CHAPTER III

METHODOLOGY
3.1. Introduction
Chapter Three describes the methodology and methods used in
this thesis to study ESL teacher motivation in New Zealand English
language schools. The chapter first operationalizes of the constructs of
intrinsic and extrinsic motivation for this study; then focuses on a
justification for adopting a mixed methods approach. Given the
importance of the methods used to validate the findings, justification
of them will be provided in detail. In short, macro level data from the
close-ended questions in the questionnaire provided background
information of the participants, a broad picture of motivating/demotivating factors and the degree of motivation/de-motivation. Micro
level qualitative data derived from the open-ended questions,
participants journals and interviews captures individual profiles and
was used to supplement and expand the quantitative data from the
questionnaire. The chapter then goes on to identify the research design
used in the study, including the participants and the methods used to
collect and analyse the data, and how reliability and validity is
protected. Finally, ethical issues associated with the study are
addressed.
3.2. Operationalization of theoretical framework
As noted in the Literature Review chapter, in order to address
the three research questions, the categories of intrinsic and extrinsic
underpin both the research instruments and the subsequent
interpretation of data. In brief, intrinsic motivation includes a sense of
achievement and challenge, while extrinsic motivation involves
considerations about salary, workload, work autonomy, relations with

13

colleagues and students, career prospects, job security, and school

policies.
To operationalize intrinsic/extrinsic motivation, Kassabgy et
al.s (2001) questionnaire used to study ESL/EFL teacher motivation
was adapted for the present study. The questionnaire assisted in the
collection and analysis of both quantitative and qualitative data. For
the quantitative data, 32 statements and their corresponding 32
statements were used in developing the questionnaire for this study
(refer to appendix c). The 32 statements in section one of the
questionnaire aim to elicit data about teachers values, and the
matching 32 statements in section two measure teachers perception
for their respective rewards. These 32 statements (in both sections one
and two) reflect or indicate different factors categorized under the
concepts of intrinsic and extrinsic. These factors were further coded
under different themes or major headings. For example, the statement
having a challenging job asked in section one, and the matching
statement My job is challenging asked in section two, were both
grouped under the intrinsic heading: Personal Challenge.
Conversely,

the

statement

Frequent

feedback

about

the

effectiveness of my performance from a person I report to asked in

section one, and I receive frequent feedback about the effectiveness
of my performance from a person I report to asked in section two,
were categorized under the extrinsic heading: Management
Communication. In addition, these established themes or major
headings assisted in coding qualitative data collected from the openended questions, journals and interviews. For instance, suggestions for
improvement such as clear communication channel and more
regular feedback were coded under that major theme of
Management Communication.

14

3.3 Justification for mixed methods methodology

Quantitative and qualitative approaches are sometimes deemed to
be indistinguishable (Nunan, 1996). However, Bryman (2008) notes
that, in general terms, quantitative approaches are artificial with a focus
on behaviour and numbers; while qualitative approaches are more
natural with a focus on the meanings participants express in words.
Nunan (1996), nevertheless, further observes that:
Quantitative is obtrusive and controlled, objective, generalizable,
outcome oriented, therefore it is hard [data]in contrast qualitative is
soft as it assumes that all knowledge is relative, that there is a subjective
element

to

all

knowledge

and

research,

and

that

holistic,

ungeneralizable studies are justifiable. (p.3) 40

In other words, quantitative research is more concerned with
cause-effect relationship, variables and attitudes while qualitative
research attempts to give voice to participants own meanings.
Most motivation studies discussed in the preceding chapter were
shown to be predominantly reliant on a quantitative approach.
Consequently, they have been fruitful in identifying variables through
scale-type questionnaires. However, they have also been criticized for a
lack of robust and in-depth data (Drnyei, 2007). On the other hand,
quantitative data could provide a basis for further research and
collection of subsequent qualitative data (Tashakkori & Teddlie, 2003).
In this regard, a subsequent qualitative approach may provide further
narrative data that captures individual values and behaviour through
interaction between researchers and participants.
The foregoing discussion indicates the incompleteness of a single
methodology as used in many empirical investigations. An appropriate
research methodology is important in finding answers to research
questions, and the adoption of only one is vulnerable to its
discrepancies. Neuman (1997) also emphasizes that most social

