Bhandari, P. (2021, July 7) An Introduction to Correlational Research. Scribbr.
[Link]
(Bhandari, 2021)
User:
[File: [Link]]
what is the research paradigm for this articel and why, justify using examples
Assistant:
Research Report
Unmasking the Research Paradigm Behind “How to Formulate a Research Strategy?” – A
Pragmatist-Constructivist Lens
Executive Summary
The article "How to formulate a research strategy?" by Ashni Walia and Priya Chetty does not strictly
confine itself to a single rigid epistemological stance. Instead, it adopts a pragmatic-pluralist
paradigm with strong constructivist leanings. The authors present research paradigms—specifically
noting positivism, post-positivism, and constructivism—as flexible tools that should be selected
based on the specific research questions and objectives rather than dogmatic adherence to one
philosophy [1]. By emphasizing that a research strategy must be "suitable with respect to the
purpose of the study" and feasible based on available resources, the article champions pragmatism
[1]. This report dissects the article's underlying paradigm, justifies this classification using direct
examples from the text, and provides actionable recommendations for researchers utilizing this
framework.
Core Concepts: Ontology, Epistemology, Axiology
To understand the article's paradigm, it is essential to establish the foundational vocabulary of
research philosophy:
Ontology: The study of reality (e.g., is there a single objective truth, or multiple constructed
realities?).
Epistemology: The study of knowledge (e.g., how can we know what is real?).
Paradigm: A basic set of beliefs that guides the researcher's actions and methodology [1].
The article explicitly defines the research paradigm as the first step in formulation, stating it involves
a "basic set of beliefs that guides the researcher regarding the way of performing the research" [1].
Dissecting the Article’s Paradigm Narrative
The article presents a pragmatic approach, foregrounding research questions over doctrinal
allegiance. It treats paradigms as functional lenses rather than absolute truths.
Direct Textual Evidence of Pluralism
The authors explicitly list multiple paradigms as viable options for researchers. They state that
"There are various types of research paradigms, including positivism, post-positivism or
constructivism" [1]. Rather than arguing for the supremacy of one, they position them as the starting
point (Step 1) that subsequently dictates the research design (Step 2) and methods (Step 3) [1].
Furthermore, the article insists that an appropriate strategy must be selected based on practical
elements like "Research questions," "Amount of time available," and "Resources at the researcher’s
disposal" [1]. This prioritization of practical utility over philosophical purity is the hallmark of
pragmatism.
Implicit Signals of Constructivism
While pragmatic overall, the article frequently uses examples that highlight a constructivist
appreciation for human experience. For instance, when explaining interpretive research, the authors
note it "uses human experience in order to understand the phenomena" [1]. When providing an
example of how paradigms flow into methods, they specifically choose a constructivist path: "if a
researcher followed a constructivist paradigm using a qualitative research design, then the data
collection method can be interviews, observation or focus group discussions" [1].
Comparative Paradigm Table
The following table illustrates how the article treats different paradigms and strategies,
demonstrating its pluralistic acceptance of diverse methodological approaches.
| Paradigm / Strategy Type | Epistemic Claim in Article | Example Used in Article |
|---|---|---|
| Positivist / Quantitative | Focuses on numerical data and objective measurement of "when, where,
what and how often a specific phenomenon occurs" [1]. | Using questionnaires, polls, and surveys to
collect numerical data [1]. |
| Constructivist / Qualitative | Focuses on understanding underlying reasons, opinions, and human
experiences without numerical data [1]. | An in-depth analysis of perceptions of occupants in an old-
age home regarding the quality of service [1]. |
| Pragmatic / Action | Aims at finding immediate solutions to practical problems applied by agencies
or governments [1]. | Determining which strategy would work best to motivate physically challenged
students [1]. |
Key Takeaway: The article validates both objective (positivist) and subjective (constructivist)
approaches, ultimately recommending that the choice be driven by the practical aim of the research
(pragmatism).
Implications for Researchers
Selecting a paradigm after defining research questions improves methodological fit and ethical
compliance. The article's framework offers significant flexibility.
