90 Exam-Style Questions covering Modules 1 & 2 (Foundations, Problem Statements, Aims/Objectives, Hypotheses, Positivism, Epistemology & Ontology). Click "View Model Answer" to reveal the exact marking criteria.
Part 1: Foundational Questions (Questions 1 - 30)
P1-Q01Easy
Define scientific research in Computer Science.
NUST Model Answer [3 Marks]: A controlled, empirical, critical, and systematic inquiry designed to solve identified problems and create new, generalizable knowledge.
P1-Q02Easy
State the 6 core characteristics of scientific research.
What does it mean for research to be "Controlled"?
NUST Model Answer [3 Marks]: Isolating independent and dependent variables while holding extraneous variables (hardware, OS, background processes) strictly constant.
P1-Q04Easy
What is the difference between a Research Aim and a Research Objective?
NUST Model Answer [4 Marks]: Aim: The single broad ultimate goal (The "What"). Objectives: The 3-4 specific, measurable action steps (The "How").
P1-Q05Easy
What action verbs should be used when writing Research Objectives?
NUST Model Answer [3 Marks]: Design, Implement, Evaluate, Measure, Compare, Analyze (never passive terms like "study" or "understand").
P1-Q06Easy
Define a Research Hypothesis.
NUST Model Answer [3 Marks]: A tentative, testable prediction about the directional relationship between variables, structured as "If X, then Y."
P1-Q07Easy
Define a Research Question (RQ).
NUST Model Answer [3 Marks]: An open-ended inquiry used in qualitative or exploratory research when directional predictions cannot yet be made.
P1-Q08Easy
What is Positivism?
NUST Model Answer [3 Marks]: A research paradigm assuming objective reality exists independently of human observation, tested via quantitative, empirical measurement.
P1-Q09Easy
What is Interpretivism?
NUST Model Answer [3 Marks]: A paradigm assuming reality is socially constructed and subjective, explored via qualitative experiences and interviews.
P1-Q10Easy
What is Pragmatism?
NUST Model Answer [3 Marks]: A paradigm focusing on practical problem-solving using whatever methods (quantitative or qualitative) best answer the research question.
P1-Q11Easy
Define Epistemology.
NUST Model Answer [3 Marks]: The theory of knowledge—how we acquire knowledge and the relationship between the researcher and the known.
P1-Q12Easy
Define Ontology.
NUST Model Answer [3 Marks]: The study of reality and existence—what constitutes the nature of real-world phenomena.
P1-Q13Easy
Define Axiology.
NUST Model Answer [3 Marks]: The study of values, ethics, and researcher bias in scientific investigation.
P1-Q14Easy
Differentiate between Objectivist and Subjectivist Epistemology.
NUST Model Answer [4 Marks]: Objectivist: Knowledge is value-free and measured externally without researcher bias. Subjectivist: Knowledge is co-created through interaction between researcher and subjects.
P1-Q15Easy
Differentiate between Realist and Relativist Ontology.
NUST Model Answer [4 Marks]: Realist: Single objective reality exists independent of human perception (e.g. CPU silicon). Relativist: Multiple realities exist shaped by human contexts (e.g. UX satisfaction).
P1-Q16Easy
Why is research funding necessary in CS projects? State two reasons.
NUST Model Answer [3 Marks]: A clear, precise statement detailing a specific difficulty to be eliminated or a gap in existing knowledge.
P1-Q18Easy
State two characteristics of an empirical conclusion.
NUST Model Answer [3 Marks]: 1. Based on hard observable data, 2. Verifiable through direct experimentation or measurement.
P1-Q19Easy
What does it mean for research to be "Critical"?
NUST Model Answer [3 Marks]: Anticipating potential flaws, edge cases, and algorithmic biases that could invalidate the study.
P1-Q20Easy
What does it mean for research to be "Systematic"?
NUST Model Answer [3 Marks]: Following a logical, step-by-step sequence where each phase builds on the preceding phase.
P1-Q21Easy
Why must research conclusions be "Verifiable"?
