This document discusses qualitative research methods for business. It addresses challenges in making qualitative data understandable for real decisions. It discusses why businesses conduct research, how to determine what data and analysis is needed, and issues with determining "truth" in business contexts. Finally, it discusses four types of qualitative data and focuses on making qualitative data more quantitative by addressing issues like validity, sample sizes, and coding consistency.
Messy Research: How to Make Qualitative Data Quantifiable and Make Messy Data Understandable
1. 1
MESSY RESEARCH
How to Make Qualitative Data
Quantifiable and Make Messy Data
Understandable
Dr. Gigi Johnson
2. Core • When to chose it
Qualitative • Major challenges in design
Issues and analysis
• How to tell messy stories for
real decisions and action by
business
3. 3
Why • Make a decision
Business • Deploy resources as an
Research organization
? • Convince others in
organization
• Rule out alternatives
• Influence certain people in
the company
• Understand change ahead of
company
4. 4
How Much Do We Need?
• How much data and what data?
• From where?
• What analysis do we need to do?
• What narrative/presentation is enough?
Often, in business research, we tend
to focus on volumes . . .
. . . missing focus on analysis and
presentation for decision-making
5. 5
What is Truth in Business?
• What is enough information to make a decision—or real
“Truth”?
• How much information of what type is “enough”?
• How messy will this be and still be “normal”?
• Who can best find the data?
• Which people know what part of the answers we need?
• Is the core data we need reachable?
• What is our role as researcher with perspective?
6. 6
4 Types of Qualitative Data
Leech, N. L., & Onwuegbuzie, A. J. , 2005
7. 7
Focus on Quantitative
Issues Design characteristic
• Validity • Sample sizes
• External Validity: • Statistical significance
generalize findings across • Sampling bias
populations, tasks, and • Coding consistency
environments (Campbell
• Control groups
& Stanley, 1966)
• Pre- and post-testing
• Internal Validity: Design
rule out other factors • Instrument design (tested
other than the validity)
Independent Variable
10. 10
Qualitative vs. Quantitative
Quantitative Qualitative
•Helpful when “answering •Looking at single case or
questions of who, where, small number of cases
how many, how much, and •Looking at in-context
what is the relationship situation, framed by words
between specific variables” and narratives
(Adler, 1996, p. 5) •Looking for in-context
•Striving for causation or for relationships and
generalizing to larger connections
populations •Creating hypotheses or
instruments for quantitative
11. 11
Qualitative Can Enrich Quantitative
Examples:
“Prebriefing” (Collins et al., 2006), checking
potential quantitative survey participants for
willingness and suitability
Pilot study to assess the appropriateness of an
instrument like a questionnaire or survey
Ruling out hypotheses
12. 12
Challenge of Qualitative
• Difficulty in capturing lived experiences via
text
• Creating a “bricolage” – an assemblage of
representations that fit a complex situation
(Denzin & Lincoln, 2005)
Use of Qualitative Analytical tools helps connect
this complex in-context environment into a way
that others can understand.
15. 15
Populations and “Level”
• Populations: Total target group
• AMR: Group could be a regional or business population, or could
be members at a level in the organization
• Sample: Group in study
16. 16
“Who” has the Data?
• Thinking in terms of Five Forces
• Vendors, Customers, Competitors
• Reasons to share
• “Knowing”
• Belief, research, or connections
• Expert does not mean “knows” real information
• Similar question: Secondary Research and connecting
Primary to it
17. 17
Research Methods
• Document Analysis
• Focus Groups
• Observations
• Interviews
• Shadowing
• Participant Observation
• Literature Review
• Oral History/Ethnography
• Social Network Analysis (SNA)
• Quantifying/mapping context
18. 18
Literature Review
• Check out what research has been done on the research
methods that you are considering, e.g., focus groups,
narrative research, document analysis
• Google Scholar: Good launching pad
19. 19
Sampling Methods and Size
• Quantitative: Concern with probabilities and similarities to
overall population
• Qualitative?
