Vous êtes sur la page 1sur 7

REPORT GUIDE

THE BLIND MEN AND THE ELEPHANT: AN INDIAN FABLE

 Six blind men go to observe an elephant. One feels the side and thinks the elephant is like a
wall. One feels the tusk and thinks the elephant is like a spear. One touches the squirming
trunk and thinks the elephant is like a snake. One feels the knee and thinks the elephant is
like a tree. One touches the ear, and thinks the elephant is like a fan. One grasps the tail
and thinks it is like a rope. They argue long and loud and though each was partly in the
right, all were in the wrong.
 Our data analysis will determine whether our research findings are partly right, or
entirely wrong.
-Are research findings reflective of what is true in the real world?
-NO. because ALL RESEARCH HAVE LIMITATIONS.
1. LIMITATIONS SET BY THE RESEARCH QUESTION
-Our scientific investigation is only within the bounds of our research questions or
statements of the problem, because this constitutes VALIDITY. That is
measuring what we purport to measure, investigating what we ought to be
investigating.
2. LIMITATIONS SET BY THE RESEARCH VARIABLES
-The number and scale of measurement of the variables will limit how the data
will be
analyzed.
-For example, when you have three or more variables, ANOVA will be the
statistical treatment used for data analysis. When you have NOMINAL OR
CATEGORICAL variables, you use non-parametric tests (chi-square) rather than
parametric tests (t tests).
3. LIMITATIONS SET BY THE RESEARCH DESIGN
-Quantitative research designs need statistical treatment for data analysis. As
mentioned above, the number and scale of the variables will determine the what
statistical treatment should be used. However, the type of quantitative design
should also be considered. For example, the statistical treatment for a
correlational research will be different from an experimental research or a survey
research.
-Qualitative research designs will either use categorization and theme based
analysis or quantitative analysis of text-based data, depending on which
qualitative framework is being used (Grounded Theory, Discourse Analysis,
Phenomenology)
4. LIMITATIONS SET BY THE DATA COLLECTION
-Sampling problems
-Interview biases
-It is the analysis of data that, in a decisive way, forms the outcomes of the research.

LEARNING OUTCOMES
1. Discuss the purpose and features of analyzing research data
2. Distinguish between quantitative and qualitative data analysis
3. Correct myths on data analysis
4. Examine qualitative data analysis
PART I. OVERVIEW OF DATA ANALYSIS
A. WHY ANALYZE DATA
1. Describe and summarize the data.
-Data analysis makes sense of the data collected. It presents information in a more
comprehensive and logical manner, in the context of the research problem/s.
2. Identify relationships between variables.
-In correlational research, the statistical treatment used will determine whether there is a
significant relationship between variables, the strength of the relationship, and identify outliers.
3. Identify the differences between variables.
-Statistical treatment can determine whether there is a significant difference between the means
of the variables investigated, or if the difference in the mean is due to chance alone.
4. Forecast outcomes.
-In correlational research, regression analysis can be done to predict the linear relationship of
variables.
-In experimental research, outcomes are predicted by virtue of establishing causality between
variables.
5. Explain a phenomena.
-In phenomenological research, the phenomena is contextualized in the context of experience.
6. Validate or Develop a Theory.
-The Grounded Theory approach utilizes data analysis to develop a theory
-Quantitative research can replicate previous studies and establish the reliability and validity of
research findings.
7. Establish patterns in a phenomena or event.
-This is true of the majority of qualitative research wherein the primary goal is to go in depth in
explaining a phenomena and finding themes and patterns.

B. FEATURES OF DATA ANALYSIS


Irrespective whether the data is quantitative or qualitative, the analysis involve:
1. INFERENCE – the use of reasoning to reach a conclusion based on evidence
2. COMPARISON AS A CENTRAL PROCESS – identification of patterns or aspects that are
similar or different.
3. STRIVING TO AVOID ERRORS, FALSE CONCLUSIONS AND MISLEADING INFERENCES.

