ArticlePDF Available

Using the Theory of Planned Behavior with Qualitative Research

Authors:

Abstract and Figures

Studies adopting the Theory of Planned Behavior (TPB) mostly use quantitative methods. Sometimes, however, researchers choose to use a qualitative method because of the nature of available data (e.g., interviews) or availability of only a limited number of cases. This paper describes a study in which the TPB was used with qualitative methods to explain differences in university teaching. It focuses primarily on the methods used: qualitative data coding, data analysis and interpretation, and methods for presenting and supporting results. The study explored factors which influence university teachers to adopt teaching models based on online social interaction when an e-learning platform is used to complement undergraduate classroom teaching. Participants were 26 university teachers (15 from Australia and 11 from Italy). They responded to a semi-structured interview based on the TPB. Three approaches to use of e-learning platforms were identified: upload of materials, use of discussion forums, and computer-supported collaborative learning (CSCL). Using this approach, it was possible to highlight substantial differences in the attitudes, social influence, and perceived behavioral control among the three groups.
Content may be subject to copyright.
ISSN 2035-2034
Carlo F. Dondena Centre for Research on Social Dynamics
DONDENA WORKING PAPERS
Using the Theory of Planned Behavior with qualitative
research
Stefano Renzi, Jane E. Klobas
Working Paper No. 12
URL: www.dondena.unibocconi.it/wp12
October 2008
Carlo F. Dondena Centre for Research on Social Dynamics
Università Bocconi, via Guglielmo Röntgen 1, 20136 Milan, Italy
http://www.dondena.unibocconi.it
The opinions expressed in this working paper are those of the author and not those of
the Dondena Centre which does not take an institutional policy position.
© Copyright is retained by the authors.
Using the Theory of Planned Behavior with
Qualitative Research
Stefano Renzi
Carlo F. Dondena Centre for Research on Social Dynamics
Università Bocconi
via Guglielmo Röntgen 1
20136 Milan
Italy
stefano.renzi@unibocconi.it
Jane E. Klobas
Carlo F. Dondena Centre for Research on Social Dynamics
Università Bocconi
via Guglielmo Röntgen 1
20136 Milan
Italy
and
UWA Business School
University of Western Australia
PO Box 1164
Nedlands 6909
Western Australia
jane.klobas@unibocconi.it, jane.klobas@uwa.edu.au
Abstract
Studies adopting the Theory of Planned Behavior (TPB) mostly use quantitative methods.
Sometimes, however, researchers choose to use a qualitative method because of the nature of
available data (e.g., interviews) or availability of only a limited number of cases. This paper
describes a study in which the TPB was used with qualitative methods to explain differences
in university teaching. It focuses primarily on the methods used: qualitative data coding, data
analysis and interpretation, and methods for presenting and supporting results. The study
explored factors which influence university teachers to adopt teaching models based on online
social interaction when an e-learning platform is used to complement undergraduate
classroom teaching. Participants were 26 university teachers (15 from Australia and 11 from
Italy). They responded to a semi-structured interview based on the TPB. Three approaches to
use of e-learning platforms were identified: upload of materials, use of discussion forums, and
computer-supported collaborative learning (CSCL). Using this approach, it was possible to
highlight substantial differences in the attitudes, subjective norms, and perceived behavioral
control among the three groups.
Keywords
Theory of Planned Behavior, TPB, qualitative methods, e-learning, Learning Management
System, LMS, university teaching, Online Social Interaction, Computer Supported
Collaborative Learning, CSCL
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
2
Introduction
The Theory of Planned Behavior (TPB; Ajzen, 1991) is a well known social psychological
theory of human behavior. It assumes that what a person does in a given context (a behavior)
the formation of an intention to perform the behavior and postulates that intentions reflect
three motivational influences, attitudes, subjective norms, and perceived behavioral control.
In addition, non-motivational factors that can condition the behavior (the actual availability of
time, facilitating conditions, etc.) are represented by actual control over the behavior, or
actual behavioral control. Relevant attitudes represent “the degree to which a person has a
favorable or unfavorable evaluation or appraisal of the behavior in question, subjective
norms represent “the perceived social pressure to perform or not to perform the behavior and
perceived behavioral control is “the perceived ease or difficulty of performing the behavior ...
assumed to reflect past experience as well as anticipated impediments and obstacles” (p188).
Underlying each of these concepts, respectively, are the person’s beliefs about the outcome of
performing the behavior, about what other people want them to do, and about their abilities
and the availability of resources that will permit them to go ahead. These key elements of the
TPB are illustrated in Figure 1. The TPV is widely used in many fields of research to explain
behaviors that range from dieting to speeding to adopting new agricultural techniques to
reproductive behavior. (An extensive bibliography can be found online at
http://people.umass.edu/aizen/tpbrefs.html.)
Figure 1. Theory of Planned Behavior (TPB, Ajzen, 1991)
Use of the TPB with qualitative research methods is not so common. Methods developed for
data collection and analysis with the TPB model are mostly quantitative (Ajzen, 2004) and
qualitative methods are suggested only for the elicitation of beliefs (Ajzen, 2002). In some
cases the choice of a qualitative research method is, however, forced by the kind of data
available (e.g., interviews) or by the number of cases available which does not allow
statistical techniques to be used.
Few published studies have used qualitative research methods with the TPB (Mynarska,
2008) and there is little or no detail about the process adopted by the researcher to obtain the
presented results. In this paper the study Differences in university teaching after Learning
Management System adoption: an explanatory model based on Ajzen’s Theory of Planned
Behavior (Renzi, 2008) is used as a basis for examination of qualitative research with the
TPB. Qualitative data coding and analysis, interpretation and presentation of results are
described in detail.
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
3
The need to transform university teaching
Digital technologies are considered strategic for improvement in the effectiveness of
education systems. Several surveys have been conducted by international bodies and research
centers that monitor changes in Information and Communication Technology (ICT) strategy
and use in higher education. One of the most recent surveys (Becker & Jokivirta, 2007) is
based on data collected in 2006 from 67 institutions, mostly members of the Association of
Commonwealth Universities and Universities UK. The findings, which compare the 2006
results with those of surveys carried out in 2002 and 2004, show that use of e-learning (in
terms of the proportion of a course made available online) is increasing among universities
over time, but this increased online presence is associated with a significant reduction in face-
to-face classroom time. This change implies, not only a shift in the proportion of face-to-face
and online activities, but also changes in teaching.
The need for change in higher education teaching is supported by the literature, which argues
that online teaching requires a significant change in pedagogy and a related shift in ICT use
from delivery of educational material delivery to a platform for social and collaborative
learning (a relationships medium). Such interactive use of educational technology (Biggs,
2003) is expected to provide a productive and engaging learning environment (Kimball, 2002;
Rudestam & Schoenholtz-Read, 2002), encouragement of reflective learning (Palloff & Pratt,
1999), and support for effective learning through discourse (Kirschner, 2004; Laurillard,
2002).
Despite a substantial body of theory, there is very little field research on the effectiveness of
learning technology used as a relationships medium, but the cost-benefit of e-learning has
been investigated in several studies (Bates, 2001, Appendix). Laurillard (2007) focused on the
problem of e-learning costs in higher education. The core of Laurillard’s contribution is a
cost-benefit model for e-learning in higher education. She finds that the most effective
solutions are obtained through re-design of courses, substituting face-to-face teaching
methods in favour of online teaching methods – but only when these are based on
relationships and, more specifically, online collaborative learning. This way, teachers can
manage larger cohorts of students, reducing costs without compromising teaching and
learning quality.