Opportunities for Mixed-Methods
By adopting the article's pragmatic stance, researchers are freed to use mixed-methods designs. If
the research objective requires both understanding human perception (constructivism) and
measuring frequency (positivism), the pragmatic paradigm allows for the integration of both
qualitative interviews and quantitative surveys.
Risks and Caveats
A potential risk of this highly pragmatic approach is the lack of deep ontological justification. The
article briefly mentions paradigms but does not delve into the philosophical friction between them.
Researchers using this guide must be careful to explicitly justify their chosen paradigm in their own
manuscripts to avoid appearing philosophically inconsistent during peer review.
Actionable Recommendations
To effectively apply the pragmatic-pluralist paradigm advocated by the article, researchers should
follow a structured alignment process.
| Step | Action | Application based on Article |
|---|---|---|
| 1. Define the Goal | Map research questions to practical constraints. | Assess the "Amount of time
available" and "Resources at the researcher’s disposal" before choosing a philosophy [1]. |
| 2. Select the Paradigm | Choose the belief system that best answers the specific question. | Opt for
constructivism if exploring opinions, or positivism if measuring frequencies [1]. |
| 3. Align the Design | Ensure the research design logically follows the paradigm. | If constructivist,
choose a qualitative design; if positivist, lean toward quantitative experiments [1]. |
| 4. Verify Ethics | Ensure the chosen strategy can be executed responsibly. | Guarantee participant
anonymity, especially in social science research [1]. |
Key Takeaway: Following this sequential alignment ensures that the research strategy remains
coherent from philosophical inception to practical data collection.
References
How to formulate a research strategy?. [Link]
strategy/
User:
what is the research paradigm for this articel and why, justify using examples
Assistant:
The research paradigm for this article is mainly interpretivism, specifically expressed through an
interpretive phenomenological paradigm.
The article is not positivist because it does not try to measure AI adoption through statistical testing,
numerical hypotheses, or objective cause-and-effect relationships. Instead, it seeks to understand
how employees experience, perceive, and make meaning of integrating artificial intelligence into
sustainability accounting and reporting.
Why the paradigm is interpretivist
An interpretivist research paradigm assumes that reality is socially constructed and that people
understand phenomena through their own experiences, meanings, values, and contexts. In this
article, the researcher is interested in the subjective views of employees working in companies listed
on the Namibia Stock Exchange.
The article itself states that the study used:
> “a qualitative research design using interpretive phenomenology to explore professional users'
perceptions of integrating AI and cloud computing in SAR.”
This is strong evidence of an interpretivist paradigm because the focus is on perceptions,
experiences, and meaning, not on numerical measurement.
Justification using examples from the article
The study explores lived experiences
The article says that phenomenology focuses on:
> “first-person experiences with an emphasis on extracting perceptions and meaning from an
individual's lived experiences.”
This shows that the researcher wanted to understand how employees personally experience AI
integration in sustainability accounting and reporting. For example, participants described how AI
changed their work by automating ESG data collection and reducing reporting time.
One participant said:
> “Using AI to automate ESG data collection has reduced our reporting timelines by nearly 40%.”
Although this example includes a percentage, it is used as part of a participant’s lived experience,
not as statistical evidence from a large quantitative survey.
The research questions focus on perceptions and challenges
The first research question asks:
> “What are the perceptions of employees from listed companies on the Namibia Stock Exchange
regarding the impact of AI on the efficiency and accuracy of SAR?”
The word perceptions is important. It shows that the study is concerned with how employees
understand and interpret the impact of AI. This fits interpretivism because the researcher is studying
meaning from the participants’ point of view.
The second research question also asks about organisational challenges in integrating AI. These
challenges are explored through employees’ accounts, such as high costs, infrastructure problems,
lack of regulatory clarity, and ethical concerns.
The study uses qualitative interviews
The article collected data through online interviews with 13 participants, including sustainability
managers, ESG analysts, accountants, and IT specialists.