Model Answer [3 Marks]: So independent researchers can replicate the experimental setup and obtain identical results.
P1-Q22Easy
Define Independent Variable in a CS experiment.
Model Answer [3 Marks]: The variable manipulated by the researcher (e.g. cache replacement algorithm type).
P1-Q23Easy
Define Dependent Variable in a CS experiment.
Model Answer [3 Marks]: The measured outcome variable affected by changes in the independent variable (e.g. latency in ms).
P1-Q24Easy
Define Extraneous Variable.
Model Answer [3 Marks]: Uncontrolled background variables that could confound the experiment if not held constant (e.g. CPU thermal throttling).
P1-Q25Easy
What is Null Hypothesis ($H_0$)?
Model Answer [3 Marks]: A statement assuming no significant difference or relationship exists between experimental groups.
P1-Q26Easy
What is Alternative Hypothesis ($H_1$)?
Model Answer [3 Marks]: A statement predicting a significant difference or directional effect between experimental groups.
P1-Q27Easy
What is a Theoretical Framework?
Model Answer [3 Marks]: The foundational structure of theories and concepts that supports and grounds the research study.
P1-Q28Easy
State the primary goal of Quantitative research.
Model Answer [3 Marks]: Generalizability, objectivity, and numerical prediction of variables.
P1-Q29Easy
State the primary goal of Qualitative research.
Model Answer [3 Marks]: In-depth understanding of human behaviors, context, and non-numerical experiences.
P1-Q30Easy
What is Generalizability in CS research?
Model Answer [3 Marks]: The extent to which experimental findings apply beyond the specific test sample to broader populations/systems.
Part 2: Medium / Applied Questions (Questions 31 - 60)
P1-Q31Medium
SCENARIO: You are benchmarking memory overhead of a custom Linux kernel scheduler vs standard CFS scheduler. Identify the Independent Variable, Dependent Variable, and 2 Controlled Variables.
NUST Model Answer [5 Marks]: Independent Variable: Scheduler Type (Custom vs CFS). Dependent Variable: Memory Overhead (MB) and Context Switch Latency ($\mu s$). Controlled Variables: 1. CPU hardware architecture, 2. Workload thread count.
P1-Q32Medium
Explain how a researcher holding a Positivist paradigm would evaluate a new biometric authentication system vs an Interpretivist researcher.
NUST Model Answer [5 Marks]: Positivist: Measures False Acceptance Rate (FAR), False Rejection Rate (FRR), and scan latency in ms. Interpretivist: Conducts user interviews to explore feelings of privacy invasion, trust, and physical discomfort.
P1-Q33Medium
A student writes an objective: "To understand how firewalls work." Rewrite this objective into NUST examination standard.
NUST Model Answer [4 Marks]: Corrected Objective: "To evaluate the packet inspection latency and throughput of stateful firewalls under simulated P2P proxy traffic."
P1-Q34Medium
Map a Realist Ontology to a CS physical system component and a Relativist Ontology to a CS human-facing component.
Model Answer [4 Marks]: Realist: Thermal throttling thresholds of CPU silicon. Relativist: User satisfaction score with a Figma UI wireframe.
P1-Q35Medium
Formulate a testable Null Hypothesis ($H_0$) and Alternative Hypothesis ($H_1$) for evaluating dynamic loop blocking in a network firewall.
Model Answer [5 Marks]: $H_0$: Implementing dynamic loop blocking causes no statistically significant reduction in unauthorized proxy traffic. $H_1$: Implementing dynamic loop blocking reduces unauthorized proxy traffic by at least 30%.
P1-Q36Medium
Explain why Pragmatism is often the default paradigm for CS Applied Research projects.
Model Answer [4 Marks]: Because practical CS solutions require both objective technical execution (benchmarks) and subjective user validation (usability/adoption).
P1-Q37Medium
How does Axiology influence algorithm design in biometric research?
Model Answer [4 Marks]: It forces the researcher to actively evaluate algorithmic fairness, prevent demographic bias in fingerprint matching, and protect data privacy.