C. CORE DIFFERENCES BETWEEN QUALITATIVE AND QUANTITATIVE DATA ANALYSIS


1. Qualitative data analysis is less standardized with the wide variety in approaches to qualitative
research matched by the many approaches to data analysis. On the other hand, quantitative
data analysis requires choosing from a specialized, standard set of data analysis techniques.
2. The results of qualitative data analysis guide subsequent data collection, and analysis is thus a
less distinct final stage of the research process. Quantitative data analysis does not begin until
all data have been collected and condensed into numbers.
3. Qualitative researchers create new concepts and theory by blending together empirical and
abstract concepts. Quantitative researchers manipulate numbers in order to test a hypothesis
with variable constructs.
4. Qualitative data is in the form of words, which are relatively imprecise, diffuse and context-
based. Quantitative data use the language of statistical relationships in analysis.

D. Common Myths
1. Complex analysis and big words impress people.
-Most people appreciate practical and understandable analyses.
2. Analysis comes at the end after all the data are collected.
-We think about analysis upfront so that we HAVE the data we WANT to analyze.
3. Quantitative analysis is the most accurate type of data analysis
-Some think numbers are more accurate than words but it is the quality of the analysis process
that matters. The most accurate type of data analysis is one that provides reliable and valid
evidences to answer the research question/problem. When using a quantitative methodology,
you are normally testing theory through the testing of a hypothesis. In qualitative research, you
are either exploring the application of a theory or model in a different context or are hoping for a
theory or a model to emerge from the data. In other words, although you may have some ideas
about your topic, you are looking for ideas, concepts and attitudes often from experts or
practitioners in the field.
4. Data have their own meaning.
-Data must be interpreted. Numbers do not speak for themselves.
5. Stating limitations to the analysis weakens the evaluation.
-All analyses have weaknesses; it is more honest and responsible to acknowledge them.
6. Computer analysis is always easier and better.
-It depends upon the size of the data set and personal competencies. For small sets of
information, hand tabulation may be more efficient.
PART II. QUALITATIVE DATA ANALYSIS
A. WHAT IS QUALITATIVE DATA?
Qualitative data are forms of information gathered in a nonnumeric form. They are related to
concepts, opinions, values and behavior of people in a social context.
Types:
1. Structured text (writings, stories, survey comments, news articles, books, etc)
2. Unstructured text (transcription, interviews, conversation, etc.)
3. Audio recordings and music
4. Video recordings (graphics, art, pictures, visuals)

B. WHAT IS QUALITATIVE DATA ANALYSIS (QDA)


 The classification and interpretation of linguistic (visual) material to make statements about
implicit and explicit dimensions and structures of meaning-making in the material and what
is represented in it
 QDA is the range of processes and procedures whereby we move from the qualitative data
that have been collected into some form of explanation, understanding or interpretation of
the people and situations we are investigating. QDA is usually based on an interpretative
philosophy. The idea is to examine the meaningful and symbolic content of qualitative data.
For example, by analysing interview data the researcher may be attempting to identify any
or all of:
-Someone's interpretation of the world,
-Why they have that point of view,
-How they came to that view,
-What they have been doing,
-How they conveyed their view of their situation,
-How they identify or classify themselves and others in what they say
 The process of QDA usually involves:
• RECORDING OF DATA
This is done by audio recording on a digital voice recorder. It is important that
back up recording is done. For example, using another tape recorder to serve as
backup of electronic failure and faults; and to ensure that all voices could be
heard. Taking notes served as further back up and provided the context to the
interviews.
• VERBATIM TRANSCRIPTION OF RESPONSES
This should be done as soon as possible. The researcher has the option to
transcribe the responses himself or let an expert do the transcription. Notations
can be used.
 CODING INTO THEMES
Looking for themes involves coding. This is the identification of passages of text (or other
meaningful phenomena, such as parts of images) and applying labels to them that indicate
they are examples of some thematic idea. At its simplest, this labelling or coding process
enables researchers quickly to retrieve and collect together all the text and other data that
they have associated with some thematic idea so that they can be examined together and
different cases can be compared in that respect.