This supports an interpretivist paradigm because interviews allow participants to explain their
experiences in depth. The researcher was not simply collecting fixed-choice answers or numerical
data. Instead, participants gave detailed responses such as:
> “Automation allows us to focus more on strategy and less on manual data entry.”
and:
> “The lack of clear guidelines on AI usage in sustainability reporting makes it difficult to ensure
compliance.”
These examples show that the data came from participants’ interpretations of their workplace
realities.
The sample size is small and purposive
The study used purposive and snowball sampling and included 13 participants. This is typical of
interpretivist research because the aim is not to generalise statistically to a large population. Instead,
the aim is to gain a deep understanding of selected participants’ experiences.
The article even explains that small samples are suitable for Interpretive Phenomenological Analysis
because the goal is to understand the meaning of participants’ lived experiences.
The data analysis is thematic and interpretive
The article used thematic content analysis to identify patterns, themes, and meanings in the
interview data. The findings were organised into themes such as:
| Strand | Themes |
|---|---|
| Benefits | Efficiency and automation; improved data accuracy; scalability and accessibility |
| Challenges | Implementation challenges; infrastructure inadequacy; regulatory obscurity |
| Considerations | Ethical considerations; regulatory considerations; training and capacitation |
This supports an interpretivist paradigm because the researcher interpreted participants’ words and
grouped them into meaningful themes.
For example, participant comments about high costs and system incompatibility were interpreted
under the theme implementation challenges. Comments about data privacy and algorithmic bias
were interpreted under ethical considerations.
Ontology, epistemology, and methodology of the article
| Element | Position in this article | Justification |
|---|---|---|
| Ontology | Relativist / socially constructed reality | The article assumes that employees may
experience AI integration differently depending on their industry, organisational size, role, and
digital maturity. |
| Epistemology | Subjectivist / interpretivist | Knowledge is gained through participants’
perceptions, experiences, and meanings. |
| Methodology | Qualitative interpretive phenomenology | The study uses interviews, purposive
sampling, and thematic analysis to understand lived experiences. |
| Methods | Online interviews and thematic content analysis | The researcher analyses participants’
narratives rather than numerical datasets. |
Why it is not mainly positivist
A positivist study would usually use large samples, surveys, hypotheses, statistical models, and
objective measurement. This article does not do that. It does not test a hypothesis such as “AI
adoption significantly improves sustainability reporting accuracy.” Instead, it asks employees how
they perceive AI’s impact.
The article does mention ideas such as efficiency, accuracy, and company size, but these are
explored mainly through participants’ accounts and thematic interpretation rather than statistical
testing.
Final answer
The research paradigm of the article is interpretivism, specifically an interpretive phenomenological
paradigm. This is because the study uses qualitative interviews to explore employees’ subjective
perceptions and lived experiences of integrating AI into sustainability accounting and reporting. The
researcher interprets participants’ views through themes such as efficiency, data accuracy,
implementation challenges, infrastructure limitations, ethical concerns, and training needs. The
article’s focus on meaning, context, experience, and interpretation clearly places it within the
interpretivist paradigm.
User:
[File: [Link]]
what is the paradigm of this article
Assistant:
The research paradigm of this article is positivism, more specifically a post-positivist quantitative
paradigm.
This is because the article studies the relationship between variables using a structured model,
numerical data, hypotheses, statistical testing, and survey responses. The researchers are not mainly
trying to understand students’ lived experiences in depth; instead, they are testing whether
measurable factors such as perceived usefulness, trust in AI, attitude toward AI, AI literacy, and
employability are statistically related.
Why the paradigm is positivist
A positivist paradigm assumes that social reality can be measured objectively through observable
variables. Researchers usually use quantitative methods, hypotheses, surveys, and statistical analysis
to test relationships between variables.
