CHAPTER 4

RESEARCH DESIGN AND METHODOLOGY

4.1 INTRODUCTION

As a preview, chapter two highlighted the literature review of the study. The chapter

emphasized what other scholars that is, from the International and global perspective

have discovered about the study topic. From the literature, diverse views can from

various authors were explained. The researcher then embarked on explaining the

theoretical framework and made use of three theories namely the McKinsey 7-S

model, Vygotsky’s socio-cultural theory and lastly Bronfenbrenner’s ecological theory

to explain the phenomena under study. The study aimed at understanding the

importance of researching on the provision of education to learners with autism from

a system perspective to determine the how diverse issues affect the process. The

study objectives centered on training needs for teachers, policy issues and

challenges faced by teachers in educating learners with autism. This chapter shall

dwell on the research design and methodology.

4.2. RESEARCH PARADIGM

In order to understand the concept of research paradigm, it is crucial to regard the

context with which it works. The researcher chose to explain the term research

paradigm in the context of research wherein there is need to describe what is and

how it should be viewed or treated. In this case, it speaks to the nature of data from

the general understanding of the worldview (Denzin & Lincoln, 2020). According to

Crawford (2020), this allows researchers to put on a different set of eyes with which

to see the data to be generated.

Since research paradigms work as lenses to researchers, they offer first of all a

standard of agreement in terms of what all researchers present their researches on

(Denzin & Salvo, 2020). Secondly, such standards also afford what Oliver (2021),

called philosophical orientations which are basically fundamental beliefs useful for

understanding research inquiry. This was very crucial as it allowed the researcher to

carry out studies with preconceived ideas of how the world works such that as they

interpret the data of any new study, it aligns to the worldviews available already.

Such therefore allows new information to make sense and understandable. Without

the worldviews, researchers would neither have the basis to carry out studies nor the

interpretation thereof. Hence, research paradigms provide the scope for

understanding research procedures, outcomes and interpretations (Oliver, 2021).

The researcher adopted the Interpretivist research paradigm for the study. The

paradigm showcased reality as a social construct which depended on the subjective

experiences of the participants which cannot be attributed as thus. An Interpretivist

paradigm allows the researcher to avoid contaminating research outcomes as it

emphasizes ownership of research outcome to participants. This is through the

detailed provisions of the research setting, that is, the participants and their social

environment. The interpretive approach is also inductive in nature as it does not limit

the scope of knowledge produced in research. This is due to the fact that it usually

allows the participants to dwell on what they think is relevant to the phenomena

under question (Creswell & Creswell, 2018).

The Interpretivist paradigm was utilized because it allowed for the detailed

explanations which enabled provisions of meanings and interpretation to originate

from the participants in the study. The Interpretivist paradigm allowed the researcher

access to reality as a socially constructed phenomena that depended on subjective

experiences (Denzin & Lincoln, 2020). This paradigm was suitable to ensure

meanings and interpretation are extracted from study participants. All the above

allowed the study to be about exactly where it was carried out and not anywhere

else. Interpretive research approaches are useful in understanding even the smallest

of minorities or study participants. Every study is done with the understanding of its

unique position in comparison with the rest of the world.

The Interpretivist paradigm thence determined the researcher to make use of

research instruments appropriate for such a study. In this case, the chosen

instruments were in the form of semi structured interviews, observations and

document analysis. The interpretive research paradigm was also chosen because it

allowed the researcher to be able to dwell on qualitative variables like efficacy and

attitudes which give due validation to participants’ subjective views (Flick, 2022).

Such subjective views allow practitioners to then be able to understand how the

Kingdom of Eswatini experience the teaching of learners with autism as a country

unlike doing so as part of the world. This therefore ensures that whatsoever

intervention is carried out becomes specific to the context thereof.

4.3. RESEARCH APPROACH

The researcher made use of the qualitative research approach as it is flexible for

pursuing phenomenon. Such a quality was a useful one as pursuing phenomenon in

its original environment allowed participants to share their lived experiences.

According to Oliver (2021), the nature of qualitative research is that it provides

valuable insight by adopting context and content in explaining research phenomena.

Denzin and Salvo (2020), posit that qualitative research takes the form and shape
that allows participants to present it in ways that make sense to them and avoids

robust and complex terms. Qualitative research is about expressing subjective

experiences (Crawford, 2020). Such is vital since it allows researchers to study on

aspects which are not quantifiable.

A qualitative research approach was relevant to the study as it entails studying

particular issues or phenomena by focusing on participant experiences within their

social milieu. According to Flick (2022), qualitative research approach gives thick

descriptions thick in detailed explanations about participants’ lived experiences and

their contexts. The matching of context and content of research is at the core of

results and justifies the need to carry out similar research as done in other regions. It

also allows for the comparison between different study contexts which allows

practitioners to learn from the best available. The variables under study that is

aspects like challenges and experiences were not quantifiable hence the usefulness

of such an approach. This also required the actual experiences as determined by the

participants themselves without a pre-determined scale of some sort.