• Snowball sampling: uses social networks and
connections to identify unknown populations
• Convenience sampling
• Judgment Sample: based on framework of variables
from researchers
• Maximum Sampling, Extreme
• How much is enough?
• Saturation (repeated patterns) (Rubin & Rubin,
1995).
20. 20
Instruments
• Creating a Questionnaire
• Focus Group – Outline, Objectives
• Surveys – may be instruments already tested for validity
• Interviews
• Open Ended
• Semi-Structured
• Test and plan coding methods upfront; what will you input the answers into?
• Grounded Theory: Grand Tour Question(s)
23. 23
Transcription as Friend and Foe
01: Exactly. And, as far as doing it, the other, I think the biggest
obstacle, is training. Is getting=
G Is opportunity.
01 It is an opportunity. But . . .
Group ((chuckles))
01 . . . it is an obstacle as far as the [district is concerned.]=
G =[It is hard to not say it.]=
01 Because they will not give that time to really teach and train.
Even, you know, I'm gonna walk in as the . . . the real Luddite.
And be able to walk out and feel like I can go out and use the
equipment. Not just say it.
11 Yeah.
24. 24
Undercurrents from Field Notes
Individual impressions
Notes before and after sessions
Bring your own biases, context, and observations to the
table
26. 26
Designing the Analysis
• Not just casually connecting
• Causality vs. Correlation
• Two analysis directions
• Old-fashioned and robust
• Excel worksheets or written on documents
• Hand coding and counting
• Alternatives
• Computer-assisted data qualitative data analysis software
(CAQDAS)
27. 27
Recursive Abstraction
• Fancy phrase for summarizing, then summarizing the
summaries
• Usual accidental business research method
• Helps to have consistent methods for summarizing
between coders/team members, or a coding worksheet
28. 28
Coding
• Chunking text data, then adding a code
• You can code and iteratively recode/emergent (Tesch, 1990).
• Method: aimed to continue to narratively code while
bridging to new ideas and surfacing new categories until
you began to find pattern codes and themes (Miles &
Huberman, 1994).
29. 29
Key Phrase Frequency
• Word counts are based on the belief that all people have
distinctive vocabulary and word usage patterns.
• “Linguistic fingerprints” (Pennebaker, Mehl, &
Niederhoffer, 2003, p. 568).
• Gives context to words like “many,” “frequently,” etc. terms
are fundamentally quantitative.
30. 30
KWIC
Keywords-in-context (KWIC; Fielding & Lee, 1998)
•Data analysis method that reveals how respondents use
words in context
•Compares words that appear before and after “key words”
31. 31
Narrative Analysis (NA)
• (Nearly) all qualitative research is filtered by contexts,
beliefs, and methods of communication
• NA evaluates patterns, threads, tensions, and themes
within the transcripts and field notes (Clandinin &
Connelly, 1994, 2000; Ryan & Bernard, 2000).
• Can pull out portions of text where themes are mentioned
(Ryan & Bernard, 2000)
32. 32
Triangulation
• Assesses the integrity of the inferences that one draws
from more than one vantage point (Lincoln & Guba, 1985)
• Use of multiple data sources, multiple researchers,
perspectives, tools, and/or methods (Denzin, 1989;
Schwandt, 2001)
• Adds confirmability, dependability, and credibility to data
collection
34. 34
2 Reasons for Tools
• Help the team gather, sort, visualize, and engage messy
and abundant qualitative data
• Explain and convince client of validity of research done
• Ability to walk through analytical process and explain
the patterns in the data
40. 40
Using Data to Tell and Be the Story
• Abundant Data (“Big Data”) from in-context data collection
in our connected world
• Social Network Analysis (SNA) – how we are all
connected
• Big company problem – large volumes of data to digest
and act upon
• “Investigating relationships” – not just for presentation, but for
research teams to visualize emerging patterns
42. 42
• Great list:
Data http://dailytekk.com/2012/02/27/over-100-incredible-infographic-tools-a
Visualizatio • Piktochart – Transforms your information into memorable
presentations.
