A mass media management dissertation is not simply a report about an interesting media issue. It is a systematic, defensible investigation that connects a research problem with assumptions about knowledge, evidence, interpretation and action. Whether the topic concerns audience analytics, newsroom culture, platform regulation, digital subscriptions, advertising effectiveness or media ownership, the quality of the study depends heavily on methodological choices made before data collection begins.

Research philosophy concerns the assumptions that guide the production and evaluation of knowledge. Research design provides the overall architecture of a study. Methodology explains the logic linking the research question, data, methods and conclusions. Methods are the practical techniques used to collect and analyse evidence. These concepts are related but not interchangeable.

Research problem → research questions → philosophical position → research design → sampling and methods → analysis → interpretation and recommendations.

Poor alignment at any stage can weaken a dissertation. For example, a study claiming to understand journalists’ lived experiences cannot be adequately addressed through audience ratings alone; conversely, a study testing whether a pricing change increases subscriptions requires systematic numerical evidence rather than only personal accounts.

Learning Objectives

From Research Problem to Methodological Choice

A research problem identifies a significant gap, uncertainty, contradiction or practical difficulty. A research question converts that problem into an answerable inquiry. Methodological choice should follow the question rather than personal preference or the popularity of a technique.

Research purpose	Typical question	Suitable emphasis
Exploratory	What is happening, and how do participants understand it?	Interviews, focus groups, observation, document analysis
Descriptive	What are the characteristics, frequency or distribution of a phenomenon?	Surveys, analytics, content coding, descriptive statistics
Explanatory	Why or under what conditions does something occur?	Experiments, statistical modelling, comparative analysis, case studies
Comparative	How do organisations, audiences or contexts differ?	Structured comparison using qualitative, quantitative or mixed evidence
Longitudinal	How does a phenomenon change over time?	Repeated surveys, panel data, archival research, time-series or sequential qualitative studies

These purposes may be combined. A dissertation might first explore why podcast listeners cancel subscriptions, describe cancellation patterns across demographic groups, and then explain whether perceived value predicts cancellation.

Philosophical Foundations

Ontology and epistemology

Ontology concerns what reality is like. A realist may argue that audience behaviour and organisational structures exist independently of researchers, although they may be difficult to observe perfectly. A constructionist may emphasise that meanings such as “quality journalism,” “trust” or “professional success” are produced through social interaction and language.

Epistemology concerns how knowledge can be acquired and justified. Some researchers privilege observable measurement and systematic testing. Others regard interpretation, context and participant meaning as essential to understanding social life. Neither position automatically makes a study rigorous; rigour depends on consistency between assumptions, procedures and claims.

Methodology is the theory and rationale of inquiry. A methodology explains why particular methods are appropriate. A questionnaire, for example, can be used within a positivist survey, a pragmatic evaluation or a mixed-method design. The instrument itself does not determine the philosophy.

Positivism and post-positivism

Positivism treats observable phenomena, measurement and systematic comparison as central sources of knowledge. In media management, a positivist-oriented study might test whether exposure to personalised news recommendations is associated with subscription intention. It may use predefined variables, probability sampling and statistical tests.

Post-positivist research retains an interest in empirical evidence but recognises that measurement is imperfect and that complete objectivity is difficult. Researchers therefore use transparent procedures, error estimation, alternative explanations and cautious claims.

Strengths include clarity, replicability, efficient comparison and suitability for estimating patterns in larger populations. Limitations include the risk of reducing complex experiences to variables, overlooking context and treating operational definitions as if they were the phenomena themselves.

Interpretivism and constructionism

Interpretivism holds that social reality is experienced and understood through meanings. A study of newsroom transformation, for instance, may investigate how editors and reporters interpret “digital-first” work, professional autonomy and algorithmic performance metrics. Interviews, observation and textual analysis are common methods.

Interpretivist research generally values depth, context, reflexivity and the situated accounts of participants. Its quality is assessed through credibility, transparency, thick description, interpretive coherence and attention to disconfirming evidence rather than statistical generalisation alone.

Its limitation is not that it is “subjective” in a simplistic sense. Rather, interpretation requires disciplined documentation and acknowledgment of the researcher’s role. Findings may be transferable to comparable contexts but are not automatically representative of all media workers or audiences.

Critical approaches

Critical research examines power, inequality, ideology and the conditions under which some voices become more influential than others. It may analyse ownership concentration, gendered labour, platform dependency, racial representation, commercial pressures or unequal access to media technologies.