Crawford (2020) is of the idea that qualitative research is characterized by studying

phenomena in its natural milieu, signifying the understanding of each research

matter in the respective setting and to the understanding of the participants. The

approach has the advantage of making research studies about the participants as

outcome of studies are presented and explained in terms of the participants

understanding and attributed meaning (Creswell & Creswell, 2018). What could be

seen as a disadvantage is the subjective nature of results as stated above which

may alienate the results from the researcher. In addition, even in environments

where people share common values, how they experience the world is always
subjective hence the likelihood of too broad experiences to make sense (Gupta &

Gupta, 2022).

The best way to cater for the weaknesses of the approach was through ensuring that

the sample frame was well represented such that various dynamics can be catered

for. In addition, all the steps followed were detailed enough to ensure the data

represents that indeed it has emanated from the participants’ experiences and was

independent of the researcher. Lastly, the method used for data analysis and

interpretation looked for common themes and not sporadic ideas or points for the

purpose of generalization.

4.4. RESEARCH DESIGN

Creswell and Creswell (2018) state that research designs are the tools used when

collecting, analysing and interpreting research data. Flick (2022) views them as the

strategy used in the planning of research activities. These strategies then provide

guidelines on procedural directions in every research study. According to various

scholars, there quite a good number of qualitative research designs which sometime

overlap amongst each other (Creswell & Creswell, 2018; Denzin & Lincoln, 2018;

Flick, 2022; Gupta & Gupta, 2022). Examples of the abovementioned include

narrative research design, case study research design, ethnography research design

and phenomenology research design.

The types of research designs available in qualitative research include the narrative

research design which looks at identifying participants with a relevant story to the

study matter. The researcher then collects their data through interviews or

documents. Though it focuses on individual experiences, it also acknowledges

multiple life experiences and interpretations. Ethnography is another option where

the researcher has to immerse themselves in a particular study group’s way of life.

That is done to try and understand why and how that particular group do their things.

There is also phenomenology research design which seeks to understand and

explore experiences and meanings attached to human life. It does not impose any

preconceived ideas on the experiences being sought (Crawford, 2020).

The choice of the research design was determined by the underpinning philosophical

assumption of the study. The study’s philosophical assumption was on the basis of

participants’ experience. The researcher therefore made use of a phenomenology

research design. This was suitable to explore the lived meanings attached to the

experiences of the participants. In addition, the study did not require the researcher

to attach any meaning or impose any ideas to the study. Hence the phenomenology

research design was the most appropriate for the study.

The study adopted a phenomenology research design since it is flexible enough,

which is a valuable attribute to the study allows one to explore the experiences and

meanings attached to them by the participants themselves (Creswell & Creswell,

2018). Phenomenology is basically the philosophy of experiences. Hence, the only

source of such meanings attached to experiences can only be derived from

participants’ lived experiences. Using a phenomenology design allows for authentic

accounts of lived experiences which are explained and interpreted by the participant

him or herself. There are different types of phenomenology design that is narrative,

hermeneutic, interpretive and descriptive phenomenology (Creswell & Creswell,

2018). The researcher thence adopted the interpretive phenomenology as it allowed

for a focus on essence and intentionality.

4.5. DATA COLLECTION METHODS

The researcher made use of two data collection methods. The two are namely semi

structured interviews and document analysis. The following sections dwelled on

describing the two methods. In addition, the reasons for choosing those particular

methods are given and an outline of what the two methods entail were also

presented.

4.5.1. Interviews

Interviews are an interaction which involves exchanging of perspectives between two

individuals or even a group (Flick, 2022). Such exchanges are meant to produce new

information or knowledge. Scholars have posited that interviews are more than just a

tool for data collection but have other qualities. The researcher used two kinds of

interviews namely semi structured interviews and also focus group discussions. The

two were unpacked below in detail as to how they were utilized.

Semi-structured interviews

A semi structured interview is a data collection technique that allows researchers to

make use of pre-determined open-ended questions when interview participants. In

addition, the same instrument allows flexibility since it also enables the researcher or

interviewer to make follow up questions from participants’ arising responses. It is

regarded as a balanced technique since it is always guided by key areas of interest

which are allowed to develop through natural dialogue. Researchers are free to
explore any new ideas as long as they are within the confines of the themes. Semi

structured data collection techniques possess some of the qualities from structured

and unstructured interview techniques (Gupta & Gupta, 2022).

