Module Overview

Twelve units, two summative artefacts, and a thread of misinformation research

This module enhanced my skills in designing and evaluating research in computing. Through twelve units, I explored the scientific method, research strategies, statistics, validity, and project risk, culminating in a literature review and research proposal on misinformation detection in social media.

Skills developed. The module outlines skills including time management, commercial awareness, critical thinking and analysis, decision-making, problem-solving, initiative, entrepreneurial thinking, and various forms of communication and literacy. Each skill is linked to a self-assessment in the Skills Matrix below, with subsequent artefacts and reflections providing supporting evidence.

Module Learning Outcomes

The four outcomes assessed by this module

⚖️

LO 1. Evaluate the professional, legal, social, cultural, and ethical issues impacting computing professionals.

📚

LO 2. Evaluate academic investigation principles and apply them to a computing research topic.

🔍

LO 3. Critically evaluate existing literature, research design, methodology, and data analysis for the chosen topic.

📝

LO 4. Critically produce and evaluate a research proposal for the topic.

Key Artefacts

Summative submissions, mandatory worksheets, and the reflective piece

📖

Literature Review — Detecting Misinformation on Social Media

UNIT 7 62% Merit LO 2 LO 3

A literature review on automated misinformation detection, including content, propagation, source-credibility, and hybrid retrieval methods, relating to the EU AI Act and UK Online Safety Act. The search adhered to PRISMA 2020 standards across ACM DL, IEEE Xplore, ACL Anthology, and Scopus.

📁 Evidence: Literature Review.odt

📊

Summary Measures Worksheet

UNIT 7 LO 3

Mandatory worksheet applying descriptive statistics — mean, standard deviation, median, quartiles, IQR — to the Diet A vs Diet B dataset, and frequency analysis on the Brand × Area dataset. Diet A had a higher and more consistent mean weight loss (5.34 kg, SD 2.54) than Diet B (3.71 kg, SD 2.77).

📁 Evidence: Summary Measures.xlsx

📈

Hypothesis Testing Worksheet

UNIT 8 LO 3

Mandatory worksheet applying inferential statistics. A paired t-test on filtration agents (t(11) = −3.26, p ≈ 0.0076) rejected the null, identifying Agent 1 as significantly more effective. A Welch t-test on bank cardholder income by sex showed no significant difference, with the small n = 3 highlighting the limits of small-sample inference.

📁 Evidence: Hyphothesis Testing.xlsx

📉

Charts Worksheet

UNIT 9 LO 3

Mandatory worksheet on chart selection: percentage frequency bar charts for brand preference in two demographics, interpreting competitive concentration. Emphasised how chart choice impacts the validity of conclusions drawn.

📁 Evidence: Charts Worksheet.xlsx

🎤

Research Proposal Presentation — Detecting Misinformation on Social Media

UNIT 10 68% Merit LO 4

A 15-slide research proposal outlining a hybrid detector with calibration and selective abstention, plus a governance mapping on transparency, audit, human oversight, security, and data protection.

📁 Evidence: Research Proposal.pptx

📝

Reflective Piece (Submission)

UNIT 12 LO 1

A 1,000-word reflection on statistical analysis skills, the research methods process, and personal/professional development, written using Rolfe et al.'s (2001) What, So What, Now What framework. Submitted as a separate Word document.

📁 Evidence: ReflectivePiece.docx

Unit Reflections

Key takeaways across all twelve units

01

Introduction to Research Methods

Differentiating inductive from deductive reasoning, I see ethical considerations as foundational obligations. This shift affects my empirical work planning.

02

Research Questions, Literature Review and Proposal

Practised turning broad topics into testable questions. The "RQ first, scope second" discipline showed when I rushed the proposal methodology to fit a slide budget.

03

Methodology and Research Methods

Compared quantitative, qualitative, and mixed-methods approaches against typical question shapes, revealing that a quantitative design suits my AI-driven misinformation thesis better.

04

Case Studies, Focus Groups and Observations

Observation methods have biases — like observer effects and sampling — so documenting protocol is as vital as data collection.

05

Interviews, Surveys and Questionnaire Design

Consider question wording as a tool that affects measurement, not just response rates. Pre- and post-testing should be habitual practices.

06

Quantitative Methods: Descriptive and Inferential Statistics

Re-grounded summary measures (mean, median, SD, IQR) were applied to the Diet A/B dataset for written interpretations.

07

Inferential Statistics and Hypothesis Testing

Applied paired t-tests, F-tests for variance, and Welch t-tests on datasets. The n = 3 income data exercise was a lesson in small-sample inference.

08

Data Analysis and Visualisation

Distinguished between descriptive and misleading visualisations. As an SRE, I read charts daily but seldom check their visual integrity.

09

Validity and Generalisability in Research

Internal and external validity, reliability, and associated threats are tied to the temporal-shift issue in misinformation detection.

10

Research Writing

Structured writing for a dissertation includes signposting, theoretical framing, and defended scope. The tutor's feedback on the literature review described it as "fragmented, itemised", impacting this unit's framing.

11

Professional Development and e-Portfolio

Completed the Skills Matrix, SWOT, and Action Plan; treated the portfolio as a continuous record.

12

Project Management and Managing Risk

Mapped a risk register to my proposal, including data licensing, compute, integration, and schedule with mitigations, influencing my dissertation scope.