Critical studies ask not only “What happens?” but also “Who benefits?”, “Who is excluded?”, “How is this arrangement maintained?” and “What alternatives are possible?” They may combine discourse analysis, political economy, interviews, historical research and statistical evidence.

A critical approach does not justify selective evidence. Normative commitments should be explicit, sources should be evaluated carefully, and competing interpretations should be considered. Its contribution lies in connecting media practices with institutional structures and possibilities for change.

Pragmatism

Pragmatism begins with the research problem and asks which forms of evidence will best answer it or support a practical decision. A pragmatic researcher may combine subscription data, a customer survey and interviews with former subscribers because each illuminates a different aspect of the problem.

Pragmatism is especially useful in applied mass media management, where managers need both measurable indicators and contextual understanding. It can support mixed methods, evaluation research and action-oriented recommendations. However, “using whatever works” is not sufficient justification. The researcher must explain how different forms of evidence are integrated and how philosophical tensions are managed.

Comparative overview

Approach	View of knowledge	Typical question	Common methods	Main quality concern
Positivist/post-positivist	Knowledge is developed through systematic observation and testing	What relationship or pattern exists?	Surveys, experiments, analytics	Reliability, validity, bias, error
Interpretivist	Knowledge emerges through contextual understanding of meaning	How do people experience and interpret this?	Interviews, observation, thematic analysis	Credibility, reflexivity, coherence
Critical	Knowledge should reveal power and support informed change	Who benefits, who is marginalised, and why?	Political economy, discourse, participatory inquiry	Transparency, positionality, explanatory depth
Pragmatic	Knowledge is judged partly by usefulness for the problem	Which evidence best informs action?	Mixed methods, evaluation, case study	Integration, fitness for purpose, practical value

Research Designs and Methodological Choices

Quantitative research

Quantitative research represents observations numerically and examines distributions, differences, relationships or effects. In media management it may measure reach, engagement, advertising response, employee satisfaction, trust or willingness to pay.

A variable is an attribute that can take different values. Operationalisation specifies how an abstract concept will be measured. “Audience loyalty,” for example, might be operationalised through renewal intention, frequency of use and length of subscription. A single indicator may be convenient but can provide a narrow representation of a complex construct.

Descriptive statistics summarise data. For observations x₁, x₂, ..., x_n, the arithmetic mean is:

Here, x̄ is the sample mean, xᵢ is observation i, and n is the number of observations. The mean is appropriate for numerical data when its use is substantively meaningful; extreme values can distort it. Frequencies, percentages and the median may be preferable for skewed or categorical data.

Quantitative designs can estimate prevalence, compare groups and test hypotheses. They require attention to sampling, non-response, measurement validity, confounding variables and the difference between association and causation. A correlation between social media engagement and subscription revenue does not by itself prove that engagement caused revenue growth.

Qualitative research

Qualitative research investigates meanings, processes, experiences and contexts using non-numerical or text-rich data. Common methods include:

Qualitative sampling is commonly purposive: participants are selected because they possess relevant experience. The sample may develop through theoretical or snowball sampling, provided recruitment and inclusion decisions are documented. Data analysis often involves coding, category development, comparison across cases and interpretation in relation to theory.

The researcher should distinguish participants’ statements from the researcher’s interpretation, preserve an audit trail and actively seek evidence that challenges initial themes. Saturation can be a useful practical consideration, but it should not be treated as a mechanical number. The adequacy of a sample depends on the research question, diversity of relevant perspectives and richness of the data.

Mixed-method research

Mixed-method research intentionally integrates quantitative and qualitative evidence in one study. It is more than placing a survey and interviews side by side. Integration may occur when one phase informs another, when findings are compared, or when data are transformed and combined in a joint interpretation.

Design	Sequence and purpose	Example
Explanatory sequential	Quantitative first, qualitative follow-up	Survey identifies low trust among young users; interviews explain why
Exploratory sequential	Qualitative first, quantitative follow-up	Interviews generate dimensions of platform fatigue; survey estimates prevalence
Convergent	Qualitative and quantitative data collected in parallel	Survey patterns and newsroom interviews address the same management issue
Embedded	One method supports a primary design	Interviews embedded in an evaluation experiment

Mixed methods is valuable when breadth and depth are both required, when findings need triangulation, or when an intervention must be assessed and understood. It is demanding: the researcher needs competence in both traditions, adequate resources and a clear integration strategy. Contradictory results should be investigated, not hidden.