15

researchers would not be ready to concur with only one methodology,

nor would they conduct any research utilizing only one.
Arguably, then, a mixed methods approach (as named in
Creswell, 2003) combining both quantitative methods and qualitative
methods can offer additional benefits for the understanding of the
phenomenon in question (Drnyei, 2007, p. 47). Hence, a combination
of the two methods can facilitate each other so that the accuracy and
robustness of a study can be strengthened
In fact, there appears to be a trend in using mixed methods to
investigate motivation. The most frequent approach is through a
combination of quantitative questionnaires, quantifiable observations,
and elicited speech samples or interviews (Creswell, 1994 as cited in
Drnyei, 2001). For instance, Senior (2006) adopted a mixed methods
approach using a questionnaire, class observation and ongoing semistructured interviews to study language teachers drive and class
strategies. She comments that this approach can help find out the
what, why and how, which provide 41 sufficient richness
and depth of understanding to further address the validity of the
findings. In other research on language teacher motivation, Pennington
emphasizes that evaluative results needed to be complemented by
comprehensive and explanatory data (Pennington, 1995). In other
words, qualitative data can be used to supplement, validate, explain,
illuminate or reinterpret quantitative data gathered from the same
participants.
However, a mixed methods approach has its critics. It is both time
and resource consuming (Drnyei, 2007), and hence is difficult to
implement. This difficulty of implementation is discussed further in the
Data Collection section. Additionally, this approach has been described
as part of an old tradition [with] over reliance on cross-tab and casestudyopportunist eclecticism (Bullock, Little & Millham, 1992,
p.86). Bryman (2007) adds a mixed methods approach may be: 1)

16

unpredictable and redundant; 2) confusing in functions; and 3) unsure

of the how, when and why in combining the data. However, Bullock et
al. (1992) agree that if it is not adopted only for convenience, but has
one methodology supporting the other, a mixed methods methodology
will allow complementary and illuminating data.
Bryman (1992) remarks that any research approach, including
mixed methods, has to be decided by the nature of the research
question. Subsequently, Bryman (2007) lists five reasons for facilitating
two types of data provided by a mixed methods approach. These
reasons are applicable to the present study and are discussed further:
a. Triangulation: to corroborate
b. Complementarity: to elaborate, illustrate, clarify or enhance
c. Development: to identify participants or to inform the other method
d. Initiation/openness: to find new perspective or question
e. Expansion: to extend breadth and range
Triangulation
Triangulation is pursued to enhance the validity and reliability of
any research by collecting data through different means or multiple
sources. Drawing on works by Anderson (1998) and Freeman (1998),
four types of triangulation can be outlined:
a. multiple methods: different data collection methods
b. multiple data sets: a survey of different sources or a same source at
different times
c. multiple investigators: uses of several investigators
d. multiple theories: employment of more than one theory
The present study is triangulated through the first type. Data is
triangulated by multiple collection methods via self-completed
questionnaires, journals and interviews. To implement the multiple
methods triangulation, there is a comparison of the dominant factors in

17

quantitative data and qualitative data. In other words, there will be a

triangulation by whether a dominant factor in the first data may
become less dominant in the second data through numerically coding,
measuring and comparison.
Thereupon, intrinsic bias caused by a single method research
will likely be overcome by the different inquiry approaches. In other
words, investigations from different perspectives will possibly
eliminate bias by researchers.
Creswell and Clark (2007) posit that triangulation is one of the
four major designs in a mixed method approach (others are the
Embedded,

the

Explanatory

and

the

Exploratory).

