ATLAS.ti:
The Qualitative Data Analysis Workbench
An overview
Ricardo B. Contreras, PhD
Applied cultural anthropologist
Director of the ATLAS.ti Training Center
Greenville, North Carolina, United States
trainingcenter@atlasti.com
Kristin Kostick, PhD
Applied medical anthropologist
Ethnopro
Certified ATLAS.ti Instructor
www.atlasti.com
The Company
• Based in Berlin, Germany.
• First commercial release in 1993.
• Currently in release 7 (June 2012).
© 2012 Ricardo B. Contreras, PhD
www.atlasti.com
Learning Resources
• To learn ATLAS.ti on your own (http://www.atlasti.com/download.html):
– Quick tour
– Chapters of the manual (at the very least):
• Data management
• Creating, opening, and saving HU
• Setting up a new project
• Project back-up and transfer
• To learn ATLAS.ti under the guidance of an experienced instructor
(www.atlasti.com/training.html):
– Online 6-hour workshops through web-conferencing
– Face-to-face 2-day workshops
– On-site 2-day workshops (on-demand)
• Additional guidance
•
•
•
Book by Susanne Friese “Qualitative Analysis with ATLAS.ti”. London: Sage,
2012.
Companion website of the book: http://www.quarc.de/qualitative-data-analysiswith-atlasti/companion-website.html
Multiple videos in Youtube.
© 2012 Ricardo B. Contreras, PhD
www.atlasti.com
What is ATLAS.ti?
• Software that assists in the process of qualitatively
analyzing research data.
• Data collected through unstructured or semi-structured
methods of data collection.
• The researcher is in control of the analysis process.
© 2012 Ricardo B. Contreras, PhD
www.atlasti.com
Ultimate Purpose of Using ATLAS.ti
in Analysis
Closing the gap between the original text and the interpretative text
Original Text,
Participants’
Voices
Final Text,
Interpretative
Text
Note:
This figure was inspired by the following book: Wolcott, Harry F. 1994. “Transforming
Qualitative Data: Description, Analysis, and Interpretation”. Thousand Oaks: Sage.
© 2012 Ricardo B. Contreras, PhD
www.atlasti.com
The Deductive-Inductive Dialogue in
ATLAS.ti
Deductive
Inductive
Generalizations
External Framework
Data
You are not forced into a deductive or an
inductive approach. The decision as to
what approach to take is a methodological
one. However, commonly people combine
both: start with a set of a priori concepts
(codes), as a general framework, but leave
enough space for discovery in the form of
emergent concepts (codes).
© 2012 Ricardo B. Contreras, PhD
Data
Inductive
Deductive
www.atlasti.com
Hermeneutic Unit
• Integrated context for data description, analysis, and
interpretation.
• Container that holds the sources of information and all of
the analytical work done around them.
• File of extenstion “hpr6/7“.
• Every research project requires a single hermeneutic unit
holding all of the sources of information to be analyzed.
© 2012 Ricardo B. Contreras, PhD
www.atlasti.com
The Objects of the Hermeneutic Unit
PDs
Quotes
Networks
These are the basic
elements of an analysis
project with ATLAS.ti.
Generally speaking, the
many other functions and
tools available to
researcher revolve around
these six objects.
© 2012 Ricardo B. Contreras, PhD
HU
Codes
Families
Memos
www.atlasti.com
Primary Documents:
Sources of Information
Text
The multiple file
formats accepted
allows for rich
triangulation.
Graphic
Multimedia
Word
JPEG
MPEG-2,
MPG
RTF
PNG
WMV
PDF
GIF
AVI
Excel
Etc.
MP4
Google
Earth
PDs
TXT
WAV
Quotes
Networks
HU
Etc.
Etc.
Codes
Families
Memos
© 2012 Ricardo B. Contreras, PhD
www.atlasti.com
Primary Documents:
Loading Multiple Primary Documents Simultaneously
•
•
•
Loading up to four documents simultaneously.
Allows to conduct rich comparative analysis (e.g., coding two or more
interviews conducted at different points in time).
Allows for rich transcription of audio or video documents.
PDs
Quotes
Networks
HU
Codes
Families
Memos
© 2012 Ricardo B. Contreras, PhD
www.atlasti.com
Primary Documents:
Multimedia and Google Earth
Video frame showing
quotations and codes
on the margin
Audio primary document
showing quotations and
codes on the margin
Graphic primary
document showing
quotations and codes
on the margin
PDs
Quotes
Networks
Google Earth primary
document showing
quotation named “My
house”
HU
Codes
Families
Memos
© 2012 Ricardo B. Contreras, PhD
www.atlasti.com
Quotations
•
•
•
Segments of the text selected by researcher or automatically (auto-coding).
Quotations can be created in all kinds of primary documents.
The quotation constitutes the basic unit of the analysis project. They can be
linked to other objects, they can be left unlinked, they can be described indepth, and their linkages can be visualized as rich graphic representations.