During data collection, the interviewer made use of open ended questions to solicit

responses from the participants. As the responses were provided, the researcher

noted other key matters arising from the responses. This opened up room for further

probing and detailed explanations. Such an approach was useful as it allowed for as

much detail as possible to be elicited from participants. This ensured that the scope

of data generated was not limited to the researcher’s guide alone but also other

important aspects which arose from the exercise.

Kumar (2018) identified some of the characteristics of semi structured interviews as

flexibility, allowing in depth exploration, and adaptable to various contexts and

situations and they also utilize predetermined open ended questions. The flexibility

allows researchers plenty of room to probe for detail as much as possible. The same

quality also advocates for its other characteristic of allowing in-depth exploration.

Most importantly, the technique is adaptable to various contexts that is whether

extremely educated or illiterate and as such has proven useful over time. Semi

structured interviews were used on school principals and also policy makers.

Focus group discussions

A focus group discussion is an interview data collection technique. It involves the

selection of specialised participants who have knowledge of a specific field who

discuss a chosen topic in an interactive manner (Oliver, 2021). The discussion

involves having the researcher as the moderator and the whole process is recorded
for later use. Characteristics of focus group discussions include a selected panel

based on expertise or desired criteria, use of a moderator and lastly interaction

between chosen participants such that the setting is conducive for such. This

research technique was utilised for teachers educating learners with autism.

The steps of carrying out a focus group discussion involve ethical clearances and

confirmation of informed consent by participants. After that, the researcher appraises

the participants on the expectations of the process. The researcher has to then

choose or select a moderator to take charge of the session. The moderator then runs

the focus group discussion whilst the session is being recorded. The interviewer can

also record separate notes as and when necessary (Kumar, 2018).

It is important to note that the focus group discussions must be more than one to

ensure data triangulation. This was not a problem for this study as the focus groups

were chosen per school such that there was a good number of the discussions. The

role of the moderator is a critical factor in the quality of data as issues of participation

are easily affected by such factors like group dynamics and or environmental

settings. The moderator must ensure that sufficient probing is done for the best

possible quality outcomes. That can only be ensured if and when the participants are

comfortable and free enough to participate.

4.5.2. Document analysis

Document analysis refers to the systematic documents review to gain understanding

of certain procedures and guidelines informing the way things are done in a given

circle. It can be done in research and also in policy studies wherein policies are
analysed through reviews and evaluation of the documentation against real life

practice (Flick, 2022).

Document analysis was very critical for this study since it allowed the researcher to

carry out a review and evaluation of the available or supporting documents with

regards to the education of learner with autism in the Kingdom of Eswatini. It was

very critical that the researcher gets to review material or documentation on teaching

policy, strategies and progress reports on the teaching of learners with autism.

Document analysis was also critical since it aided the researcher determine the

progress on the education of learners with autism from when the practice began in

the Kingdom to date. In addition, documentation on other important variables such

as teacher qualifications for teaching learners with autism, the deployment of

teachers skilled in teaching learners with autism as well as documentation of

workshops for pre and in-service trainings on the education of learners with autism

were other crucial information only reviewable through document analysis (Flick,

2022).

The technique is very related and matched to the different questions as it addresses

all the research questions in more than one way. All the documentation should be

able to provide what the required skills to educate learners with autism are.

Secondly, the documentation should also be able to provide answers to questions of

challenges facing educators and also the policy issues guiding the education of

learners with autism. In conclusion, it is true that this data collection technique is

matched to the research questions and literature all the way. Just by doing a

document analysis alone, one can be left with a detailed overview of the current

state of affairs as far as the education of learners with autism in the Kingdom of

Eswatini is concerned.
4.6. POPULATION AND SAMPLING

4.6.1. Population

The term population can be understood within the confines of further specifications

or demarcations. It can be a target population or a study population. A target

population therefore refers to the specific or total group of individuals carrying

particular traits identified for a particular research problem. This is therefore the

whole broad group that will benefit from the given study (Hennink & Kaiser, 2022).

For this study, the target population was all the primary school teachers in grades

one to three in the Kingdom of Eswatini. The second group were all primary school

principals in the Kingdom of Eswatini. And lastly, all the policy makers associated

with policy making for primary school education in the Kingdom of Eswatini. Study

populations determine scope and boundaries of any given study. Such scope and

boundaries ensure that research outcomes can be inferred to particular contexts,

situations or groups of people.

On the other hand, a study population refers to a selected group of individuals

characterizing similar traits of interest to be subjected to a research study. Study

populations are determined by issues of accessibility, feasibility and also practicality.

Data or results emanating from a study population is attributable or can be inferred to

the target population (Oliver, 2021). The researcher focused on four primary schools

each from the four regions of the Kingdom of Eswatini’s primary education system.