Through

triangulation design, quantitative and qualitative data are collected for

supplementary purposes either to directly compare and contrast
quantitative statistical results with qualitative findings or to validate or
expand quantitative results with qualitative data (ibid, p. 62). For the
present study, the results from the quantitative closed questions in
sections 1 were validated with or illustrated by the qualitative results
from the open-ended questions in section 3 of the questionnaire and the
subsequent data from journals and interviews.
Having said that, it has to be noted that there have been
arguments about whether data from multiple methods can be consistent
with results in triangulation, or only complementary to them. Bryman
(2007) suggests that even with the same aim, data from these two
approaches may not be comparable because of the different
preoccupations and highly contrasting strengths and weaknesses.
In the confines of this study, only one out of ten questionnaire
participants had joined the journal keeping group (n = 7) and three of
them were interviewed. Hence, arguably, data triangulation in this
mixed methods approach may play a less significant role than if
providing complementary data, openness and clarifications.

18

Complementarity
The preceding literature review has pointed out that most studies
regarding general teacher and ESL teacher motivation were conducted
quantitatively and hence lacked robustness and comprehensiveness.
For this study, the quantitative research identifies areas of focus
(motivating/de-motivating factors) while qualitative data gives
substance to the focus. For example, the subsequent qualitative data
from journals and interviews will likely provide further explanation,
illustration or variables some previous researchers were unaware of.
This function is supported by some studies. Sosu, McWilliam and Gray
(2008), for example conducted a research into teacher commitment
using a mixed methods approach and comment that Preliminary
analysis for the survey informed some questions in the interview
processfind[ing] out both general and specific factors[and]
compared to the use of a single method, this should help expand the
scope of our understanding of the situation (p. 172).
Development
Neuman (1997) advocates for the application of a mixed-method
and justifies it with having the advantage of creating access to
participants and understanding of their broad circumstances. This
method will allow systematic sampling of the questionnaire returnees
and ability to locate and interview specific individuals who are of
interest to the researcher. For the present study, the quantitative data
from questionnaires assisted in identifying specific ESL teachers and
further data was collected through subsequent journal keeping and
interviews.

19

Initiation/openness
As noted in the previous chapter, there seems to be a lack of
consensus on a universally acceptable theory on work or teacher
motivation. Moreover, ESL teacher motivation in New Zealand (as
elsewhere) is under-researched. There is a need for discovery, via a
qualitative approach, before any hypotheses or conclusion can be
reached as there is possible unanticipated information.
Expansion
There have been a few longitudinal studies on teacher motivation
(for example, Dinham & Scott, 2000; Senior, 2006). The use of journal
may allow a longitudinal survey data collection over time.
To sum up, a mixed methods approach is the most appropriate for
addressing the issues in the present study because:
a. Quantitative questionnaires have been predominantly used for
research in teacher motivation offering very little complementary
data, new perspectives or extended breadth which can be provided
by the inclusion of qualitative methods
b. Corroboration of two sets of data via the mixed-methods may
provide triangulation
c. Openness is important for examining the uncharted territory of ESL
teacher motivation in the New Zealand context
3.4 Research design
The design of a research study includes the establishment of a
boundary for the study; that is, the identification of the participants, use
of instruments and data collection methods and the procedure for data
recording and analysis (Creswell, 2003).

20

3.4.1 Participants
The participants in the study were practising language teachers
teaching

English as

L2

in New

Zealand

language

schools

(administrators and former teachers were excluded). Also excluded

from the study were English teachers in community contexts, such as
Adult Community Education, or volunteer tutors with ESOL Home
Tutors (now renamed English Language Partners New Zealand).
At the time of the study, Statistics New Zealand (2008) estimated
there were about 121 operating language schools in New Zealand.
Schools approached for this study were located through three major
sources:
1)

website of English language schools in New Zealand

([Link] );