THIETART (2007:139) and NEUMAN (2011:510-514): STEPS IN CODING


1. OPEN CODING – involves the identification and naming of segments of meaning from
the field notes and transcripts in relation to the research topic.
2. AXIAL CODING – means reviewing and examining the initial codes that were identified
during open coding.
3. SELECTIVE CODING – involves selective scaning of all codes that were identified for
comparison, contrast and linkage to the research topic (question) as well as for a
central theme or “key linkage” that might occur.

C. APPROACHES IN QDA
1. DEDUCTIVE APPOACH
-Using your research questions to group the data and then look for similarities and differences.
-Used when time and resources are limited.
-Used when qualitative research is a smaller component of a larger quantitative study
2. INDUCTIVE APPROACH
-Used when qualitative research is a major design of the inquiry
-Using emergent framework to group the data and then look for relationships
D. PRINCIPLES IN QDA
1. People differ in their experience and understanding of reality (constructivist-many meanings).
2. A social phenomenon can’t be understood outside its own context (Context-bound).
3. Qualitative research can be used to describe phenomenon or generate theory grounded on
data.
4. Understanding human behaviour emerges slowly and non-linearly.
5. Exceptional cases may yield insights into a problem or new idea for further inquiry.
E. PROCESS OR STEPS OF QDA
The traditional QDA is labour intensive. After gathering data, the researcher will:
a. Transcribe the source material with a word processor.
b. Make multiple photocopies of the text,
c. Painstakingly read through and assign codes to the material
d. Cut the pages up into coded passages and then,
e. Manually sort the coded text in order to analyze the patterns they find.
**More efficient QDA Process

1. ORGANIZE THE DATA


 Transcribe the data*
 Translate the data*
 Data cleaning
 Label or code the data (structuring and familiarizing)
*Can be done manually or thru a computer program.
*Manual Methods:Notes and interviews are transcribed and transcripts and images etc. are
copied. The researcher then uses folders, filing cabinets, wallets etc. to gather together
materials that are examples of similar themes or analytic ideas. This facilitates easy retrieval
of such linked material, but necessitates two things: (1) Making multiple copies of the
original data as the same data may represent two or more themes or analytic ideas. (2) A
careful method of labelling the material in the folders or files so that it is possible to check
back and examine the broader context in which that data occurred. The analyst needs to
know where the snippets of data in the files came from so that they can be re-
contextualised.
* With the advent of the personal computer that proved excellent at manipulating text, it was
clear that with the right software much of the manual organisation could be done efficiently
with a PC. Thus many researchers have replaced physical files and cabinets with computer
based directories and files along with the use of word processors to write and annotate
texts. Many analysts now also use dedicated computer assisted qualitative data analysis
(CAQDAS) packages that not only make the coding and retrieval of text easy to do, but can
add other functions like searching that computers do quickly but which takes humans ages
to do or in some cases, which humans have never done. At first the focus of CAQDAS was
on text since that was easy to handle on PCs, but now that much audio and video is in
digital form too, software has been developed to support the analysis of audio and video
data.

2. IDENTIFY A FRAMEWORK
Framework will structure, label and define data.
a. Explanatory – guided by the research question/s
b. Exploratory – guided by the data
The framework will define the coding plan.

3. SORT DATA INTO FRAMEWORK


a. Code the data.
b. Modify the Framework.
c. Data entry if use computer packages

4. USE FRAMEWORK IN DESCRIPTIVE ANALYSIS


a. Arrange the responses in categories and
b. Identify recurrent themes.

Note: Stop here if research is exploratory.

5. SECOND ORDER ANLAYSIS


a. Identify recurrent themes
b. Notice patterns in the data
c. Identify clusters (Search for causality and identify related themes)
d. Build sequence of events
e. Search data to answer research questions
f. Develop hypothesis and test

F. QDA WITH SOFTWARE


With qualitative software, your workflow will be similar, but
each step will be made easier by the computer’s data storage
capability, automated searching and display. You can use text,
picture, video and audio source files directly. You can assign
codes manually (auto-code) to any section of the text, audio or
video or part of a picture. Analysis is easy with the report
feature, where you can select a subset of cases and codes to
work with, choose what data to use, and sort your reports
automatically

6.

What is Qualitative Data Analysis (QDA)?