This article fits that paradigm because it uses:
| Feature | Evidence from the article |
|---|---|
| Quantitative method | The article states that “a quantitative method with a survey was
employed.” |
| Structured questionnaire | Students completed a questionnaire using a seven-point Likert scale. |
| Hypotheses | The study tested eight hypotheses, such as “AI literacy positively affects internal
employability.” |
| Statistical analysis | The researchers used Partial Least Squares Structural Equation Modelling, or
PLS-SEM. |
| Measurement scales | The study used established scales for AI literacy, trust in AI, attitude toward
AI, and employability. |
| Numerical results | The findings are reported using coefficients, T-values, P-values, R-squared
values, VIF, and effect sizes. |
These features show that the researchers aimed to measure, test, and explain relationships between
variables.
Examples that justify the paradigm
The article tests hypotheses
The study developed and tested eight hypotheses, including:
> “Perceived usefulness positively affects the acceptance of using AI.”
> “AI literacy positively affects internal employability.”
> “AI literacy positively affects external employability.”
This is typical of a positivist approach because the researchers begin with theory, formulate
hypotheses, collect data, and then test whether the hypotheses are supported or rejected.
For example, the article found that H1, H3, H5, H6, H7, and H8 were supported, while H2 and H4
were not supported. This shows a deductive, theory-testing approach.
The study uses measurable variables
The article breaks the topic into measurable constructs, such as:
• Perceived usefulness
• Ease of communication
• Trust in AI
• Fear of AI
• Acceptance of AI
• Attitude toward AI
• AI literacy
• Internal employability
• External employability
Each construct is measured using questionnaire items. For example, AI literacy is measured through
dimensions such as awareness, usage, evaluation, and ethics.
This supports a positivist paradigm because the researchers treat abstract ideas as variables that can
be measured numerically.
The researchers used a survey and Likert scale
The article says:
> “All measurement scales in the above research model were from established constructs using a
seven-point Likert scale.”
A Likert scale converts students’ views into numerical data, for example from 1 = strongly disagree to
7 = strongly agree. This is a quantitative measurement technique associated with positivist research.
For instance, students responded to statements such as:
> “I can skillfully use AI applications or products to help me with my daily work.”
> “I trust that my privacy is protected when using AI.”
> “I am generally confident of success in job interviews and selection events.”
These responses were then analysed statistically.
The study uses statistical modelling
The article used:
> “Partial least square structured equation modeling.”
This is strong evidence of a positivist or post-positivist paradigm. Structural equation modelling is
used to test causal or predictive relationships between variables.
For example, the researchers tested whether:
• Perceived usefulness affects acceptance of AI
• Trust in AI affects acceptance of AI
• Acceptance of AI affects attitude toward AI
• Attitude toward AI affects AI literacy
• AI literacy affects internal and external employability
The results are presented through statistical indicators such as:
| Hypothesis | Result |
|---|---|
| H1: Perceived usefulness affects acceptance of AI | Supported |
| H2: Ease of communication affects acceptance of AI | Not supported |
| H3: Trust in AI affects acceptance of AI | Supported |
| H4: Fear of AI affects attitude toward AI | Not supported |
| H5: Acceptance of AI affects attitude toward AI | Supported |
| H6: Attitude toward AI affects AI literacy | Supported |
| H7: AI literacy affects internal employability | Supported |
| H8: AI literacy affects external employability | Supported |
This confirms that the study is concerned with testing relationships rather than interpreting
subjective meanings in depth.
Ontology, epistemology, and methodology
| Element | Position in this article | Justification |
|---|---|---|
| Ontology | Objective / realist | The article assumes that constructs such as AI literacy and
employability can be measured as real variables. |
| Epistemology | Objectivist / post-positivist | Knowledge is produced through measurement, survey
data, and statistical testing. |
| Methodology | Quantitative | The study uses a questionnaire and numerical analysis. |
| Method | Survey and PLS-SEM | Data were collected from 208 students and analysed statistically.
|
| Reasoning approach | Deductive | The researchers developed hypotheses from existing theory and
tested them. |
Why it is not interpretivist
This article is not mainly interpretivist because it does not use interviews, focus groups, or narrative
analysis to explore students’ meanings and lived experiences. It does not ask, “How do students
experience AI literacy in relation to employability?” Instead, it asks whether there is a measurable
relationship between AI literacy and employability.