Quotation
always
shown in
context
Codes
linked to
quotation
shown on
margin
Interactive list of
quotations shown in the
“Quotation Manager”,
each one identified by
unique ID
PDs
Quotes
Networks
“Comment” space
where the quotation
can be described.
HU
Codes
Families
Memos
© 2012 Ricardo B. Contreras, PhD
www.atlasti.com
Codes
•
•
Concepts that can either derive from external frameworks (deductively
created) or emerge from the text (inductively created).
Codes can be linked to quotations, memos, arranged in families, and
represented in networks. They can also be left alone, unlinked.
Codes shown in the “Code
Manager” specifying how
grounded in the evidence they
are, how connected they are to
other codes (density), their
authors, and their definitions.
PDs
Quotes
Networks
Codes
shown on
an optional
side panel
© 2012 Ricardo B. Contreras, PhD
Codes that
are linked to
quotations
shown on
the margin
HU
Codes
Families
Memos
www.atlasti.com
Memos
•
•
•
•
•
Spaces for reflection, analysis, integration, and interpretation.
Spaces where to make sense of the data.
Writing memos should accompany the process of coding.
Memos can be linked to quotations, codes, and other memos.
If memos are systematically linked to supporting quotations, it follows that
the reflections, analyses, and interpretations contained in them are grounded
in evidence.
Memo that is
linked to quotation
is shown on the
margin
Memos shown in “Memo
Manager”. This shows how
grounded the memo is in
the evidence (number of
quotations linked to it)
PDs
Quotes
Networks
HU
The memo text
Codes
Families
Memos
© 2012 Ricardo B. Contreras, PhD
www.atlasti.com
Families
•
•
Grouping of primary documents, codes, and memos according to shared
characteristics.
Families allow for comparison across cases, conceptual categories, themes,
etc.
Network view of
primary
document family
Network view of code family
PDs
Quotes
Networks
HU
Network view of memo
family
Families
Codes
Memos
© 2012 Ricardo B. Contreras, PhD
www.atlasti.com
Networks
Weak link network showing quotations
linked to codes as well as the primary
document from which some of the
quotations originate
Weak-link Network
Linkages between elements
created through the normal
analytical process.
Code-to-code network
showing semantic
relations between
codes
Quotation-to-quotation network
(hyperlinks) showing semantic
relations between quotations
PDs
Strong-link Network
Graphical representations of a
semantic type linking codes to codes
and quotations to quotations through
specific meanings. Researcher’s
understanding of the problem.
© 2012 Ricardo B. Contreras, PhD
Networks
Quotes
HU
Codes
Families
Memos
www.atlasti.com
Outputs
Qualitative
The Query Tool, which produces
qualitative outputs through Boolean,
Semantic, and Proximity operators
Rich Text Format written
report produced through
the Query Tool
© 2012 Ricardo B. Contreras, PhD
Quantitative
The co-occurrence
table showing number
of co-occurrences
between codes
Excel graph with data from the Cooccurrence table. All quantitative
reports can be exported into Excel
www.atlasti.com
Exporting in Different Formats
XML, HTML, SPSS, EXCEL, RTF, PDF, PNG, etc.
© 2012 Ricardo B. Contreras, PhD
www.atlasti.com
Team Work
Mary
Mary + Peter
Peter
Final
HU
Base
HU
Susan
Susan + Paul
Paul
© 2012 Ricardo B. Contreras, PhD
www.atlasti.com
Project Management:
The Copy Bundle Function
• If working with documents that have been “added” (version 7) or
“assigned” (version 6.2) into the project, it is necessary to create
back-up files using the Copy Bundle function.
• Save the Copy Bundle file in an external drive.
• Use the Copy Bundle function to back-up and to migrate the project
between computers.
A. Create Bundle
© 2012 Ricardo B. Contreras, PhD
B. Save Bundle
www.atlasti.com
Document Management in Version 6.2
•
•
•
Create a new folder in a location of your preference in Windows
(e.g., desktop, My Documents, shared drive, server). Name it with
the name of your project.
Copy all your source document files into this folder.
Save the HU (extension “hpr6”) inside of the same folder.
Project A
HU
.doc .docx .pdf
.rtf .jpg .wmv, etc.
The HU file (extension ‘hpr’) saved
inside of the project folder together with
the source documents.
© 2012 Ricardo B. Contreras, PhD
Project B
.doc .docx .pdf
.rtf .jpg .wmv,
etc.
HU
.doc .docx .pdf
.rtf .jpg .wmv,
etc.
The HU file (extension ‘hpr’) saved inside of the project
folder but not inside of the individual sub-folders
where the source documents are stored.
www.atlasti.com
Thank you!
ATLAS.ti Scientific Software Development GmbH
Hardenbergstr. 7
D-10623 Berlin
Germany
trainingcenter@atlasti.com
© 2012 Ricardo B. Contreras, PhD
www.atlasti.com