Authors of this page: Ann Lewins², Celia Taylor¹ and Graham R. Gibbs¹
Affiliation: ¹University of Huddersfield and ²University of Surrey
Date written: 23rd Nov 2005
Updated 1st December 2010
Ref: Taylor, C and Gibbs, G R (2010) "What is Qualitative Data Analysis (QDA)?",
Online QDA Web Site, [onlineqda.hud.ac.uk/Intro_QDA/what_is_qda.php]

The process of QDA usually involves two things, writing and the identification of themes. Writing of some kind
is found in almost all forms of QDA. In contrast, some approaches, such as discourse analysis or conversation
analysis may not require the identification of themes (see the discussion later on this page). Nevertheless
finding themes is part of the overwhelming majority of QDA carried out today.

Writing
Writing involves writing about the data and what you find there. In many cases what you write may be analytic
ideas. In other cases it may be some form of précis or summary of the data, though this usually contains some
analytic ideas.

Coding into themes


Looking for themes involves coding. This is the identification of passages of text (or other meaningful
phenomena, such as parts of images) and applying labels to them that indicate they are examples of some
thematic idea. At its simplest, this labelling or coding process enables researchers quickly to retrieve and
collect together all the text and other data that they have associated with some thematic idea so that they can
be examined together and different cases can be compared in that respect.
Examples showing a short passage of coded text.
Interpreting
It is easy, when starting QDA both to write and code in ways that are nothing more than descriptive summaries
of what participants have said or done. Inevitably even description involves some level of interpretation though
the trick is to move away from the kinds of descriptions and interpretations that people would use in the milieu,
community or setting you are investigating to a categorisation and analytic understanding that begins to explain
why things are as you have found them.
Organising
The data sets used in QDA tend to be very large. Though samples may be quite small compared with those
used in quantitative approaches such as surveys, the kinds of meaningful data collected (field notes, video
recordings and interviews, for example) tend to be very lengthy and require the kind of intensive examination,
understanding and reading that only humans can do. In order to keep a clear mind and not become
overwhelmed by the sheer amount of data and analytic writings, the analyst needs to be organised.
Researchers tend to approach this organisation in one of two ways.
Manual methods
Notes and interviews are transcribed and transcripts and images etc. are copied. The researcher then uses
folders, filing cabinets, wallets etc. to gather together materials that are examples of similar themes or analytic
ideas. This facilitates easy retrieval of such linked material, but necessitates two things:
1. Making multiple copies of the original data as the same data may represent two or more themes or
analytic ideas.
2. A careful method of labelling the material in the folders or files so that it is possible to check back and
examine the broader context in which that data occurred. The analyst needs to know where the snippets
of data in the files came from so that they can be re-contextualised.
Computer based
With the advent of the personal computer that proved excellent at manipulating text, it was clear that with the
right software much of the manual organisation could be done efficiently with a PC. Thus many researchers
have replaced physical files and cabinets with computer based directories and files along with the use of word
processors to write and annotate texts. Many analysts now also use dedicated computer assisted qualitative
data analysis (CAQDAS) packages that not only make the coding and retrieval of text easy to do, but can add
other functions like searching that computers do quickly but which takes humans ages to do or in some cases,
which humans have never done. At first the focus of CAQDAS was on text since that was easy to handle on
PCs, but now that much audio and video is in digital form too, software has been developed to support the
analysis of audio and video data.
What does qualitative data analysis involve?
If you have NEVER performed any qualitative data analysis before, AND if you have NEVER attended any
methods training there are some aspects of QDA that it might be useful to consider. (If you are considering
using CAQDAS software then also take a look at the page on what the software can and cannot do.
 Are you interested in interpreting the data in terms of themes / concepts / ideas / interactions / processes?
See How and what to code.
o Then YOU have to do the thinking, the analysis.
o There is no software that can actually do the thinking for you.
 Data may be messy – textual – multimedia.
 You need to give thought to efficient data management.
 You need to find out what literature there is around your research topics.
 Qualitative data usually cannot be reduced to numbers.
 If you ARE just trying to reduce the data to numbers,
o have you properly understood the reasons for doing qualitative research?
o will the sample size and/or sampling method be telling you anything of value at all? (Many qualitative
samples are small and not proper random samples).
o see the numbers you are generating only as pointers to more thinking and researching about where
and why there are anomalies or exceptions. This may mean more data collection, more thinking,
more testing.
 There are many ‘approaches’ to analysing qualitative data. See Methodologies.
 Do you have theories you wish to test, challenge or enlarge upon? See How and what to Code..
 Are you seeking rather to generate theory or an account which emerges from the data (a bottom up
approach)? See glossary for e.g. Grounded Theory. See How and what to Code.
 Are you more interested in the way respondents use language to construct their world and themselves
and the other people in it? See Social Constructionism or Constructivism described in the Glossary.