Thus, to obtain high quality learning and decrease costs, higher education institutions need to
decrease the proportion of face-to-face teaching methods in their courses in favour of a higher
proportion of online teaching methods mostly based on computer-supported collaborative
learning (CSCL). Given the slow rate of transformation reported in international surveys, the
actors involved in this change process (faculty, administrators, and policymakers) should
benefit from an understanding of the factors influencing teachers to adopt CSCL in their
teaching.
The study described in this paper specifically examined why, once a university has adopted a
Learning Management System (LMS)1, some teachers transform their teaching and adopting
CSCL, while others use the LMS only to support their classroom teaching. With CSCL,
teachers integrate classroom teaching with online activities designed to encourage social
learning through student-student interaction. Teachers who use the LMS only to support their
1 A learning management system is “software designed to provide a range of administrative and pedagogic
services (related to formal education settings e.g. enrolment data, access to electronic course materials,
faculty/student interaction, assessment, etc.)” (OECD, 2005). Examples include WebCT and Moodle.
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
4
classroom teaching typically use the LMS to manage educational material, and in some cases,
assignments or quizzes or both. Some of these teachers also use the discussion forum feature
provided in most LMS software to answer specific questions raised by students, to clarify
assignment or exam details, or to permit students to continue outside the classroom (on a
voluntary basis) discussions about issues raised during classroom sessions.
The TPB as a framework for studying teaching model adoption
When using the TPB, it is important to define the behavior that is to be explained or
predicted. For this study, the behavior of interest was characterized as adoption of an online
teaching model by individual undergraduate teachers at universities that have already adopted,
at institutional level, an LMS.
Educational theorists normally distinguish between two dichotomous behaviors when teachers
adopt ICT; the information management and relationships medium approach (e.g., Rudestam
& Schoenholtz-Read, 2002). For this research, three behaviors were defined on the basis of
(a) the level of online social interaction (OSI) adopted by the teacher, (b) the presence of
course activities based on online social learning, and (c) the presence of CSCL and its
integration in course design. Table 1 describes the characteristics of each LMS teaching
model. In the table, a short name is associated with each model.
Table 1. LMS teaching models.
LMS teaching model short name LMS teaching model description
Teaching material upload (TMU) The LMS is primarily used to store records of what happened in the
classroom (audio-video recorded lectures, slides, and handouts) and to
publish course schedule information (announcements and deadlines). Some
course management activities such as assignments and quizzes may be
automated.
Either the discussion forum feature is not activated or it has been activated
as part of the default configuration provided by the School or University
and is used by the teacher as a kind of public e-mail to answer students’
questions.
Online discussion (OD) The LMS is used to store classroom related material, to publish course
schedule information, and to automate some course management activities
as in TMU. The discussion forum is included among LMS features by
decision of the teacher either taken or alone as a participant in a project
which included the discussion forum use.
The discussion forum plays a key role in OSI. It is a place where teachers
stimulate and encourage students to continue outside the classroom, on a
voluntary basis, discussions about issues raised during classroom sessions.
Teachers also use the discussion forum to answer specific questions raised
by students and to clarify assignment or exam details.
CSCL Nearly all the LMS features are used, especially the ones available to
support OSI, in addition to the features used in TMU for course material
upload and for course management, and in OD for student-student
interaction.
A key role is played by planned CSCL activities which are based on peer to
peer learning through OSI that takes place in the discussion forum or using
other suitable LMS features. CSCL activities are an integral part of the
course structure, designed with their own learning goals and usually
assessed (as distinct from the use of the discussion forum simply to extend
classroom discussions as in OD).
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
5
The elements of the TPB were used as a framework to explain the adoption of different
models. There is, however, an important difference between the model adopted in this study
and the TPB. The TPB is normally used to predict behavior, and a pivotal concept is the
intention to perform the behavior as an antecedent of the behavior itself. This study was not
concerned with prediction, but with explaining the influences on adoption of CSCL as distinct
from other online teaching models. The behavioural change had already occurred, and the
researcher was seeking to explain why. For this reason, intention is not included in the
framework adopted for this study.
Thus a teacher’s adoption of an LMS teaching model is expected to be influenced by one, or
all of, attitudes, subjective norms, perceived behavioral control, and actual behavioral control.
In addition to considering the relative influences of each of these higher level factors, it
should be possible to identify how specific attitudes, subjective norms, perceived behavioral
control factors, and actual behavioral control variable factors might explain adoption of each
of the models.
Methodology
Sample
Teachers’ adoption of an online learning model occurs within the context of the university
and the national system in which the university is immersed (in this case, the national
university system). This context may affect any or all of the influences on adoption of a
teaching model. For this reason, participants were selected from two different university
systems, Australia and Italy, and four universities (two in Australia and two in Italy).
Information about the characteristics of the two university systems and the universities
included in this study was collected to verify that the teachers’ work environments were
sufficiently compatible to permit the results from all universities to be pooled for analysis. In
fact, Australian and Italian university systems have degrees structured in a different way but
the three years of undergraduate studies after high school are very similar and comparable.
LMS use was not compulsory for teachers in any of the universities included in this study.
Recruitment of teachers was managed through the teaching and learning unit of each
university. Although, between them, the participating universities had around one thousand
undergraduate teachers, it was possible to identify fewer than 50 teachers who were thought to
be using their universities’ LMS to support online social interaction (OSI) either through use
of the discussion forum or CSCL. Interviews were held with 26 teachers, 15 of them
Australian and 11 Italian. Their characteristics are summarised in table 2.
Several environmental and individual factors were identified as potential confounding
variables in this study. The environmental variables were the national university system,
subject taught, if the course is mandatory or elective, class size, the number of teaching staff
available for the described course, and teaching horizon (i.e., if the course was already taught
by the teacher or if they expected to teach the course again). Individual factors include
demographic information, teaching experience, general technology experience, and
experience in teaching with technology.
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
6
Table 2. Characteristics of the participating teachers.
Variable Item Frequency Percent (%)
Country Australia 15 58%
Italy 11 42%
Sex F 8 31%
M 18 69%
Age 31-40 7 27%
41-50 14 54%
51-60 3 11%
Over 60 2 8%
Position Lecturer 20 77%
Full/Ass Professor 6 23%
Teaching experience (years) 5 2 8%
10-15 10 38%
16-20 7 27%
21-30 5 19%
over 30 2 8%
Class size (no. students) < 50 7 27%
50-100 5 19%
120-170 7 27%
over 250 7 27%
IT for teaching use Novice 2 8%
Average 12 46%
Expert 12 46%
LMS experience (years) 2-3 3 11,5%
4-6 20 77%
over 6 3 11,5%
Total 26 100%
Australian and Italian appointments are not perfectly matching but it was possible to divide
the teachers into two broad categories: 20 of them were lecturers (or similar) and six were full
professors (or similar). Their teaching experience ranged from five to over 30 years and they
managed classes of very different sizes, ranging from less than 50 to over 250 students.
Teachers rated themselves as experts in using information technology for teaching in 12
cases, while 12 rated themselves as average, and two as novice. Most the teachers had 4 to 6
years of experience in LMS use, while three had 2 to 3 years of experience, and three had
over 6 years of experience. None of the teachers were novice users of LMS
Data collection
The data for this study were obtained mainly through interviews. The interviews usually
lasted about one hour, but in some cases, teachers described their experiences in greater detail
(and with great enthusiasm) and those interviews lasted up to two hours.