For example, the study does not deeply analyse individual students’ personal stories about AI.
Rather, it converts their responses into numerical data and tests a research model.
Final answer
The paradigm of this article is positivist, or more precisely post-positivist quantitative. The article
uses a survey, Likert-scale measurements, hypotheses, and PLS-SEM statistical analysis to test
relationships between AI literacy and perceived employability. It assumes that concepts such as AI
literacy, attitude toward AI, trust in AI, and employability can be measured objectively and analysed
statistically.
User:
what is the research methodology
Assistant:
The research methodology of this article is quantitative research methodology, using a cross-
sectional survey design and Partial Least Squares Structural Equation Modelling (PLS-SEM) for data
analysis.
In simple terms, the researchers collected numerical data from university students through a
questionnaire and used statistical techniques to test the relationships between AI literacy and
perceived employability.
Research methodology used in the article
| Methodology element | What the article used |
|---|---|
| Research approach | Quantitative |
| Research design | Cross-sectional survey |
| Sampling method | Purposive sampling |
| Data collection method | Structured questionnaire |
| Measurement tool | Seven-point Likert scale |
| Sample size | 208 valid student responses |
| Data analysis method | Partial Least Squares Structural Equation Modelling |
| Software | SmartPLS 4.0 |
| Reasoning approach | Deductive, hypothesis-testing |
Why it is quantitative methodology
The article clearly states:
> “A quantitative method with a survey was employed in a higher educational institution in Hong
Kong.”
This shows that the study used a quantitative methodology because it collected measurable data
from respondents and analysed the data statistically.
The study did not use interviews, focus groups, or observations. Instead, it used a structured
questionnaire with numerical response options.
Cross-sectional survey design
The study is cross-sectional because data were collected at one point in time from students in 2024.
The article also confirms this in the conclusion:
> “This is a cross-sectional study with the students as respondents.”
This means the researchers did not follow the students over a long period. They measured students’
perceptions of AI literacy and employability at a single point in time.
Sampling method
The researchers used purposive sampling.
The article states:
> “The purposive sampling method was used.”
This means participants were selected because they fitted the purpose of the study: they were full-
time university students in a higher educational institution in Hong Kong.
The sample included 208 valid responses from students. Around 250 students were invited, giving a
response rate of 83.2%.
Data collection method
Data were collected using a structured questionnaire.
The questionnaire measured constructs such as:
• Perceived usefulness of AI
• Ease of communication with AI
• Trust in AI
• Fear of AI
• Acceptance of AI
• Attitude toward AI
• AI literacy
• Internal employability
• External employability
The article states that the measurement scales were based on established constructs and used a
seven-point Likert scale, where respondents rated their agreement from 1 = strongly disagree to 7 =
strongly agree.
Data analysis method
The researchers used Partial Least Squares Structural Equation Modelling, also called PLS-SEM.
This method was used to test the proposed research model and hypotheses. For example, the study
tested whether:
• Perceived usefulness affects acceptance of AI.
• Trust in AI affects acceptance of AI.
• Acceptance of AI affects attitude toward AI.
• Attitude toward AI affects AI literacy.
• AI literacy affects internal and external employability.
The article also states that SmartPLS 4.0 software was used for the analysis.
Hypothesis-testing approach
The methodology is also deductive because the researchers developed hypotheses from existing
theory, especially the Technology Acceptance Model, and then tested those hypotheses using survey
data.
For example, the article tested hypotheses such as:
> “Perceived usefulness positively affects the acceptance of using AI.”
and:
> “AI literacy positively affects internal employability.”
This shows that the study moved from theory to data, which is typical of a deductive quantitative
methodology.
Final answer
The research methodology of the article is a quantitative, cross-sectional survey methodology. The
researchers used purposive sampling to collect data from 208 university students through a
structured questionnaire using a seven-point Likert scale. The data were analysed using PLS-SEM
with SmartPLS 4.0 to test hypotheses about the relationship between AI literacy and students’
perceived employability.