Major Study Purposes

Exploratory studies

Exploratory research is appropriate when concepts, mechanisms or relevant variables are not yet clear. It is often used for emerging technologies, new audience practices and poorly understood organisational problems. Exploratory findings usually generate propositions or a conceptual framework rather than definitive population estimates.

Descriptive studies

Descriptive research documents what exists: audience composition, content characteristics, workflow patterns or levels of trust. Good description requires precise definitions, appropriate sampling and transparent measurement. Description is not intellectually inferior to explanation; reliable description often makes later explanation possible.

Explanatory studies

Explanatory research examines reasons, mechanisms or causal relationships. Strong causal claims require more than a before-and-after comparison. Researchers should consider temporal order, plausible alternative explanations, selection effects and measurement quality. Experiments can strengthen causal inference through random assignment, while observational studies may use statistical controls and carefully reasoned designs, though controls cannot eliminate every source of bias.

Comparative studies

Comparative research analyses similarities and differences between media organisations, countries, platforms, audience segments or time periods. Comparisons require equivalence: concepts and measures should have sufficiently similar meanings across cases. A comparison of “engagement” is weak if one organisation defines it as comments and another defines it as page views.

Longitudinal studies

Longitudinal research examines change. A trend study samples different individuals from the same population at multiple times. A panel study follows the same individuals. A cohort study follows people sharing a defining experience or period. Repeated cross-sectional designs can show population-level change, whereas panel studies can show individual-level change but face attrition and repeated-measurement effects.

Time must be theoretically meaningful. A short observation window may capture a campaign but not durable audience behaviour. Researchers should distinguish genuine change from changes in measurement, sample composition or external events.

Sampling and Data Quality

A population is the full group to which a study may reasonably refer; a sample is the observed subset. Probability sampling gives units a known, non-zero chance of selection and supports statistical generalisation under appropriate assumptions. Non-probability sampling can be suitable for qualitative inquiry or hard-to-reach groups but requires cautious claims.

A sampling frame is the operational list or procedure used to identify eligible units. Coverage error occurs when the frame excludes relevant members. Non-response bias occurs when participation differs systematically from non-participation. In digital research, platform users who volunteer for a survey may differ substantially from less engaged users.

Validity concerns whether a measure or design supports the intended inference. Reliability concerns consistency of measurement. A measure can be reliable but invalid: a broken scale may give the same incorrect weight repeatedly. In qualitative research, analogous concerns include credibility, dependability, confirmability and transferability.

Analysis and Interpretation

Quantitative analysis should begin with data cleaning, coding decisions and examination of missing values. Researchers should report appropriate descriptive statistics before presenting inferential results. Statistical significance is not the same as practical importance, and a statistically detectable association may be too small to matter managerially.

Qualitative analysis should explain how raw material became themes or interpretations. A strong report includes representative evidence, attention to variation and a clear link between data and claims. Coding may be deductive, based on an existing framework, inductive, developed from the data, or a combination of both.

Triangulation can involve different data sources, researchers, theories or methods. It does not mean that all sources must agree. Divergence may reveal different perspectives, measurement limitations or a genuinely complex phenomenon.

Ethics, Governance and Responsible Research

Media research may involve employees, audiences, minors, vulnerable communities, confidential business information or politically sensitive content. Ethical practice includes informed consent, voluntary participation, the right to withdraw, secure data storage and minimisation of harm. Participants should understand the study’s purpose and how quotations or identifiable data will be used.

Researchers must protect confidentiality while recognising that anonymisation may be difficult in small organisations or public online communities. Public availability of a post does not automatically remove ethical responsibility. Terms of service, institutional review requirements, copyright and data-protection law may also apply.

AI tools can assist with transcription, translation, coding support and exploratory analysis, but outputs require human verification. Researchers should disclose substantial AI assistance, avoid uploading confidential data to unapproved systems, check for fabricated citations and remain accountable for interpretation. Automated sentiment or content classification may reproduce linguistic, cultural or political biases.

Applied Case: Designing a Study of Digital Subscription Churn

Background

This illustrative case concerns MetroView, a fictional regional news organisation experiencing an increase in digital subscription cancellations. Managers know the monthly cancellation rate but do not know whether the main causes are price, content value, technical friction, competing services or declining trust.

Problem and research questions

Design choice

A pragmatic explanatory-sequential mixed-method design is appropriate. First, the organisation analyses anonymised account records and administers a survey to current and recently cancelled subscribers. The quantitative phase describes cancellation patterns and identifies associations. Second, purposively selected participants are interviewed to explain motivations behind those patterns.