The interviews were informed by teachers’ responses to a questionnaire which prompted them
to reflect on the issues to be raised in the interview, by notes taken during and after the
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
7
interview by the researcher, course outlines supplied by teachers, and other information about
teaching courses and departments retrieved online from the universities’ Web sites.
Questionnaire development
The interview outline was defined on the basis of the TPB model adopted for this study. One
or more questions were developed to address each element of the model (behavior of interest,
attitudes, subjective norms, perceived behavioral control, and actual behavioral control).
To uncover attitudes, teachers were asked to list what they considered to be positive and
negative outcomes of teaching in general, technology, online teaching, and OSI.
Subjective norms were identified by asking questions about the influence of people and
institutions (university, department director, colleagues, students) on use of the LMS and
adoption of OSI (in general and for the described course).
Perceived control factors came from questions about workload (including the influence of
teaching load on course design), technology (possible weaknesses of LMS features, factors
encouraging or blocking the move to another LMS, plans to develop further online activities
in general, and online activities including group activities and OSI (factors encouraging or
blocking the change of course design).
Actual behavioral control was investigated by asking questions about the degree of autonomy
and flexibility the teacher had in designing the course, organizational factors associated with
LMS adoption by teachers, organizational factors that limit the choice or adoption of a new
LMS for the course, the kind of support available at the university (to design online activities
and to solve problems while the course was running), and other support services that could be
useful to support online teaching activities. This information was cross-checked where
possible with the university technical or teaching development unit staff who supported the
LMS.
The interview outline was completed by adding questions about potential confounding
variables (e.g. subject taught, class size, age, sex, teaching experience, technology experience,
etc.) and other information to complete each teacher’s profile.
Once defined, the questions were grouped into homogeneous sections. Based on the estimated
duration of the full interview, the questions were broken into two parts: a self-completion
questionnaire to be sent to the teacher before the interview, and questions to be used in a face-
to-face interview. To complete the questionnaire development, an Italian version of the
questionnaire was also prepared.
Coherence between interview outline and research design was checked using mind mapping
software (MindManager). The research design was illustrated in a map that included the
elements of the research model together with a list of potential confounding variables, and
other information to be recorded. After preparation of the interview outline, each question was
included in the map to check coherence. Appendix 1 shows the map matching interview
questions with variables.
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
8
Questionnaire and interview administration
A first group of 18 interviews took place face-to-face (eight in Italy and 10 in Australia) over
a three-month period. Subsequently, eight interviews were conducted at a distance by phone
or Skype (a common software allowing computer-to-computer and computer-to-phone calls).
The procedures adopted for face-to-face interviews with university teachers in Italy and
Australia were the same. After the teachers had fixed a meeting for the interview, they were
sent an e-mail with three documents attached: a Research Information Sheet, a Consent Form,
and a short questionnaire they could fill out before or during the interview.
During the interview, the researcher went through the answers to the questionnaire with the
teacher and asked the additional questions in the interview protocol. All interviews were
digitally audio-recorded, with the teachers’ permission. In addition, key points raised in
response to interview questions were handwritten on a printed copy of the interview form. All
the teachers were kind, collaborative, and passionate as they shared their experiences. After
each interview, important information collected during the interview was recorded in an Excel
file along with some basic data about the teacher and their course.
Qualitative data analysis
For each participant, information was collected in different ways: the self-completion
questionnaire, the course outline, the additional information from the university Web site,
transcripts of the audio-recording, and field notes. All this information was merged into a
single Word document which followed the structure of the interview outline for each
participant.
Preliminary analysis
The main goal of the preliminary analysis was to classify the LMS teaching models adopted
by the teachers participating in the study. Additional goals were to obtain an initial
description of the sample and to identify indicators (such as demographic characteristics) that
might be used to group teachers by characteristics not associated with their behavior.
Data associated with each of these goals were extracted manually from the aggregated Word
documents and typed into an Excel file. Navigation through the data was achieved with
simple features like highlighting data with different colours and applying Excel data filters.
Mind mapping was used to obtain a simple graphical representation of teachers with different
characteristics.
Following preliminary analysis, participants were grouped by the LMS teaching model they
adopted (as described in Table 1). All the subsequent data analysis was performed analysing
the three groups of participants (and, thus, the three behaviors) separately.
Data coding
QSR NVivo 2, a Computer-Assisted Qualitative Data Analysis Software (CAQDAS)
package, was used to code the data. The software was suitable for this study because the basic
analytical framework was already defined and the work was more about classifying the
collected data and finding similarities in cases rather than uncovering new theory or exploring
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
9
complex concepts. Easy to follow guides to the use of the software were found with the
software download and the support of some books (e.g. Bazeley & Richards, 2000; Gibbs,
2002).
Three main activities were performed in NVivo before starting to work with the aggregated
documents: set up the node tree, prepare a coding guide (inside NVivo) to support coding of
the aggregated documents and import aggregated documents into NVivo.
A node (in NVivo) is an object representing an idea, a theory, or other characteristics
associated with data contained in a document. Nodes can be linked in a hierarchical way to
form a node tree. The researcher prepared the nodes, linked in a node tree, to code the text of
the aggregated documents. In parallel, coding guidelines were also created. Node tree
preparation was based on the structure of the map prepared to match the interview outline
with the research design (see “Questionnaire development” in this paper). The final
preparation activity was to import the aggregated documents from Word. Word documents
were converted to RTF, the format required by NVivo.
Coding was done by associating the text in the documents with nodes. Because the nodes
reflected the elements of the TPB and the documents were also ordered by these elements,
most coding involved identification of concepts referred to by the teacher in their responses to
each question. A response may refer to one or several concepts. All concepts were associated
with nodes as described below. Figure 2 shows a typical NVivo session working on an
aggregated interview document. Four windows are open at the same time on the screen.
Figure 2. Working with integrated interview documents in NVivo.
The window open at the right shows the document on which the researcher is working. For
each question, there is the title of the question followed by the teacher’s answer as written on
the self-completion questionnaire, the transcribed interview answer to the question, and
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
10
additional notes. In the same window, on the right hand side, NVivo shows nodes already
associated with this text.
The top window in the centre contains the coding memo document. The document records,
for each question, the title of the question followed by the node name and the full path within
the node tree.
In the bottom centre window, the case memo document is open. Here all the ideas, thoughts,
and observations that emerged during the document analysis are written.
The window on the left side of the screen is the NVivo coder, the tool that associates a part of
the document text with a node. The coder shows the full node tree from which the human
coder can select the node to be used. If the available nodes do not adequately cover a concept
recorded in the text, the researcher can add a new node.
The link between a piece of text and a node is made by selecting a piece of the document in
the window on the right (highlighted in Figure 2), selecting a node from the node tree (also
highlighted in Figure 2), and clicking on the Code button at the bottom of the NVivo coder (in
Figure 2 the cursor is shown over this button). The text-node association is then shown in the
right side of the document window (in Figure 2 it appears at the same level as the highlighted
text).
During the document coding process for this study, nodes increased from the initial 100 to
508. Once coding of all documents was completed, a node review was performed to merge
nodes with similar meaning. The nodes were reduced to 384. The reorganization was
performed with the Node explorer tool which easily allows nodes to be moved and merged.
An extract from the list of nodes is available in Appendix 2.