• Infogr.am - Create interactive charts and infographics.
n
Infographics for
• Gephi – Like Photoshop for data. Graph visualization and
manipulation software.
Decisions • Tableau Public - Free data visualization software.
• Free Vector Infographic Kit – Vector infographic elements from
(others in MediaLoot.
Appendix slides) • Weave – Web-based analysis and visualization environment.
• iCharts – Charts made easy.
• ChartsBin – A web-based data visualization tool.
• GeoCommons – See your data on a map.
• VIDI – A suite of powerful Drupal visualization modules.
• Prefuse – Information visualization software.
• StatSilk – Desktop and online software for mapping and
visualization.
• Gliffy – Online diagram and flowchart software.
• Hohli – Online charts builder.
• Many Eyes – Lets you upload data and create visualizations.
• Google Chart Tools – Display live data on your site.
43. 43
Questions
?
Dr. Gigi Johnson
@maremel
Maremel Institute
44. 44
Playing • Wordle – http://www.wordle.com
with – fun tool to turn words from
Words documents into word maps
• Tagxedo --
http://www.tagxedo.com – similar
to Wordle, Tagxedo lets you
create word clouds and
sculptures from URLs, Tweets,
and other social media
documents, as well as export
them into a variety of formats.
45. 45
We can tinker with maps, both as pre-
Playing made images as well as data-driven
with Maps tools.
•Web Resources Depot --
http://www.webresourcesdepot.com/free-vector-w
-- shares a variety of world map images for
use
•Free PSD Files --
http://freepsdfiles.net/3d-renders/3-isolated-3d-us
-- This site has easy images for further
editing for presentations
•GunnMap -- http://www.gunn.co.nz/map --
creates world maps with your data
•StatPlanet Map Maker --
http://www.statsilk.com/software/statplanet --
also creates interactive maps
46. 46
• Several tools let you expand how you lay out
Playing concept maps and linked ideas:
• FreeMind -- http://freemind.sourceforge.net – I enjoy
with this free tool. Graphically simple, it lets you play with
a free tool for mind mapping that can be adapted
Concept into all sorts of other applications.
• Webspiration – http://www.webspirationpro.com – I
Maps miss its freemium mode; it now has a trial period and
then costs $6/month. I found Inspiration and
Webspiration wonderful for group presentations and
immediate work.
• MindManager -- http://www.mindjet.com – this
concept management tool starts at $20/month for
one and discounts for group collaboration.
• MindNode -- http://www.mindnode.com – This tools
for Mac computers comes at a moderate price -- $20
for the mac and $10 as an iOS Apps.
• VUE by Tufts -- http://vue.tufts.edu -- I really enjoy
this “Visual Understanding Environment” tool, which
combines concept maps with search and graphics.
•
47. 47
• Prezi -- http://www.prezi.com -- My recent
Presentations undergraduate class spent half of their
projects in Prezi, which has a zooming
camera metaphor across a vast digital
white board. They enjoyed putting in
music, video, and other embedded
content. I got a bit dizzy, but enjoyed the
creativity.
• Sliderocket -- http://www.sliderocket.com --
Several of my students enjoyed using
Sliderocket for class presentations. It gave
them a robust and elegant toolset to work
with.
• Brainshark -- http://www.brainshark.com --
Friends who are professional business
development executives heartily
recommend Brainshark as a way to pre-
package and present content at a
distance.
48. 48
• Google Charts API - http://code.google.com/apis/chart/ -- you can use
Graphs Google Charts to create animations in charts, dashboards, and lots of
other goodies
• Gliffy -- http://www.gliffy.com/ -- I just found Gliffy, a great diverse creator
and of charts and graphs. Different versions of it work with different social
workspace/sharing software:
• Hohli -- http://charts.hohli.com – free online chart builder
Charts • Creately -- http://creately.com -- (paid but cheap at $5/month/person) is
a online tool to build charts, and collaborate around them
• Many Eyes -- http://www-958.ibm.com/software/data/cognos/manyeyes/
-- an experiment by IBM Research and the IBM Cognos software group
let users create and evaluate data visualizations.