User:
what is the methodology for the first article
Assistant:
The methodology for the first article, “Integrating Artificial Intelligence in Sustainability Accounting
and Reporting: An Interpretive Phenomenological Analysis”*, is a qualitative research methodology
using an interpretive phenomenological approach.
The study aimed to understand employees’ perceptions, experiences, and meanings regarding the
integration of artificial intelligence into sustainability accounting and reporting among companies
listed on the Namibia Stock Exchange.
Methodology used in the first article
| Methodology element | What the article used |
|---|---|
| Research approach | Qualitative |
| Research design | Interpretive phenomenology |
| Sampling method | Purposive and snowball sampling |
| Participants | 13 employees from companies listed on the Namibia Stock Exchange |
| Participant roles | Sustainability managers, ESG analysts, accountants, and IT specialists |
| Data collection method | Online semi-structured interviews |
| Interview duration | 45 minutes to 1 hour 20 minutes |
| Data recording | Interviews were recorded and transcribed using Zoom audio transcription |
| Data analysis method | Interpretive Phenomenological Analysis and thematic content analysis |
| Main themes found | Benefits, challenges, and considerations of AI integration |
Why it is qualitative
The article used a qualitative methodology because it did not collect numerical data for statistical
testing. Instead, it collected detailed interview responses from participants.
The article states that the study used:
> “a qualitative research design using interpretive phenomenology.”
This means the researcher wanted to explore how participants understood and experienced AI
integration in sustainability accounting and reporting.
Research design: interpretive phenomenology
The study used interpretive phenomenology, which focuses on people’s lived experiences and how
they make meaning of those experiences.
In this article, the researcher explored how employees experienced AI in relation to:
• Efficiency and automation
• Data accuracy and integrity
• Scalability and accessibility
• Implementation challenges
• Infrastructure limitations
• Regulatory uncertainty
• Ethical concerns
• Training and capacity-building needs
For example, one participant said:
> “Using AI to automate ESG data collection has reduced our reporting timelines by nearly 40%.”
This type of response shows that the study was interested in participants’ practical experiences and
perceptions.
Sampling method
The article used a combination of purposive sampling and snowball sampling.
Purposive sampling means participants were selected because they had relevant knowledge or
experience of sustainability accounting, ESG reporting, accounting, IT, or AI systems.
Snowball sampling means some participants may have helped identify other suitable participants.
The final sample consisted of 13 participants, which the article explains is suitable for an interpretive
phenomenological study because the aim is depth of understanding rather than large-scale
generalisation.
Data collection method
Data were collected through online interviews.
The researcher used a structured interview guide to ask participants about:
• The benefits of AI in sustainability accounting and reporting
• The challenges of AI adoption
• The potential of AI to improve sustainability outcomes
• Ethical and regulatory concerns
• Training needs
The interviews lasted between 45 minutes and 1 hour 20 minutes. They were recorded, transcribed,
edited to remove identifying details, and participants were given pseudonyms.
Data analysis method
The article used Interpretive Phenomenological Analysis, supported by thematic content analysis.
The researcher followed Smith et al.’s framework by:
Listening to recordings and reading transcripts repeatedly.
Making initial notes in a research diary.
Conducting close line-by-line analysis.
Colour-coding participants’ responses.
Identifying patterns and themes.
Grouping findings into broader themes.
The analysis produced three main strands:
| Strand | Examples of themes |
|---|---|
| Benefits | Efficiency and automation; improved data accuracy; scalability and accessibility |
| Challenges | Implementation difficulties; infrastructure inadequacy; regulatory obscurity |
| Considerations | Ethical issues; regulatory considerations; training and capacitation |
Final answer
The methodology of the first article is a qualitative interpretive phenomenological methodology. The
researcher used purposive and snowball sampling to select 13 employees from companies listed on
the Namibia Stock Exchange. Data were collected through online interviews and analysed using
Interpretive Phenomenological Analysis and thematic content analysis. This methodology was
appropriate because the study aimed to understand employees’ lived experiences, perceptions, and
meanings regarding AI integration in sustainability accounting and reporting.