A first attempt to look at data patterns
After the node review process, the researcher used different NVivo tools to read the text or to
‘look inside’ the data. Because the basic explanatory framework was already established for
this study, the main reason to look at the data was to review or verify the text-node
associations. NVivo allows this operation to be performed in a very simple way using the
Node explorer tool. In Figure 3, at the left, is the Node explorer window (with the selected
node “school teaching yes”). All the pieces of text linked to that node from all the documents
can be browsed in the right hand window.
Qualitative coding and analyses done through NVivo contributed considerably to
understanding the phenomena investigated in this study, but the tools available in NVivo to
explore data and discover data patterns did not produce remarkable results. A further attempt
to discover patterns in the data was performed by exporting the data to other software for
further analysis.
The coded data were exported from NVivo to the SPSS statistical software package. To do
this, a data table was generated in NVivo through the Assay Scope feature of the Search
function. The resulting table lists participants by rows and nodes by columns. For each
subject-node cell the presence of a ‘1’ means that for that subject that node is coded, i.e. that
characteristic is present for that subject. On close inspection of the file, though, the researcher
concluded that the data were too sparse to support analysis using statistical software.
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
11
Figure 3. Using Node explorer to find all the texts associated with a node.
Changing strategy: Looking for patterns in qualitative data using
interpretative analysis
To find common data patterns, the researcher decided to continue qualitative data analysis
without the mediation of CAQDAS software. The decision to move in this direction was also
supported by the opinion of other researchers who think that the use of software for
qualitative data analysis makes the researcher distant from data, makes data fragmented, and
forces the researcher to make the data homogeneous for computer coding needs (Holbrook &
Butcher, 1996; cited by Vidovich, 2003).
Participants were thus grouped by the LMS teaching model they adopted and the aggregated
documents (the Word files) were analysed to identify patterns shared by teachers who adopted
the same LMS teaching model.
For each LMS teaching model, the researcher prepared a Word document which reproduced
the mind map structure of Appendix 1: at the highest one, the explanatory variables (i.e.,
attitudes, subjective norms, perceived behavioral control, and actual behavioral control),
followed by the specific aspects of each variable addressed in the associated questionnaire and
interview questions. Under each question, the related answers and notes from all participants
who adopted the same LMS teaching model were copied.
Table 3 contains an example of the new document’s structure and content. The first row
(marked as 1 in the left hand column) contains the behavior (the LMS teaching model), in this
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
12
example CSCL. The second row (marked as 2 in the left hand column) contains the name of
the first level explanatory variable, in this case, “attitudes”. This is followed by a list of the
specific aspects of the first level variable that were explored in interview questions. In the
case of attitudes they were: attitudes to teaching in general, online teaching, and OSI. In the
example, the specific aspect considered for attitudes is online teaching attitudes (3). For each
aspect of the independent variable, the related questions and the answers obtained are
recorded. In the example, the questions related to online teaching attitudes are “What role
does online teaching play in your teaching?” and “About the online teaching activities
designed in your unit, what are the positives and negatives? (For teacher/student)” (4). The
answers and notes from the aggregated document for each subject adopting the same LMS
teaching model are recorded below the questions (5).
Table 3. Steps of data analysis after data were grouped by the TPB model structure.
No. Description Document content
1 Behavior (LMS
teaching model) CSCL
2 Level one independent
variable Attitudes
3 Specific aspect of
independent variable Online teaching attitudes
4 Questions about
“online teaching
attitudes”
What role does online teaching play in your teaching?
About the online teaching activities designed in your unit, what are the
positives and negatives? (for teacher/student)
5 Answers to questions
about “online teaching
attitudes” (each answer
is followed by
participant codes)
I think it’s a good chance for all to have a voice. If you have a class
discussion, many students say nothing (face-to-face).Um for the
teacher, it’s interesting I think to see what students are thinking (having
online interaction). Most often they won’t tell you (face-to-face) [laughter].
(Teacher SL-01, AU, female)
More people can contribute to their learning so they’re not only having me
as a lecturer, they may have other students teaching them. (Teacher SL-02,
AU, male)
An additional positive for the student that has come out in our evaluations
is that students who are not on campus don't feel isolated because they get
to know the other students who are here on campus and even though they're
not, they can get a bit of a feel for it, that's a positive for me too, the
students are happier. (Teacher SL-04, AU, female)
I sort of thought for the students as well, student to student interaction and
student teacher interaction are positives (in online teaching). (Teacher SL-
05, AU, female)
6 Description of relevant
aspects discovered
from answers
Another (positive) aspect which is shared by all the teachers in this group is
interaction with students (or among students). We have to remember that
the questions were about online teaching and the fact the teachers talk about
student interaction means that they strongly associate online teaching with
OSI.
Teachers included this element as positive for the teacher, as positive for
student, or for both cases. They used different kind of words, sometimes
just few a words demonstrating how online teaching is, for them, obviously
associated with OSI.
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
13
Answers presented in (5) are an example of results obtained through an iterative process
based the constant comparative method of qualitative data analysis (Glaser & Strauss, 1967).
With this method, the researcher codes quotations from the transcripts that exemplified a
concept. As more quotations are coded, these quotations are compared and combined with
similar quotations in the transcript. The pattern coding process was based on the procedures
suggested by Miles and Huberman (1994). The elements take shape as concepts and are
completed by a description of all relevant aspects as analysis continues (6).
At the end of this iterative process, the researcher was able to identify the specific elements
shared by teachers who adopted the same LMS teaching model. Figure 4 shows an example of
a concept discovered through this process. The title of the discovered concept is followed by a
short description and comments, and by the extract of answers supporting the concept along
with each participant’s identifying code so the reader can find the other characteristics of the
participant in the tables.
Figure 4. An example of concept discovered from answers after analysis.
Classroom teaching and online teaching are integrated
Most of the teachers belonging to this group (six out of eight) stated explicitly that classroom
teaching and online teaching are integrated.
My general belief is that … face-to-face with appropriate use of online is the best mix.
(Teacher SL-01, AU, female)
So you offer a more rounded learning experience. … I believe from my educational
experience, by washing yourselves with as many different mediums - book, text, writing,
online, typing, reading, chatting, you are learning far better.
(Teacher SL-02, AU, male)
It’s another space for teaching. Yeah, it’s just like a classroom, except it’s virtual. (Teacher
SL-03, AU, male)
It's completely integrated. (Teacher SL-05, AU, male)
(Online teaching) is not different … is not something added (to the classroom teaching) or
an alternative. (Teacher SL-07, IT, male, translated from Italian)
It is difficult for me to think about online teaching as something simply added on to
classroom teaching … it is completely integrated.
(Teacher SL-08, IT, male, translated from Italian)
Looking at this example, it is quite clear that it would have been much more difficult to obtain
the same results with the computer-based software process described earlier where answers
are coded all together by node. In fact, the results emerged quickly from the beginning of this
analysis, demonstrating clear differences among influences on adoption of each of the three
LMS teaching models.
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
14
Presenting the results
On completion of the analysis, the discovered concepts, completed with further descriptions
and quotations from answers were packed together with a header description. These are the
findings which identify the second level of the explanatory variable for each high level
variable in the TPB framework.
In the example presented in Table 3, the concept discovered was “Online teaching includes
interaction with students (or among students)”. It was accompanied by the relevant aspects
described by the researcher in row 6 and by the quoted answers in row 5. The discovered
concept “Online teaching includes interaction with students (or among students)” contributes
to understanding of which attitudes influence adoption of the CSCL model.