The quantitative sample should not simply consist of the most accessible social-media followers. The sampling process should include active subscribers and recent cancellations, with response rates and missing data reported. Survey items should distinguish price dissatisfaction from perceived lack of value rather than assuming they are identical.

The interview guide might explore the subscription journey, use of content, payment experience, competing media, trust and attempted retention. Analysis would involve thematic coding, comparison of current and former subscribers, and integration with quantitative findings.

Possible outcome

Suppose survey results show that cancellation is more common among subscribers with low weekly use, but interviews reveal that low use often follows repeated login failures. The managerial implication is not merely to publish more content. It may involve improving authentication, communicating product value and testing retention offers. The evidence should not claim that technical problems alone caused every cancellation unless the design supports that conclusion.

Learning points

The case illustrates alignment between problem, design and methods. It also demonstrates why mixed methods can be valuable: numerical patterns identify the scale and distribution of the issue, while interviews illuminate mechanisms and meanings. Ethical governance is essential because account records and cancellation histories are sensitive personal data.

Best Practices for a Dissertation Methodology

Challenges and Limitations

No design eliminates all uncertainty. Quantitative studies may suffer from weak measures, non-response and overreliance on platform metrics. Qualitative studies may face access problems, selective participation and interpretive disagreement. Mixed-method studies may become too broad for available time or resources. Comparative research can impose inappropriate categories across contexts, while longitudinal research is vulnerable to attrition and historical changes.

A rigorous dissertation does not conceal these limitations. It evaluates their likely effect on the findings and identifies what future research should do differently. Methodological modesty strengthens credibility: conclusions should extend only as far as the evidence permits.

Chapter Summary

Research philosophy provides assumptions about reality, knowledge and the role of the researcher. Positivist and post-positivist approaches emphasise systematic measurement and testing; interpretivism emphasises meaning and context; critical research examines power and inequality; pragmatism selects and integrates evidence in relation to practical problems.

Quantitative designs are useful for measurement, prevalence, comparison and hypothesis testing. Qualitative designs provide depth, context and insight into experience and meaning. Mixed-method designs integrate both, but require explicit sequencing and interpretation. Exploratory, descriptive, explanatory, comparative and longitudinal purposes represent different forms of inquiry and may be combined when logically justified.

A sound dissertation demonstrates alignment among research questions, philosophy, design, sampling, methods, analysis and claims. It also addresses validity or credibility, ethics, privacy, limitations and the responsible use of technology. In mass media management, methodological competence enables students to transform organisational problems into evidence-based decisions.

Key Terms

Positivism: An approach emphasising systematic observation, measurement and testing.

Interpretivism: An approach seeking contextual understanding of meanings and experiences.

Critical research: Inquiry focused on power, inequality, ideology and social change.

Pragmatism: An approach that selects evidence according to the research problem and intended use.

Operationalisation: Defining how an abstract concept will be measured or observed.

Reflexivity: Critical examination of the researcher’s influence and assumptions.

Descriptive research: Systematic account of characteristics, frequencies or distributions.

Triangulation: Use of multiple sources, methods, researchers or theories to deepen inquiry.

Mixed-method research: Deliberate integration of qualitative and quantitative evidence.

Sampling frame: The operational list or process from which a sample is selected.

Review Questions

Short Answer Questions

Long Answer Questions

Analytical / Application Questions

Multiple Choice Questions with Answers

Question 1

Which philosophical approach is most directly concerned with interpreting participants’ meanings and experiences?

Question 2

A study first surveys cancelled subscribers and then interviews selected participants to explain the survey patterns. What design is this?

Question 3

Which research purpose is most appropriate when little is known about a new media practice?

Question 4

Correct Answer: B. Defining how an abstract concept will be measured or observed

Question 5

Question 6

A researcher follows the same subscribers for three years to study changing renewal behaviour. This is a:

Question 7

Correct Answer: C. It indicates association but does not by itself prove causation

Question 8

Correct Answer: B. Selecting and integrating evidence according to the research problem

Question 9

A measure gives consistent results but does not actually capture audience trust. It is best described as:

Question 10

Which is an important ethical practice when using confidential subscriber records?

Correct Answer: C. Minimising harm and securing the data with appropriate governance

Chapter 6 Research Philosophy, Design and Methodological Choice

Research Philosophy, Design and Methodological Choice

Introduction