• GGobi -- http://www.ggobi.org/ -- free data visualization tool for your
datasets
• Mondrian -- http://www.rosuda.org/Mondrian/ -- open source toolset for
charting and graphing data plots and more complex graphs and data-
driven visuals
• OpenDX -- http://www.opendx.org -- Older open source software, based
on IBM’s visualization data explorer.
• Spotfire -- https://silverspotfire.tibco.com – a whole visualization suite,
free for individuals for the first year, then $99/year thereafter.
• Visualizefree -- http://www.visualizefree.com/ -- Sampler of more complex
system; shows real-time images from the FAA of flights as a sample
• Mycrocosm -- http://mycro.media.mit.edu/ -- quirky tool to create displays
of your own personal data that you can input by cell or email and track
49. 49
• Hans Rosling’s Gapminder
Playing Foundation worked with Trendalyzer, which
then was sold to Google in 2007, then
with folded away when Google Labs.
Motion • VIDI -- http://www.dataviz.org/ -- VIDI Data,
run by the Jefferson Institute, provides a
Charts visualization module for Drupal CMS to
show motion charts, timelines, geodata,
and comparative data.
• TrendCompass
-- http://epicsyst.com/trendcompass -- lets
you add your own data to their data
visualization tool if you register
• Eurostat Explorer
-- http://www.ncomva.se/flash/explorer/eur
o/ -- sample with EU data that can be
played with using a motion graphic.
50. 50
Playing • Tweakersoft’s Vector Designer
-- http://www.tweakersoft.com/vect
with ordesigner.html -- This $20 Mac
Images App helps users create simple
vector designs.
• GIMP -- http://www.gimp.org --
For those who would want to tinker
with Photoshop, but wince at the
pricetag, GIMP (“GNU Image
Manipulation Program”) is an open
source alternative.
• Inkscape -- http://inkscape.org/ --
open source vector graphics
•
51. 51
Playing There are lots of extensive tools to
work with large public databases.
with Data •Google Public data
Resources -- http://www.google.com/publicdata
-- From the creators of abundant
and specialized search comes
search just for public data sources
•Visualizing.org
-- http://www.visualizing.org/data/br
owse and http://www.visualizing.org/
data/channels -- Visualizing
provides links to all sorts of sample
and interesting data sets
52. 52
• KDnuggets News newsletter on Data Mining
Additiona and Knowledge Discovery
l Tools -- http://www.kdnuggets.com/software/visualiza
tion.html -- longer list of free and paid data
visualization tools
53. 53
Select References
• Leech, N. L., & Onwuegbuzie, A. J. (2007). An array of qualitative data
analysis tools: A call for qualitative data analysis triangulation. School
Psychology Quarterly, 22, 557-584.
• Lewins, A., & Silver, C. (2007). Using Software in Qualitative Research: A
Step-by-Step Guide, Sage.
• Lewins, A. (2008). CAQDAS: Computer Assisted Qualitative Data
Analysis' in (ed) N. Gilbert, Researching Social Life (ed.)(3rd ed). London:
Sage.
• Lewis, R.B., & Maas, S.M. (2007). QDA Miner 2.0: Mixed-Model
Qualitative Data Analysis Software, Field Methods 19: 87-108
• Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Beverly Hills, CA:
Sage.
• Silver, C., & Fielding, N. (2008). Using computer packages in qualitative
research. In C. Willig & W. Stainton-Rogers (eds.), The Sage Handbook
of Qualitative Research in Psychology. London: Sage.
Editor's Notes
When should we use qualitative research versus other methods? What are the major challenges in design and analysis ? What should we consider for an AMR Project with qualitative research versus what we might do for a published research project or for our own professional work?