For each TPB factor (attitude, subjective norm, and perceived behavioral control), the results
were gathered together in a table which shows the findings detailed by second level variable
and teacher. Figure 5 shows an extract from the summary table of the attitudes of teachers
who adopted the CSCL teaching model. The “Variable” column lists the concepts that
emerged from the analysis along with some basic demographic factors to enable comparison
of teachers in each group. The concept that emerged from the analysis shown in Table 3 is
highlighted and the number in parenthesis shows how many teachers shared that concept; an
“x” in a column shows that a teacher expressed the concept. In the extract of the attitudes
table in Figure 5, the concept “Online teaching includes interaction with students (or among
students)” was shared by 8 teachers (all the teachers who adopted the CSCL teaching model)
and in the figure is visible the “x” showing that the teachers coded as SL-01 and SL-02 shared
that concept. The full table is available in Appendix 3.
Figure 5. Extract from an attitudes table (with example concept highlighted).
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
15
Concepts obtained in the way described above required, in some cases, a further
transformation to be expressed in the form of an attitude, subjective norm, or aspect of
perceived behavioral control. For example, the concept associated with perceived behavioral
control for circumventing LMS constraints was recorded in the analysis as “LMS has some
constraints but it is ok”, using the words of the participants who expressed this concept. For
final reporting, it was expressed more directly as an aspect of behavioral control: “I am able to
circumvent LMS limits”. For each LMS teaching model the researcher prepared a table with
the findings obtained from data analysis and the same concept transformed in TPB form. The
table associated with the above examples is available in Appendix 4.
The final concept descriptions for each of the independent variables were gathered together in
a diagram that summarized the second level independent variables identified for each LMS
teaching model. Thus, the analysis produced three TPB diagrams, one for each teaching
model. The TPB diagram for the CSCL teaching model is available in Appendix 5.
Identifying differences in behaviors
The aim of this study was to identify factors explaining why, in an LMS environment, some
university teachers adopt an LMS teaching model based on online social learning or CSCL,
rather than a teaching model that simply supported classroom-based activities. Therefore,
while presenting the results of the qualitative data analysis the focus was on the TPB factors
that explained the adoption of each LMS teaching model. The discussion of results, instead, is
focused on the differences and similarities among behaviors and on possible relationships
among TPB elements and other data (e.g., demographic data).
To support the discussion, the researcher prepared a table that summarised all the relevant
findings related to demographic variables and to each TPB factor (attitude, subjective norm,
and perceived behavioral control, detailed by second level variable) for each behavior. Figure
6 shows an extract from that table. The full table is available in Appendix 6.
Figure 6. Extract from table showing online teaching model differences.
The extract shows two of the three LMS teaching models considered (“CSCL” and “Online
discussion”). The concepts that were discovered in the analysis are listed for each aspect of
the research framework (e.g., Attitudes, Online teaching) against each LMS teaching model.
The concept is followed by the number of teachers in the group who shared it. Comparing this
number with the number of teachers belonging to the group (at the top of the column), the
reader can see how much that concept is shared among members of the group.
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
16
The position of concepts in rows helps the reader see if the concept is unique for a teaching
model or if it is shared by other teaching models. For example, the concept “Online teaching
includes interaction among student” is listed in the column “CSCL” in the row “Attitudes -
Online teaching” but there is an empty cell in the “Online discussion” column because there is
no similar concept. In the next row instead, the concept “Online teaching permits flexibility”
in the “CSCL” column is in the same row of the concept “With online teaching there are
benefits” in the column “Online discussion” because the two concepts are considered similar
by the researcher.
Results
In the full report of this study, the results of the data analysis were described by LMS teaching
model and included: a description of the teaching model (the behavior of interest);
demographic data for the teachers who used the model; the results obtained for the second
level of the explanatory variables grouped by attitudes, subjective norms, and perceived
behavioral control; a “conversion table” showing how concepts were re-written in the TPB
format; and, finally, the graphical representation of the TPB model showing the findings for
that LMS teaching model (Renzi, 2008). Because the focus of this paper is detailed
description of the qualitative research method, only a synthesis of the discussion is presented
here below.
There are clear differences in the attitudes, subjective norms, and perceived behavioral control
of the teachers who adopted each of the three teaching models. These differences cannot be
explained by differences in environmental conditions or in attitudes to teaching in general. In
fact, all the teachers who participated in the study have a strong positive attitude to teaching.
Demographic variables such as age, sex, position, country, years of experience in teaching
and in LMS use, and course characteristics had no observable influence on adoption of LMS
teaching model. The only relevant demographic difference among teachers in the three groups
concerns the co-presence of pedagogical skills and confidence with technology use for
teaching.
In the TPB, personal skills are considered antecedents of attitudes, subjective norms, and
perceived behavioral control and, thus, can contribute to different behaviors (Ajzen, 1991).
This appears to be the case with adoption of LMS teaching models.
All the teachers who adopted CSCL have both pedagogical skills and confidence in their
ability to use technology in teaching. Their attitudes toward online teaching include that
Classroom teaching and online teaching are integrated, Online teaching includes interaction
among students, and Students learn more with OSI2. Few are not conscious that others have
influenced their adoption of the CSCL model (i.e., OSI adoption for online teaching had no
subjective norms for them), instead focusing on other motivations for adoption. Their
perceived behavioral control shows that they are confident they have the pedagogical skills to
adapt course designs to the technology available; having good technology skills, they are
prepared to change the current LMS for a more effective LMS. The combination of
pedagogical skills and technology skills is also evident in their belief that I would be able to
develop more online activities with OSI.
2 Italics are used in this section to highlight concepts that emerged from the analysis.
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
17
Teachers who adopted the Online Discussion (OD) model believe that Online teaching is a
complement of classroom teaching and OSI allows information exchange. They refer to few
social influences on their teaching model adoption. Half of the teachers in this group are able
to adapt their course design to the technology available, while it seems that the other half is
not looking for advanced LMS features (LMS does what I need it to do). While CSCL
teachers are prepared to develop more online activities without conditions, OD teachers
require more resources associated with recognition by their university. They are prepared to
develop more OSI-based activities, but they are less confident with pedagogy and need
support to design and run such activities and need to understand how much effort is required.
The teachers who adopted the TMU model have different attitudes toward online teaching,
including a strong preference for face-to-face teaching and refusal to use OSI because it is not
needed or can generate problems. Their high perceived behavioral control for technology is
consistent with the simple features required to upload teaching material.
The greatest differences in the attitudes, subjective norms and perceived behavioral control of
teachers who adopted each of the three models can therefore be seen in:
Attitudes. There are marked differences in attitudes to the role of online teaching. Teachers
who adopted CSCL see classroom teaching, online teaching, and social interaction as
essentially related with one another. Teachers who use OSI as a complement to classroom
teaching (those who adopted the OD model), also believe that this is the appropriate role for
OSI. Teachers in both these groups believe that OSI has benefits for students, even when it
involves more work for the teacher. Attitudes to adoption of OSI among teachers who do not
adopt it (those who adopted the TMU model) vary, but generally they do not see that OSI
benefits students or teachers, and indeed may consider OSI problematic. They prefer face-to-
face teaching to online teaching, but like many teachers who adopted the OD model,
recognize the benefits that automation of teaching processes with the LMS can provide.
Subjective norms. While teachers who adopt the OSI-based models appear to be largely self-
motivated, teachers who use the LMS primarily for teaching material upload seem more likely
to be influenced by others.