These included grade one to three teachers from four government schools, the
school principals, and policy makers for primary school education. The study

population were selected on the basis of meeting desired criteria matching the

research questions. The school teachers represent the primary outlet in terms of

educating learners with autism. The principals were selected as they are the

overseers of all educational activities within the school environment including the

educating of learners on the spectrum. Lastly, the policy makers were chosen due to

their role of making of the policies guiding the education of learners with autism.

In total, semi structured interviews were carried on two policy makers. Similarly, a

total of four semi structured interviews were also carried out on the selected four

principals representing one school in each of the four regions of the kingdom of

Eswatini. Lastly, the researcher carried out one focus group discussion per school

wherein all the selected teachers from each school participated in the sessions. Thus

all in all, four focus group discussions were carried out.

4.6.2. Sampling

Sampling involves finding researchable units from the entire research population.

Sampling as a procedure is very critical to the quality of research outcomes hence

the need for it to be carried out carefully. As a rule, sampling is done differently with

regards to the nature of the study for instance, qualitative studies are reminiscent

with small sample sizes whereas quantitative have bigger quantities.

According to Creswell and Creswell (2018), although there are various techniques in

sampling, each usually matches a specific research tradition such as case study.

The various sampling techniques fall under either probability sampling or non-

probability sampling. In probability sampling, there is random selection of participants

or subjects such that population units have similar and calculable chance of making

it into the sample. It therefore means that any of the units within the population

stands a chance to be selected into the study sample.

The researcher guided by the needs of the study used a non-probability sampling

strategy. Non-probability sampling entails that it is impossible to have a

representative sample through random selection of units to be in a sample. As such,

non-probability sampling simply states that there is no equal chance of making it to

the sample among population units. Therefore, for this study, the researcher had to

be purposeful hence the use of purposive sampling wherein the participants were

chosen by meeting specific purpose. The researcher followed a subjective sampling

technique based on judgement which ensure that there was no fixed probability of

population units being selected into the sample (Creswell & Creswell, 2018).

As already stated, a study sample is the definition of items or objects representing

the bigger target population. The sample is characterized by the traits of the larger

population. For this study, the researcher determined that the sample size for the

three populations were two policy makers accessed through the Ministry of

education and the Deputy Prime Minister’s office who are key ministerial custodians

of the education portfolio in the Kingdom of Eswatini. Only the two participants were

selected into the sample which according to Kumar (2018), was acceptable because

it satisfied the minimum 10% expected sample size. The second group of units in the

sample were primary school teachers. These were determined through

comprehensive purposive sampling (Creswell & Creswell, 2018) of grades one to

three classes of learners with autism. The acceptable sample units for this category

were three teachers per school that is one from each grade multiplied by four

schools totaling a sample of twelve participants. Last but not least, principals were
chosen into the sample through comprehensive purposive sampling. Selection of the

principals was based on belonging to the selected respective schools thus the units

in numbers were four participants.

The researcher chose to use purposive sampling to ensure selected participants are

directly involved with educating of learners with autism or have specific knowledge

crucial to address the research questions. This particular area of study is not

particularly overflowing with participants as they only form a relatively small and

unique population in Eswatini. Schools educating learners with autism may be limited

and hence the adoption of purposive sampling ensured the efficient use of resources

and time.

The criteria for selection was based on meeting at least one of the three different

purposes that is, being an educational policy maker within the kingdom of Eswatini or

being a primary school teacher for grades one to three classes of learners on the

spectrum within the Kingdom of Eswatini or a principal in a primary school where the

teachers educate learners with autism.

4.7. DATA ANALYSIS, DISCUSSION AND PRESENTATION

The researcher utilized inductive thematic analysis for this study. Inductive thematic

analysis enables the researcher to get meaning through data induction in the

process of generating of themes and patterns. Since it uses an inductive analysis,

inductive thematic analysis formulates theoretical foundations from using data as the

basis. According to Braun and Clarke (2021), researchers have freedom during data

analysis and are able to generate tailor made meanings against the a priori codes.
Inductive data analysis allows for the capture of complex dynamics in any qualitative

data set which allows for easy of comparisons, descriptions and even reflections on

the showcased themes. Themes are a representation of interpretations of data

based on established data or ideas formulated from the research phenomenon which

determines how the researcher comes up with the data analysis (Braun & Clarke,

2021). Inductive thematic analysis involves the researcher comparing frequencies in

themes, examining theme relationships and identifying co-occurrences of themes.

From the above processes, the removal of insignificant themes, the merging of

similar themes and creation of new ones could then be done.