Evaluation of Submissions

Critical self-assessment of the two summative artefacts

Literature Review

62% Merit

The review highlighted major task families and included current governance frameworks. However, three weaknesses emerged. First, the PRISMA search lacked a flow diagram, reducing methodological traceability. Second, I relied on books and industry guidance instead of the expected peer-reviewed journals. Third, some sections became generic and list-like, compressing depth into surface coverage. In future work, I would focus on a clear research question with a theoretical framework, conduct a documented PRISMA search with clear criteria, and write in my voice from the first draft.

Research Proposal Presentation

68% Merit

The proposal presented clear aims, objectives, and research questions, with the hybrid architecture diagram and governance mapping as key strengths. Two issues impacted the grade: the methodology lacked detail, describing the evaluation without procedural steps for reproducibility, and I missed a references slide, a presentation-level oversight. Both can be resolved with a methodology slide and a formatted references slide in Harvard style.

Professional Skills Matrix

Self-assessment against the module competency framework

Level key: Aware general understanding · Trained applies independently in some contexts · Proficient broad in-depth knowledge, minimal supervision · Expert leads and trains others
Competency Area Skill Level
Communication & LiteracyExpress information to technical and non-technical audiencesProficient
Communication & LiteracyCreate reports, diagrams, plans, manualsProficient
Commercial AwarenessKeep current with industry tools and emerging technologyProficient
Commercial AwarenessFamiliarity with codes of conduct (BCS / industry)Trained
Critical Thinking & AnalysisCritically analyse complex ideas in computingTrained
Critical Thinking & AnalysisRecognise gaps and seek additional informationProficient
Ethical AwarenessComply with applicable laws; maintain privacy and confidentialityProficient
Cultural AwarenessAct in the interest of the wider communityProficient
Teamwork & LeadershipCollaborate effectively in diverse teamsProficient
Teamwork & LeadershipGive and receive constructive feedbackTrained
Decision Making & InitiativeDecide on complex matters using multiple sourcesProficient
NumeracyInferential statistics and hypothesis testingTrained
IT & Digital — SQLDatabase queryingProficient
IT & Digital — PythonProgramming and scriptingProficient
IT & Digital — JavaProgrammingAware
IT & Digital — RStatistical computingAware
IT & Digital — noSQLDocument and key-value storesTrained
IT & Digital — GitRepository development and maintenanceProficient
IT & Digital — VLE / OfficeMoodle, Word, Excel, e-libraryProficient
Project ManagementRisk register, project life cycle, change managementTrained
Critical ReflectionSelf-assess and adjustTrained
ResearchLiterature search, synthesis, methodology designTrained

SWOT Analysis

Honest assessment of where I stand entering the dissertation

💪 Strengths

With a decade of site reliability engineering experience, my work is rooted in production reality. My multilingual skills and cloud certification enhance my commercial awareness. Daily practices include Git, scripting, and incident-driven decisions. Feedback from my MSc cohort and tutors has fostered a habit of embracing critique instead of deflecting it.

⚠️ Weaknesses

Inferential statistical reasoning lags my descriptive skills; I read percentile charts daily but seldom conduct paired tests on the differences I assert as real. My academic writing is evolving, but under pressure, I revert to list-oriented drafting, as noted by my tutor. My R and qualitative methods are less developed compared to my quantitative tools.

🚀 Opportunities

The MSc dissertation formalises evaluation methods (calibration, time-split, paired tests with confidence intervals) for work. Maintaining the reflection habit compounds module-on-module.

🛑 Threats

Time pressure from full-time work and dissertation leads me to templated drafting, the tutor's concern. Without clear structure at the start, originality diminishes.

Action Plan

Concrete, time-bound goals derived from the Skills Matrix and SWOT

  1. PRISMA discipline. Before drafting a review chapter for the dissertation, create a search log, inclusion/exclusion criteria, and a PRISMA flow diagram.
  2. Voice over template. Write prose in the first draft; use bullets only for genuine lists. Review each chapter draft.
  3. Statistical formalisation. For system evaluations in the dissertation or at work: use paired tests against baselines, 95% confidence intervals, calibration metrics (Expected Calibration Error), and time-split evaluation. Review the dissertation evaluation chapter.
  4. Living portfolio. Update the Skills Matrix, SWOT, and Action Plan after each module review instead of archiving.
  5. Targeted skill gaps. Move R from Aware to Trained with a tutorial cycle; enhance qualitative methods literacy through one text. Review: mid-dissertation checkpoint.

References

Berenson, M.L., Levine, D.M., Szabat, K.A. and Stephan, D.F. (2019) Basic Business Statistics: Concepts and Applications. 14th edn. Harlow: Pearson.

Page, M.J., McKenzie, J.E., Bossuyt, P.M., Boutron, I., Hoffmann, T.C., Mulrow, C.D., Shamseer, L., Tetzlaff, J.M., Akl, E.A., Brennan, S.E., Chou, R., Glanville, J., Grimshaw, J.M., Hróbjartsson, A., Lalu, M.M., Li, T., Loder, E.W., Mayo-Wilson, E., McDonald, S., McGuinness, L.A., Stewart, L.A., Thomas, J., Tricco, A.C., Welch, V.A., Whiting, P. and Moher, D. (2021) 'The PRISMA 2020 statement: an updated guideline for reporting systematic reviews', BMJ, 372, n71. Available at: https://doi.org/10.1136/bmj.n71

Rolfe, G., Freshwater, D. and Jasper, M. (2001) Critical Reflection in Nursing and the Helping Professions: A User's Guide. Basingstoke: Palgrave Macmillan.