Perceived behavioral control. Differences in perceived behavioral control reflect teachers’
differences in pedagogical and technology skills. The teachers who adopted CSCL are able to
circumvent LMS constraints and are keen to develop more online activities with OSI. Their
technology skills make them prepared to look at more effective LMSs. OD teachers have
weaker PBC, reflecting what appear to be weaker pedagogical skills. These teachers are
prepared to change LMS, but they need training. To develop more OSI-based activities, they
need support to design and run such activities and need to know how much effort is required.
The PBC of TMU teachers reflects the simple characteristics of their model. They do not feel
need to change technology and they require more resources to develop more online activities.
Discussion and Conclusion
The choice to adopt a qualitative research method requires a difficult decision for a researcher
because, compared with quantitative data analysis, it requires more effort to demonstrate the
quality of the results obtained (Hoepfl, 1997). Leading researchers do not agree on apparently
simple issues, such as recording interviews (e.g., Lincoln & Guba, 1985; Patton, 1990). The
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
18
discussion about how to evaluate qualitative research is still open. There are several relevant
issues to consider.
Sampling is a big issue in qualitative research. Some of the classic literature suggests that 30
to 50 interviews be conducted (e.g., Denzin & Lincoln, 1994) while other authors claim that
there are no strict criteria for sample size (Patton, 1990). Some papers published in well
ranked journals include papers with only a few cases (e.g., Grant, 2004 who used four cases).
In some cases, as here, the researcher wants to study a phenomenon that is not widespread, so
the number of potential participants is limited by a small population.
The emergence of regularities or commonalities during the analysis is one of the criteria
recommended to stop data collection (Guba, 1978). This is the case of the study presented
here. Several data patterns which demonstrate clear and strong differences among the teachers
who adopted different teaching models emerged during data collection and were confirmed by
the analysis.
Transferability is considered another critical issue in qualitative research, analagous to
generalizability in quantitative research. According with Lincoln and Guba (1985),
transferability cannot be specified by the researcher. The researcher can only provide detailed
information that can be used by the reader to decide if the findings can be applied in a new
context.
Detailed description of methodology, interim results, and transformations in the various
phases of analysis, as presented in this paper, can help readers determine the transferability of
the results. In terms of content, the strong and consistent differences in the TPB factors
associated with the three teaching models suggest that the results are transferable, although
this can only be confirmed by a more extensive study. On the other hand, the shared positive
attitude toward teaching of all the teachers participating in this study points to the need for
care in designing new research to be sure that voluntary participation does not imply self-
selection of only the most enthusiastic teachers keen to share their online teaching
experiences. The risk of finding results that apply only to enthusiastic teachers is partially
confirmed by the dropout of the teachers who indicated their availability to be interviewed to
their universities’ teaching and learning units but did not answer the researcher’s request to
participate in the study.
The context in which the research is conducted is a key factor in considering transferability.
In this study, context was represented by actual behavioural control, but no differences were
found. The absence of effects of context on teaching model adoption in this study could be
associated with the personal characteristics of the teachers, such as pedagogical and
technological skills, but it still seems strange that evident differences between national
university systems has no effect. Broadening the context through more extensive research
would enable differences among university systems, including the effects on university
teachers of pedagogical support and training, to be investigated.
Despite the small sample of teachers in this study, the TPB framework and qualitative
research method combined to provide insights into teaching model adoption. Sampling is
often considered a major shortcoming that can limit the researcher’s ability to capture critical
aspects of studied behaviors and, consequently, affecting the transferability of the results of
qualitative research. The positive outcome of this research nevertheless supports Patton’s
(1990) proposition that the credibility of qualitative research can be evaluated through the
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
19
richness of information gathered, triangulation of data, and thorough analysis and
presentation, rather than sample size.
Acknowledgements
We would like to thank Prof. Laura Bernardi who invited Stefano Renzi to share his
experience in using TPB with qualitative methods at the REPRO WP5 working meeting:
“Coordinating comparative qualitative research for the REPRO”, 22-23 September 2008,
Rostock, Germany, Max Planck Institute for Demographic Research. The presentation
prepared for the meeting and the comments of the participants inspired this paper.
References
Ajzen, I. (1991). The Theory of Planned Behavior. Organizational Behavior and Human
Decision Processes, 50, 179-211.
Ajzen, I. (2002). Constructing a TpB questionnaire: Conceptual and methodological
considerations. Retrieved October 13, 2008, from the World Wide Web: http://www-
unix.oit.umass.edu/~aizen/pdf/tpb.measurement.pdf
Ajzen, I. (2004). Frequently Asked Questions [Web page]. Retrieved October 13, 2008, from
the World Wide Web: http://people.umass.edu/aizen/faq.html
Bates, T. (2001). National strategies for e-learning in post-secondary education and training.
Paris: UNESCO/IIEP. Retrieved, 21 September 2007, from the World Wide Web:
http://unesdoc.unesco.org/images/0012/001262/126230e.pdf
Bazeley, P., & Richards, L. (2000). The NVivo Qualitative Project Book. London: Sage.
Becker, R., & Jokivirta, L. (2007). Online learning in universities: selected data from the
2006 Observatory survey - November 2007 [Online report]. The Observatory on
Borderless Higher Education (OBHE). Retrieved, 21 September 2007, from the World
Wide Web: http://www.obhe.ac.uk
Biggs, J. (2003). Teaching for Quality Learning at University: What the Student Does
(Second ed.). Buckingham: Society for Research into Higher Education and Open
University Press.
Denzin, N. K., & Lincoln, Y. S. (1994). Handbook of qualitative research. Thousand Oaks,
CA: Sage Publications, Inc.
Gibbs, R. G. (2002). Qualitative Data Analysis: Explorations with NVivo. Buckingham: Open
University Press.
Glaser, B., & Strauss, A. (1967). The Discovery of Grounded Theory: Strategies for
Qualitative Research. Chicago: Aldine.
Grant, M. M. (2004). Learning to teach with the web: Factors influencing teacher education
faculty. The Internet and Higher Education, 7(4), 329-341.
Guba, E. G. (1978). Toward a methodology of naturalistic inquiry in educational evaluation.
Los Angeles: University of California, Center for the Study of Evaluation.
Hoepfl, M. C. (1997). Choosing qualitative research: A primer for technology education
researchers. Journal of Technology Education, 9(1), 47-63.
Holbrook, A., & Butcher, L. M. (1996). Uses of qualitative data analysis software in
educational research: the literature, the hard questions and some specific research
applications. Australian Educational Researcher, 23(3), 55-80.
Kimball, L. (2002). Managing distance learning: New challenges for faculty. In R. Hazemi &
S. Hailes (Eds.), The Digital University - Building a Learning Community (pp. 27-40).
London: Springer-Verlag.
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
20
Kirschner, P. A., Martens, R. & Strijbos, J.W. (2004). CSCL in Higher Education? In
J.W.Strijbos & P. A. Kirschner & P. A. R. Martens (Eds.), What We Know About
CSCL: And Implementing it in Higher Education (pp. 3-30). Norwell MA: Kluwer.
Laurillard, D. (2002). Rethinking University Teaching: A Framework for the Effective Use of
Educational Technology (2nd ed.). London: Routledge.
Laurillard, D. (2007). Modelling benefits-oriented costs for technology enhanced learning.
Higher Education, 54(1), 21-39.
Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Beverly Hills, CA: Sage
Publications, Inc.