As already indicated, data from the interviews and documents was analysed through

inductive thematic analysis which according to Braun and Clarke (2019), posit that it

is about locating ranges of differences, relationships and commonalities which is also

termed coding in a given data set. There two domains that is, emergent and priori

codes which were determined during the inductive thematic analysis. Before data

generation, the researcher embarked on the development of priori codes. The

emergent codes were generated from the responses or data.

Inductive thematic analysis is a useful tool in the identification and interpretation of

patterns or themes in a data set. According to Elliot (2018), the use of such a

technique often leads to the development of new insights and understanding. Such a

quality is in adherence to the principles of qualitative research which endeavours to

establish context specific research outcomes simply unique to each specific or given

social milieu as supported by Fryer (2022).

In addition, the inductive thematic analysis aligns enhances trustworthiness of

qualitative research through thematic analysis principles. Scholars agree that seven-
stage procedures go over and above the inductive analysis of data. Important

features brought by the technique include the clarity it brings which establishes

trustworthiness, coherence in the analysis process, and the substantiation of findings

on the basis of solid evidence (Fryer, 2022).

As already highlighted the study followed Braun and Clarke’s (2021) seven stage

inductive thematic analysis. These were used on both for both priori and emergent

themes. The seven steps of analysis started with transcription, followed by reading

and familiarization. The researcher then coded the data to enable easy searching of

themes. The stages then follow a reviewing of themes process. Following the review

of themes were the last two stages concerned with the defining and naming of the

themes and final analysis stage.

Each of the ten interview recordings from both semi structured interviews and focus

group discussions were transcribed as separate files. The researcher ensured the

identifying information was properly anonymized. After use, all material like the

recordings and other subsequent files used in the analysis were secured safely

under a locked cabinet.

The interviews and documents were used as the primary units for analysis. These

were transcribed using the seven phase thematic analysis (Braun and Clarke, 2021)

to identify common and repeated a priori and emergent themes. Initial coding and

theming was done through the inductive analysis. The inductive process also arose a

priori themes. From the initial findings from the interviews, theme descriptions were

initiated. From the interview transcripts, the researcher clarified emergent themes

through defining and naming them. The final phase is addressed in the findings

reported below (Elliot, 2018). Importantly, the researcher also noted on top of the
textual analysis, gestures and other non-verbal as communicated during the

interviews.

The transcriptions of data was done from the verbatim from both literature material

and interview residue. This enabled easy coding wherein through coding priori and

emergent themes will be established. Scholars indicate that there are different types

of codes that are commonly regular in qualitative research (Elliot, 2018; Fryer, 2022;

Kiger & Varpio, 2020). Qualitative research has room for researchers to use already

existing codes with your data. These are otherwise known as a priori codes and are

developed prior to the examination of generated data. Usually the majority of

qualitative researchers opt for the development codes during data coding. These are

termed inductive codes developed by the researcher through examination of

verbatim from interviews. Through coding priori and emergent themes, the final

outcome was established (Kiger & Varpio, 2020).

4.8. TRUSTWORTHINESS

This study focused on the more complex aspects of human behavior which are

rather difficult to quantify for instances challenges and experiences. Although

complex, research variables are also important to determine and when studying such

phenomena, researchers cannot use any other ways for ensuring the

appropriateness of the study. In qualitative research researchers use the concept of

trustworthy to determine the truthfulness and accuracy of a study. Trustworthy is

ensured through the correct structural use of measures which enable transparency

and evaluation in the research field and such that findings are credible, reliable and

truthful. Kumar (2018) posit that the measures for ensuring trustworthy are issues of
confirmability, transferability, credibility and lastly dependability. The mentioned four

elements make up the trustworthiness framework wherein credibility refers to the

level of truth of the study, dependability refers to consistency of the study,

transferability refers to applicability of the study and lastly but not least confirmability

which speaks to the neutrality of the study.

4.8.1. Credibility

The issue of credibility refers to the level of confidence that is placed in the

truthfulness of findings in a research study. This concept emphasizes or seeks to

establish if a given study’s results are believable and have been correctly interpreted

with respect to the original study participants’ views. This measure is very crucial to

research as it borders on aspects of quality of and confidence in research as

believable and of acceptable standards.

This was ensured through researcher’s extended involvement in data generation.

The researcher manage to gain the trust of participants through spending more time

on the ground, which formed a bond with the participants. This was useful for

member checking which a useful tool is also used for ensuring credibility of the study

findings. The procedure of member checking entailed taking final research outcomes

study participants to approve whether the findings were accurate and represent their

beliefs.

Another criteria used by the researcher was the harnessing of the triangulation

technique (Kumar, 2018). The researcher made use of semi structured interviews,

focus group discussions and document analysis to generate data and the data which

emerged were merged to come up with credible themes. Secondly, data was also
triangulated from different sources as participants wore different portfolios such that

there was policy makers, primary school principals and primary school teachers.