Noticing, Collecting and Thinking model


Seidel (1998) developed a useful model to explain the basic process of qualitative data analysis. The model
consists of 3 parts: Noticing, Collecting, and Thinking about interesting things. These parts are interlinked and
cyclical. For example while thinking about things you notice further things and collect them. Seidel likens the
process to solving a jigsaw puzzle. Noticing interesting things in the data and assigning ‘codes’ to them, based
on topic or theme, potentially breaks the data into fragments. Codes which have been applied to the data then
act as sorting and collection devices.

Figure 1. The Data Analysis Process (Seidel, 1998)

Noticing and Coding


In Kelle and Seidel (1995) codes are differentiated in two basic ways; they can act as “objective, transparent
representations of facts” or they are heuristic tools to enable further investigation and discovery. At one level
the codes are acting as collection points for significant data. At another level the code labels themselves are
acting as markers or pointers to the way you rationalise what it is that you think is happening. At a third level
they enable you to continue to make discoveries about deeper realities in the data that is referenced by the
codes.
Development of codes
The way codes are developed and the timing of this process will depend on whether your research project and
your approach is inductive or deductive. This will be one implication of the methodology used in your research
project.
If you are working inductively (for instance when using Grounded Theory) you may let codes emerge from the
data as part of the noticing process that Seidel describes in his model.
If your approach is deductive you may be seeking to test existing theories or expand on them. In this case you
may develop codes which represent the sensitizing ideas concepts and themes within that theory, before you
start assigning passages of the data to those codes.
When might coding not be appropriate?
You may decide that your approach and your data do not suit a coding process.
Discourse analysis: certain traditions of discourse which might include the micro-analysis of small amounts of
data (e.g. Conversation Analysis - see Methodologies page) rely much more on the patterns, structures and
language used in speech and the written word. For particular types of discourse analyses handling large
amounts of data, there may be a place for coding of a kind as a data management device though usually not
for the purposes of thematic analysis and managing ‘interpretive’ annotations to the data as described in the
model above.
The analysis of narrative: where the researcher needs to track sequences, chronology, stories or processes
in the data (coding is often too clumsy a tool as it disregards the backwards and forwards nature of much
narrative). If working in hard copy, you might draw lines to connect different parts of a narrative together; if
working with software of some sort, you might use a hypertext approach. Do this by creating sequenced
hyperlinks between multiple places in the data as an aid to keeping track of the ‘connectedness’ of stories in
the data and the ideas that are informing your analysis. Although arguing in the context of using software for
QDA, Coffey, Holbrook and Atkinson (1996) challenged the dominance of coding paradigm “It is, therefore,
part of the attraction of hypertext solutions that a sense of dense interconnectedness is preserved, enhanced
even, while linearity is discarded”.

References
Coffey, A., B. Holbrook and P. Atkinson (1996) 'Qualitative Data Analysis: Technologies and Representations',
Sociological Research Online, vol. 1, no. 1.
Available online at: http://www.socresonline.org.uk/1/1/4.html
Gibbs, G R (2002) Qualitative Data Analysis: Explorations with NVivo. Buckingham: Open University Press.
Seidel, J. & Kelle, U. (1995) 'Different Functions of Coding in the Analysis of Textual Data' in U. Kelle (editor)
Computer-Aided Qualitative Data Analysis: Theory, Methods and Practice. London: Sage.
Seidel, J (1998) Qualitative Data Analyisis. The Ethnograph v5 Manual, Appendix E.
Available online at: http://www.qualisresearch.com/