Mynarska, M. (2008). Qualitative studies and TPB (unpublished presentation). Rostock,
Germany: Working Meeting: coordinating comparative qualitative research for the
REPRO WP5 European Project, 22-23 September 2008, Max Planck Institute for
Demographic Research.
OECD. (2005). E-Learning in Tertiary Education: Where do We Stand? Paris: Organisation
for Economic Co-operation and Development (OECD).
Palloff, R. M., & Pratt, K. (1999). Building Learning Communities in Cyberspace: Effective
Strategies for the Online Classroom. San Francisco: Jossey-Bass Publishers.
Patton, M. Q. (1990). Qualitative Evaluation and Research Methods. Newbury Park, CA:
Sage Publications, Inc.
Pratt, M. G. (2008). Fitting Oval Pegs Into Round Holes
Tensions in Evaluating and Publishing Qualitative Research in Top-Tier North American
Journals. Organizational Research Methods, 11(3), 481-509.
Renzi, S. (2008). Differences in University Teaching after Learning Management System
Adoption: An Explanatory Model Based on Ajzen's Theory of Planned Behavior.
Unpublished PhD Thesis, The University of Western Australia, Perth, Western
Australia.
Rudestam, K. E., & Schoenholtz-Read, J. (2002). Overview: The coming of age of adult
online education. In K. E. Rudestam & J. Schoenholtz-Read (Eds.), Handbook of
Online Learning: Innovations in Higher Education and Corporate Training (pp. 3-
28). Thousand Oaks, CA: Sage.
Vidovich, L. (2003). Methodological framings for a policy trajectory study. In T. O'Donoghue
& K. Punch (Eds.), Qualitative Educational Research in Action - Doing and
Reflecting (pp. 70-96). London: Routledge Falmer.
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
21
Appendix 1: Matching interview outline with research
design
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
22
Appendix 2: Extract from the list of NVivo nodes
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
23
Appendix 3: Attitudes of the teachers who adopted the
CSCL model
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
24
Appendix 4: From data analysis to TPB model for the
teachers who adopted the CSCL model
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
25
Appendix 5: TPB diagram for the CSCL teaching model
Dondena Working Paper 12 Theory of Planned Behavior and Qualitative Research
26
Appendix 6: Differences between online teaching models
... The data collection and analysis methodologies developed for the TPB model are primarily quantitative (Ajzen et al., 2004). Nonetheless, in some circumstances, the choice for qualitative research methods is dictated by the type of data available (e.g., interviews), or the quantity of the cases available, which precludes the use of statistical approaches (Renzi & Klobas, 2008). Few published articles have used qualitative research methodologies with the TPB (Mynarska & Brzozowska, 2022), and there is little or no data on how the researcher arrived at the conclusions given. ...
... The interview outline was created using the TPB model, which was used for this investigation. Each aspect of the model, such as attitude, the subjective norm, and perceived behavioral control, was addressed with one or more questions (Renzi & Klobas, 2008). The study's interviewees were asked to express positive (or negative) feelings about digital change, in order to reveal their attitude to it. ...
... The COVID-19 pandemic forced the Ministry of Finance to rapidly adopt a digital transformation and new working practices. Changes in the work behavior of civil servants at the Ministry of Finance have occurred, along with changes in their attitudes, subjective norms, and perceived behavioral control (Renzi & Klobas, 2008). Attitude refers to an individual's overall evaluation of behavior, subjective norms refer to social pressure from others, and perceived behavioral control (PBC) refers to an individual's perception of his/her ability to perform the behavior. ...
Article
Full-text available
Introduction/Main Objectives: The COVID-19 pandemic forced the Ministry of Finance of the Republic of Indonesia to carry out a digital transformation by employing the concept of the Flexible Working Space (FWS) to ensure its civil servants can use the Work From Anywhere (WFA) system, including the Work From Home (WFH) system. These concepts are expected to help civil servants to optimize their performance. Background Problems: FWS has been practiced for two years, but detailed information on how the civil servants’ green work behavior has helped in maintaining their performance has not been investigated. Novelty: The researchers also studied how the digital transformation has changed the working practices among civil servants at the Ministry of Finance, and do they employ more green practices now? Discussion about the civil servants' green work behavior during this digital transformation is still limited. Research Methods: This study applied a qualitative technique to analyze the data obtained from an online qualitative survey, focus group discussions (FGD), observations and docu­mentation. These research data were validated through a triangulation approach before being analyzed using the theory of planned behavior (TPB). Finding/ Results: The positive aspects of the Ministry of Finance’s digital transformation, which occurred during the pandemic, are civil servants can work faster, more efficiently, more flexibly, and they are more focused, but they have less contact with other people. While the identified disadvantage is that the internet network is occasionally unstable, thus disrupting the learning and discussion processes. The civil servants at the Ministry of Finance had used several green practices before the pandemic began, such as saving water, saving electricity, and to some extent they had gone plasticless and paperless. After the pandemic happened and FWS was implemented, these practices became more prominent, producing better results compared to the results before the pandemic. The pandemic also brought other advantages, including carbon emissions reductions and time savings. Conclusion: The FWS concept in the Ministry of Finance changed the work behavior related to the digital transformation, and the green work behavior of the civil servants. This is based on the TPB which includes attitudes, subjective norms, and behavioral control. Digital transformation using the FWS concept has had positive and negative impacts on the civil servants' green work behavior, and in maintaining their performance.
... After the vignette was read, the interview consisted of questions grouped into three categories based on the Theory of Planned Behavior by Ajzen (1991): "Behavioral Outcomes," "Normative Concerns," and "Control Factors," with an additional final question about the participants' experience. This methodological approach, aligning with the Theory of Planned Behavior, has been effectively employed in past research (Renzi & Klobas, 2008). For the Behavioral Outcomes, participants were asked four questions: one on the advantages, one on the disadvantages, one on positive feelings, and one on negative feelings concerning the inclusion of participants with IDD into their judo classes over the next three months. ...
Article
Full-text available
This qualitative study investigates the behavioral beliefs and attitudes of judo teachers toward including individuals with intellectual developmental disorders (IDD) in their judo classes. Conducted across Slovenia, Portugal, France, and France-Polynesia, the research, within the Theory of Planned Behavior (TPB) framework, involved interviews with 21 judo teachers using a vignette of a judo practitioner with IDD. Data analyzed using NVivo (2020) highlighted that judo teachers generally view inclusion positively, noting enhancements in pedagogical skills and sensitivity alongside significant benefits for all participants. Judo teachers expressed that inclusion fosters empathy and group cohesion but also raises concerns about potential slowdowns in group progress and increased cognitive demands. Emotional responses varied from fulfilment and pride to concerns over inadequacy and group management. These findings suggest a need for targeted training and accommodations to address challenges and enhance the effectiveness of inclusion. Keywords: Judo, Intellectual Developmental Disorders, Inclusion, Physical Education, Qualitative Research, Teacher Attitudes, Theory of Planned Behavior, Behavioral Beliefs
... Although uncommon, using qualitative methodologies with the TPB allows for a more comprehensive exploration of these aspects. Previous studies, such as those by Renzi and Klobas (2008) This study aims to use TPB to understand auditees' intentions regarding the implementation of audit recommendations. By evaluating their attitudes, subjective norms, and perceived behavioral control, the research seeks to identify factors crucial for developing their behavioral intentions toward implementing audit recommendations. ...