That was yet another way of finding the study credibility as the generated data can

be attributed to at least three different sources (Flick, 2022). The researcher made

sure that all the processes followed were clear, coherent and rigorous enough to

ensure authentication.

4.8.2. Transferability

The measure of transferability speaks to the arguing for generalizing or transfer of

the study findings to other settings with different participants (Flick, 2022). Scholars

however argue that transferring study findings to other situations is rather

challenging due to specificity issues with which qualitative research is synonymous

with (Heslop, Burns & Lobo, 2018; Kumar, 2018; Oliver, 2021). In qualitative

research, researchers emphasize the contextual issues surrounding or influencing

any study such that aspects like culture or even economic or political status are

accounted for when providing details of research studies (Gupta & Gupta, 2022).

Despite the above, in some unique cases, some findings may be applicable to

similar situations and that is why rich and thick details are provided in every research

to ensure comparisons and matching of contexts are done where possible.

According to Kumar (2018), the importance of transferability is based on the

applicability of research studies to other contexts. Thereby, determining the practical

usefulness of the research studies in relation to issues of practice and or policy

making. It answers to the usefulness of research studies over and above the
population or context it is intended to. Whether a study can be generalised to other

situations or context is of utmost importance.

The details of the study were provided to the last comprehensively to the last core.

(Oliver, 2021). The research settings were described in rich and thick details in order

to give a full account of the research context. The researcher also made sure to give

a full detailed account of the participants’ details and characterization as well as any

other important study material or variable for instance the highlights on the education

curriculum, school infrastructure, educational teacher training programs and policies.

The facilitation of transferability by providing dense description of information about

the participants, the research context and setting will ensure other potential users

are capacitated to assess and evaluate how applicable the study outcomes are to

other contexts. Thick descriptions ensures that the researcher provides clearly and in

detail all the followed research procedures from data collection, study context, to the

research outcome. Such practice of providing thick descriptions creates avenues for

other researchers to also replicate similar studies in other settings.

4.8.3. Dependability

The dependability of research are determined by how its’ findings over a given

period (Kumar, 2018). In addition, a dependable study should be repeatable under

different situations. The idea behind research dependability is centred upon the idea

that research studies must end with recommendations and such should be based on

data applicable or derivable in various conditions.

The Dependability measure was ensured through adherence to systematic

provisions of research principles. The researcher followed specific steps and created
an audit trail for review, analysis and even critical checking. The research project

was presented precisely and methodologically to form a dependable comprehensive

audit trail. An audit trail ensures that independent parties are able to analyze,

evaluate and interpret the research process and or outcome even for purposes of

repeating similar studies in other contexts (Kiger & Varpio, 2020).

4.8.4. Confirmability

The conformability measure presents the degree to which a study has neutrality.

Neutrality is the margins with which the study findings are a reflection of the

participants’ perceptions, opinions or experiences and not influenced by the

researcher’s biases (Kumar, 2020). Qualitative research can be easily immersed into

the researcher’s own identity, norms and belief systems especially data generation

and interpretation or analysis (Denzin & Salvo, 2020).

The researcher harnessed such techniques as member checking, reflexive

journaling and peer debriefing were utilized to ensure confirmability. The above tools

are very useful as they ensure different stakeholders can also confirm the sync

between the research process and research outcomes. Different parties include

participants themselves through member checking and other researchers through

peer debriefing (Kiger & Varpio, 2020). The researcher through kept residues from

raw data such as audio recordings field notes also did reflexive journalism.

Last the researcher also ensured that none of their personal biases and prejudices

impacted either the study procedures or its findings through determining that there

were none that could be of influence. In addition, methodological triangulation used

in the study in the form of semi structure interviews and focus group discussions also

ensured adherence to conformability.

4.9. RESEARCH ETHICS

Walker (2018) define ethics as the considerations relating to the social codes which

communicate integrity and consistent values to govern rules of engagement between

human beings and or other species. Scholars argue that ethical measures are

necessary in research since its nature sometimes may intrude in the lives of study

subjects (Crawford, 2020; Flick, 2022; Oliver, 2021). Adherence to acceptable

standards of professional research practices or ethics is thus a research norm in

itself. As a precursor, the researcher sought all the necessary approvals and

permissions to carry out the study from the relevant gate keepers. Ethical measures

adhered to included informed consent, anonymity and confidentiality, voluntary

participation, do no harm and honesty and integrity.

Heslop et al., (2018), posit that moral principles or ethics which guide research

activities are crucial as they aim to protect against aspects of uncertainty and risk

associated with research activities. Such protection is necessary for the sake of all

the stakeholders’ wellbeing, that is, participants, researchers and or any other

beneficiary either directly or indirectly is not affected by the research procedures or

activities.