2)

website

of

finda

New

Zealand

([Link] and
3)

the TELECOM Yellow Pages directory. In total, 99 language

schools were identified (55 in the Auckland region, and 44 mainly
in Dunedin, Christchurch and other cities) by this study. Persons in
charge of language teaching in these schools were contacted by
telephone calls or emails. Amongst them, nine schools were
initially uninterested and six were unable to be contacted. For the
remaining 84 schools, in accordance with ethical guidelines,
participant questionnaire information sheets (see appendix A) and
consent forms (appendix B), as well as the questionnaire (appendix
C), were sent via emails or mail to respective contact persons to be
distributed to their ESL teachers. In addition, through the secretary
of the Teachers of English to Speakers of Other Languages
(TESOLANZ), a notice with details of this study and a contact
were

posted

on

its

website

([Link] in July,
2009 to draw their members attention. The questionnaire data

21

collection process lasted from June to August. In total, 72 returns

from participants in 26 language schools (from Auckland,
Wellington, Christchurch, Dunedin, Hamilton and Napier) were
received and were useable. It was impossible to ascertain the
participant response rate; however, ESL teachers in 26 language
schools out of the contacted 84 schools participated in the present
survey.
Sample size is a crucial factor in the credibility of a quantitative
survey. Fraenkel and Wallen ask for a minimum of 100 for descriptive
educational studies (2003, as cited in Mackey & Gass, 2005). However,
Mackey and Gass suggest that in L2 studies, the number of participants
could be smaller as long as the techniques for analysis take the
numbers into account (p. 124). Drnyei (2003b) suggests that in
accordance with statistical significance, a number of 50 participants
should be required in a credible L2 study.
Given a response of 72, the sample size for the present study may
be considered to be substantial. There are no official statistics on the
number of practicing ESL teachers in New Zealand. However,
TESOLANZ, the professional organization of language teachers in New
Zealand, covering the primary, secondary and tertiary sectors, had a
membership of 350 members in 2009. Therefore, this number of 72
participants could be a fair representation (21%) taking into account
this population of ESL English teachers, although it must be
remembered that only schools in the tertiary sector were contacted (i.e.
post secondary sector), and that community-based ESOL teachers
involved with adult immigrant and/or refugee groups were not included
in the study. TESOLANZ membership also includes such teachers.
Nevertheless, 72 questionnaires might be deemed as an adequate
sample after taking into consideration the scope of this study, and other
research regarding ESL teacher motivation such as:

22

a.

14 interviews and 199 questionnaires by Kim and Doyle (Kim &

Doyle, 1998, as cited in Drnyei, 2001)

b.

21 questionnaires by Jacques (2001)

c.

107 questionnaires by Kassabgy et al. (2001)

Questionnaire participants
A breakdown of the 72 questionnaire participants is listed in Table
3.1 below. Almost two out of three respondents were female. Over half
of them were over 41 years old, and only six were under the age of 30.
Only three had English as their L2. Regarding education and training,
83% had undergraduate degrees with certificates in ESL teaching,
while five participants did not give any educational qualification.
Furthermore, 38% had postgraduate degrees. 72% were full time staff.
Half of them have been language teachers for over 10 years, while 11%
had worked for five years or less. Finally, there was a fair split between
private language schools and schools run by tertiary institutes.
Table 3.1: Bio-data of 72 questionnaire participants Bio-data
Gender
Male
Female
Age
20-30
31-40
41-50
51 & over
Unknown
English
As first language
As second language
Qualifications (>1)
Undergraduate degrees
Postgraduate degrees
TESOL certificate
Unknown
Tenure

Number
24
48

In percentage
33%
67%

6
19
20
26
1

8%
27%
28%
36%
1%

69
3

96%
4

60
27
61
5

83%
38%
85%
7%

23

Full time staff

Casual staff
Both
Unknown
Years of teaching
under 1 year
1 to 5
6 to 10
over 11
Unknown
Under 1 year
Types of school
Tertiary language schools
Private language schools
Unknown

52
17
1
2

72%
24%
1%
3%

1
7
24
38
2
1

1%
10%
33%
53%
3%
1%

38
33
1

53%
46%
1%

24