write, hard to decipher
Easy to lose concentration while observing and writing
oNotes supplimeneted w/ photos of artifacts, events,
environment
Audio & Still Camera
oAdvantage: Allows observer to focus on ativity, rather than
spoken word
Less intrusive than video
oDisadvantage: transcribing audio to text = time-consuming
Not necessary to transcribe everything b/c use as
reminders
oCan also be supplimeneted w/ photos
Video
oAdvantage: capture both visual and audio
Positioning camera requires e9ort, varies per activity
Camera focus need to be considered
Need eldwork to familiarized w/ environment
Video recording has impact on particpants
Interviews
Unstructured Interviews – open-ended, do not follow scripts
oExploratory in nature, more like conversations around
particular topic
Open questions  no expectation on format/content
Not replicable btwn participants
oMay generate rich data
Structure Interviews – scripted, predetermined questions
oQuestions are the same btwn participants (standardized)
Questions usually short, clearly worded
Closed questions  require answer from
predetermined set of alternatives (yes/no)
Need to know range of possible answers
oUsed when goals clearly understood, and speciec
questions identiable
Ex. which of the following website do you visit most
often…
Semi-structured Interviews – basic script, ensure same
topics are covered
oHas Open and Closed questions
oAnswers can be explored more in depth
Probe interviewee until no new relevant information is
forthcoming
oPitfulls to avoid:
Ex. "You seemed to like this use of color…"
Body language of interviewer have strong inuence
on whether interviewee will agree w/ a question
Give interviewee time to speek, don't move on too
quickly
Focus Groups – interview people in groups (3 – 10)
Facilitators need to encourage queit people, stop
verbose ones
oSelected participants is representative sample of target
population
oPreset agenda guide discussion
But exibility for facilitator to explore unanticipated
issues
oFocus group allows divers/sensitive isues raised that might
be missed
Method allow individuals develop opninion w/i social
context by talking to others
Used mostly for marketing, poltical campaigning, social
science
Planning and Conducting an Interview
oDeveloping Interview Questions, guidelines
Compound sentences confusing  split it up