Research procedures implore that data generation follows ethical procedures. Such

requirements include approval to carry out data generation by authorities from

different institutions such as the Kingdom of Eswatini’s Ministry of education and

respective primary schools. To adhere to the above-mentioned standards of

procedures, the researcher that the necessary ethical clearances were met. First by

engaging the University of South Africa’s ethical clearance board to where the study
was cleared and approved to commence. This was done because research

procedures indicate that no data collection activities shall be done without ethical

clearances from the University of South Africa’s ethics clearance board.

After accessing the clearance letter from the University of South Africa approved by

the University ethics clearance board, the researcher then sought gate keeper

approval from the authorities where data collection procedures were to be

implemented. The researcher submitted the letter from UNISA’s ethics clearance

board to the Education Ministry of the kingdom of Eswatini which then issued a

supporting or confirmation letter to the researcher to submit it to the authorities

where data generation was to be done. The confirmation letter was evidence used

as permission to conduct research.

4.9.1. Informed consent

According to Walker (2018), informed consent speaks to the dialogue a researcher

and a research participant which allows participants be informed to discern on either

agreeing or disagreeing to participate in a research study. Such a decision would be

based on full knowledge about the purpose of the study provided. Scholars have all

agreed of the need for participants appraised, aware and knowledgeable about

research studies before they can agree. According to Walker (2018), researchers

have an obligation of providing sufficient information to participants regarding the

study.

The researcher ensured participants were provided with informed consent through

provision of complete study details. In addition, the researcher availed an open

access for addressing any matters arising during the study. All the chosen
participants indicated and confirmed their willingness to partake in the study as

evidenced by the signed consent forms. After providing all the study details, the

researcher was convinced that participants understood what was expected of them.

After following the procedures as indicated, the researcher believed that the

participants participated freely and openly throughout the study.

4.9.2. Anonymity and confidentiality

Anonymity and confidentiality are two related concepts wherein anonymity is a

measure that seeks to the concealment of participants’ identity in the research

documents. According to Kumar (2018), the privacy of participants should be

honoured. Confidentiality on the other hand refers to how the anonymity of

participants shall be enabled (Walker, 2018). Confidentiality deals with participants

details or information should not be identified within the data set they have provided

(Oliver, 2021). Confidentiality enables participants to partake in research activities

with freedom of expression with the knowledge that their person shall not be

identified on the basis of information provided.

The researcher ensured that access to information related to participants was limited

to only the researcher, peers and academic research supervisors. Secondly, the

researcher avoided use of the real names of study participants and also the study

setting were not named. Pseudonyms were utilised when referencing participants

such that it was impossible to identify any of the participants.

Lastly, the study was also carried out in ways that ensured anonymity and

confidentiality through conducting data generating activities like the interviews in

unoccupied classrooms. Spaces where interaction was limited to participants gave

them freedom and guarantee of anonymity. What was said was left in the rooms

where they said it.

4.9.3. Voluntary participation

Voluntary participation is the non-compelling of participants to participate in a

research study (Walker, 2018). Explicit and clear demarcations about voluntary

participation must be provided. When in order, research study details must be

available and discernible to study participants.

The researcher thus provided participants with the full study detail including the

study rationale. Before the commencement of the study, the researcher also

appraised participants on the importance and how the research outcome would

benefit the system. A comprehensive explanation about the features of the study

was given to the participants.

The researcher also ensured that participants were provided with even details like

the level of the study and how their contributions were to be presented in the study.

Lastly, the researcher also pointed out how the research study would mature into a

research report as well as into a published journal.

4.9.4. Do no harm

Harm during research is whereby participants are exposed to situations that may

lead to psychological or physical harm as a result of research activities. There are

various reasons or situations which may lead to participant distress or strain. Such

should be avoided to ensure adherence to issues of doing no harm.

In the current study, the safety of all participants was maintained throughout the

study. The researcher ensured that non-sensitive questioning was used. Secondly,

procedures were carried out in environments which did not pose any physical,

psychological or mental harm. A session that activated the debriefing exercise was

done before, during and after the data generation procedures to ensure participants

were protected all the time and that concerns if any, are addressed promptly. In a nut

shell, the researcher did all they could to ensure participant welfare at all times

during the study (Crawford, 2020).

4.9.5. Honest and integrity

Research should be based on honest and integrity. Researchers ensure that the

methods, procedures, findings and publications from their study meet the standards

of honesty and integrity. Presenting truthful findings is of paramount importance in

research (Crawford, 2020). The researcher study was conducted in the spirit of

honesty and integrity. The researcher protected the participants from any form of

deceit. The researcher also ensured that only data generated from the study were

presented and used for conclusions and recommendations. Proper referencing was

also done to cater for any material borrowed from other scholars.