Article
Full-text available
Previous studies on the implementation of audit recommendations have primarily focused on identifying influencing factors without delving into the specific behaviors of individuals responsible for their implementation. This study aims to analyze auditees' intentions to implement audit recommendations using Ajzen's Theory of Planned Behavior (TPB). While TPB is typically applied with quantitative methods, this research employs a qualitative approach to gain a deeper understanding of respondents' attitudes, motivations, and values. The findings reveal five main themes that influence auditees' intentions to implement audit recommendations: lack of engagement, unorganized organizational pathways, the role of decision-makers or policymakers, mutual reluctance toward coordination, and gaps in knowledge and qualifications. Addressing these themes simultaneously is essential for the effective implementation of audit recommendations. Auditees' attitudes toward implementing audit recommendations significantly impact their intentions. However, subjective norms and perceived behavioral control can pose barriers, diminishing their commitment to the implementation process. This nuanced understanding of the factors affecting auditee behavior is crucial for developing strategies to enhance compliance with audit recommendations.
... The more favorable one's attitude and subjective norms, and the greater their perceived control, the stronger their intention to perform the behavior. TPB has been frequently used to understand clinicians' behaviors [15], and also to examine medical educators' behaviors, including incorporating non-biomedical focused subjects into medical curricula [16,17], including qualitatively to understand teaching in higher education [18]. Therefore, the aim of this study is to explore medical educators' perceptions about what factors contribute to their intention to teach public health in Indonesian undergraduate medical schools. ...
Article
Full-text available
Purpose: Medical educators are central in ensuring future doctors have sufficient public health skills. Attitudes, norms, and perceived control about the significance of teaching a subject determines whether or not it is taught and how well. This qualitative study aims to explore medical educators’ perceptions about what factors influence their intention to teach public health in Indonesian undergraduate medical schools.Methods: Semi-structured interviews were conducted with eighteen medical educators from different Indonesian medical schools. Interviews were analyzed thematically using the Theory of Planned Behavior domains: attitudes, subjective norms, and perceived behavioral control.Results: Five subthemes emerged under these domains: attitudes (defining public health); subjective norms (room in the medical curricula; teaching and assessment); and perceived behaviour control (medical educator confidence; institutional support). Most participants had a limited understanding about the scope of public health. This coupled with an already overcrowded medical curriculum made it challenging for them to incorporate public health into the medical curriculum dominated by clinical and biomedical content. Although believing that public health is important, medical educators were reluctant to incorporate public health because they were not confident incorporating or assessing content.Conclusion: Strong institutional support is to improve public health quality and content in the medical curriculum. Including public health educators in discussions is critical.
... Depending on the behaviour in question, important others may include family, friends or spouse (Ajzen, 1991). In professional fields, important others may include job supervisors (Renzi & Klobas, 2008) or lecturers in a university environment concerning students (Cooper, Kenny & Fraser, 2012). Of particular relevance to this study, Taylor (2015) reported that the two main normative influences on students' subject choices in their study in the UK, were parents and teachers. ...
Conference Paper
Full-text available
Since the beginning of 2015, national exam in Indonesia was no longer used as graduation standard. This policy was made because of negative effects of this standardized test toward educational system in Indonesia. One of the effects is the teaching method used by teacher that rely on memorization and drill practice. The aim of this study is to see whether this new policy has an impact in teaching learning process in mathematics classroom. The samples of this study were 17 mathematics teachers from six different provinces in Indonesia. Before this policy issued, 12 teachers use conventional learning, while five teachers use unconventional learning methods such as problem solving, open ended problems, and contextual learning. The results of questionnaire reveal that from 12 teachers who use conventional method, only two teachers that change their teaching method after the new policy about national exam released. It means that 83% of teachers who use conventional learning in this study keep using the same method. Easy in the implementation and easy to understand by students are the main reasons of teachers decision to keep implementing this method. This finding shows that the changing of intended curriculum by government as decision maker is not followed by the changing of implemented curriculum-pedagogy- by teachers. Based on the result of this study, the recommendations are given to incorporate the vision and mission of government and teachers in order to reach the desired goal as well as for doing further research on this issue.
Article
L’inactivité physique chez les personnes avec limitations fonctionnelles est une préoccupation importante de santé publique. La pratique d’un sport peut être un moyen d’augmenter le niveau d’activité physique des personnes avec handicap. Ce projet vise à : (1) définir le profil identitaire des adeptes de para hockey, et (2) mieux comprendre les mécanismes associés aux intentions de poursuivre la pratique du sport. Le projet, mené à l’aide d’une méthode mixte, est constitué de deux études, basées sur le modèle de l’identité sportive et la théorie du comportement planifié. Dans la première étude, 115 para hockeyeurs ont rempli un questionnaire portant sur l’identité sportive. Ensuite, sept athlètes ont participé à une entrevue individuelle, pour aider à comprendre leurs intentions de maintenir la pratique sportive. Quatre-vingt-dix pour cent des répondants ont une identité sportive élevée, alors que 20 % se disent enclins à essayer d’autres disciplines. Les résultats de l’étude 2 indiquent que les attitudes, les normes sociales et le contrôle perçu incitent à continuer la pratique du para hockey mais aussi à s’initier à d’autres disciplines sportives. Néanmoins, les participants sont plus prédisposés à surmonter les obstacles pour le para hockey que pour s’initier à de nouvelles activités sportives.
Article
Full-text available
Workarounds are a common practice in a broad range of organizational and technological contexts, which has received much attention in information systems research. They are sometimes considered negatively, associated with business risks and noncompliance, and sometimes positively, as a source of innovation and improvement. Although workaround-based process improvements typically adopt the workaround if its consequences are not harmful, this is not always the best option. The paper aims to pave the way for leveraging workarounds for process improvement by understanding problems that motivate them, so alternative solutions can be developed for these problems. Despite many proposed explanations, an in-depth understanding of the reasons underlying employee decisions that lead to workarounds is still needed. To this end, following a qualitative research approach, data were collected in six case study organizations, investigating workaround decisions made by employees through the lens of the theory of planned behavior. The principal unique theme running throughout the findings is that workaround motivation stems from perceived misalignment between organizational, local unit, or personal goals and their realization in business processes. Workarounds are enabled by various factors, including perceived workaround opportunities in the organization. These findings are abstracted to form an explanatory model as a contextual adaptation of the theory of planned behavior to explain workaround intentions. The proposed model explains workaround decisions in a way that can leverage this understanding to promote concrete process improvements.
Article
Full-text available
A study of qualitative researchers who have submitted and/or published their research in top-tier North American organizational and management journals reveals the evaluative criteria-in-use at these journals. Specifically, when asked to take the perspective of an (a) author, (b) evaluator (reviewer or editor), and (c) judge of the comparisons between qualitative and quantitative research, three publication tensions were evident. The author offers the metaphor of ``fitting a round peg into an oval hole'' as a means of capturing these tensions and as a vehicle for organizing the various ways these tensions can be managed.
Article
Professional development at postsecondary institutions has primarily been supported through a central faculty development center. With purposive sampling, this small case study explored the ways in which teacher education faculty members were influenced to participate in decentralized professional development focused on learning to teach with WebCT. Results suggest that university faculty were influenced by both intrinsic factors (including convenience, comfort, common interests, and future purposes) and an extrinsic factor (external pressures). This research suggests that these factors also contributed to the faculty's satisfaction with the professional development. A follow-up interview explored how faculty coped with budgetary limitations and encouragement to pursue other distance education technologies.