4.10. LIMITATIONS AND DELIMITATION OF THE STUDY

4.10.1. Limitations
The first study limitation the sample size which was determined by a limited number

of regions which also translated to also to few schools. The study focus was on one

selected school per region. As a way of trying to increase depth and breadth of data,

the researcher selected to gather data from more of the teachers in different grades

of the same school. This was done as a measure to control the possible low

numbers in the selected sample.

In addition, the study did not incorporate all the factors or variables with the potential

to influence the practice of educating learners with autism. For this limitation, the

researcher emphasized on amplifying the outcomes of research within the confines

of the chosen variables. Results were presented against the back drop of the

researched on variables.

Another limitation was that the study scope was not comprehensive enough. Other

identifiable participants who could have been useful in the study include parents and

other support staff. The study only dwelled on three roles significantly limits

generasability of research outcomes. The researcher determined to limit

recommendations and conclusions to the specific roles associated with the study.

It also possible that the variables under study are influenced by different governing

bodies outside of the Kingdom of Eswatini due to how the world has become a global

village, thus teachers may be trained in other regions or countries. Therefore the

level of training, its’ inadequacy of vice versa, may not reflect failures of Eswatini.

The researcher accepted that this limitation was beyond controllable and assumed

that standards of training are based on similar expectations hence wherever when is

trained, they have to meet certain expectations.

4.10.2. Delimitation

The study followed a systemic investigation into the teaching of learners with autism

in primary schools in the Kingdom of Eswatini. It focused on all the four regions of

the Kingdom of Eswatini with each of Hhohho, Manzini, Shiselweni, and Lubombo

regions providing one school for study. Participants for the study were policymakers,

primary school principals, and primary school teachers, specifically grades one

through three.

4.11. CHAPTER SUMMARY

The chapter highlighted the selected research methodology for the study. It

presented in detail the research approach used which was the qualitative approach.

The chapter also offered insight on the study’s research paradigm. The study used

the Interpretivist research paradigm. The chapter also covered the research design

and the research techniques which informed the data collection process. Methods

used for determining the population and the sampling strategy were covered in this

chapter too. A comprehensive or detailed explanation was provided for all the steps

taken during the study including data collection and analysis. The last sections or

issues to be covered in the chapter were trustworthiness and its measures as well as

ethical guidelines. Chapter five will dwell on presentation of data as well as data

analysis.
References

Braun, V., & Clarke, V. 2019. Reflecting on reflexive thematic analysis. Qualitative

research in sport, exercise and health, 11(4), 589–597.

https://doi.org/10.1080/2159676x.2019.1628806

Braun, V. & Clarke, V., 2021. Thematic Analysis: A Practical Guide. London: Sage.

Crawford, M. L. 2020. Foundations of research design. Thousand Oaks, CA: Sage

Publications.

Creswell, J. W. & Creswell, J. D. 2018. Research design: qualitative, quantitative,

and mixed methods approaches. (5th ed.). Los Angeles: Sage Publications.

Denzin, N. K. & Lincoln, Y.S. (Ed.). 2020. The sage handbook of qualitative

research. (5th ed.) London: Sage Publications.

Denzin, N. K., & Salvo, J. 2020. New directions in theorizing qualitative research:

Theory as resistance. Myers Education Press.

Elliott, V. 2018. Thinking about the coding process in qualitative data analysis.

Qualitative Report, 23(11), 2850–2861. https:// doi.org/10.46743/2160-

3715/2018.3560

Flick, U. 2022. The Sage handbook of qualitative research design. Sage.

Fryer, T. 2022. A critical realist approach to thematic analysis: Producing causal

explanations. Journal of Critical Realism, 21(4), 365–384.

https://doi.org/10.1080/14767430.2022.2076776

Gupta, A. & Gupta, N., 2022. Research methodology. SBPD Publications.

Hennink, M., & Kaiser, B. N. 2022. Sample sizes for saturation in qualitative

research: A systematic review of empirical tests. Social science & medicine, 292,

114523.

Heslop, C., Burns, S. & Lobo, R., 2018. Managing qualitative research as insider-

research in small rural communities. Rural and Remote Health.

Kiger, M. E., & Varpio, L. 2020. Thematic analysis of qualitative data: AMEE Guide

No. 131. Medical teacher, 42(8), 846-854.

Kumar, R., 2018. Research methodology: A step-by-step guide for beginners. Sage.

Oliver, P. 2021. Using qualitative methods to answer your research question (First).

Open University Press.

Walker, T., 2018. Consent and autonomy. In Muller, A. & Schaber, P. (eds.) The

Routledge Handbook of the Ethics of Consent, pp.131–139. Oxford: Routledge.