P1-Q38Medium
Distinguish between Explanatory Sequential and Exploratory Sequential mixed methods.
Model Answer [4 Marks]: Explanatory: Quantitative data first → Qualitative follows to explain why. Exploratory: Qualitative interviews first → Quantitative survey follows to test generalizability.
P1-Q39Medium
Give an example of Convergent Parallel mixed methods in a mobile app study.
Model Answer [4 Marks]: Collecting app crash logs (quantitative) and user frustration surveys (qualitative) concurrently during a live trial.
P1-Q40Medium
Why must a problem statement include quantitative baseline stats from literature?
Model Answer [4 Marks]: To empirically establish that the problem is real, significant, and backed by cited evidence rather than personal assertion.
P1-Q41Medium
Explain the role of Delimitations in Chapter 1 of a research report.
Model Answer [3 Marks]: Explicitly defining the boundaries and scope of the study (what the researcher chose to exclude).
P1-Q42Medium
What is the risk of an un-controlled environment when measuring server throughput?
Model Answer [4 Marks]: Background network traffic or OS daemons introduce confounding noise, destroying internal validity.
P1-Q43Medium
How does an Objectivist Epistemology influence data collection tool design?
Model Answer [4 Marks]: Tools are automated scripts or calibrated sensors designed to record raw numerical data without human intervention.
P1-Q44Medium
Why are open-ended questions preferred in Qualitative Research Questions?
Model Answer [3 Marks]: To allow participants to describe rich context and unexpected perspectives without forcing binary answers.
P1-Q45Medium
Contrast Deductive vs Inductive reasoning in hypothesis testing.
Model Answer [4 Marks]: Deductive: Theory → Hypothesis → Empirical Test → Confirmation. Inductive: Observation → Pattern Recognition → Tentative Theory.
P1-Q46Medium
Scenario: Benchmarking Rust vs C++ memory safety. Is this basic or applied research?
Model Answer [3 Marks]: Applied Research—it evaluates specific practical language implementations to solve software safety issues.
P1-Q47Medium
Define External Validity in CS experiments.
Model Answer [3 Marks]: The degree to which experimental results can be generalized to real-world production environments.
P1-Q48Medium
Define Internal Validity in CS experiments.
Model Answer [3 Marks]: The degree to which observed changes in the dependent variable are strictly caused by the independent variable.
P1-Q49Medium
Why must CS research objectives be measurable?
Model Answer [3 Marks]: So the examiner can objectively verify whether each objective was successfully completed in Chapter 5.
P1-Q50Medium
What is a Construct in research?
Model Answer [3 Marks]: An abstract concept (e.g. "System Trust" or "Usability") operationalized into measurable indicators.
P1-Q51Medium
State two common sources of CS research topics.
Model Answer [3 Marks]: 1. Gaps identified in recent journal papers, 2. Real-world industry/institutional bottlenecks.
P1-Q52Medium
How does a pilot study enhance research validity?
Model Answer [3 Marks]: It tests instruments and experimental procedures on a small scale to identify flaws before full deployment.
P1-Q53Medium
Why should a title avoid acronyms unless standard in the field?
Model Answer [3 Marks]: To maintain clarity, indexability, and professional academic standards for global readers.
P1-Q54Medium
What is the role of an Institutional Review Board (IRB)?
Model Answer [3 Marks]: To review and grant ethical clearance for research involving human participants or sensitive data.
P1-Q55Medium
Why is convenience sampling flawed in quantitative research?
Model Answer [3 Marks]: It introduces severe selection bias, making results non-generalizable to the broader population.
P1-Q56Medium
Explain Triangulation in mixed methods research.
Model Answer [3 Marks]: Using multiple data sources/methods to cross-verify findings and eliminate single-method bias.
P1-Q57Medium
What is operationalization of a variable?
Model Answer [3 Marks]: Defining an abstract variable in terms of specific, measurable metrics (e.g. defining "speed" as throughput in Mbps).
P1-Q58Medium
What is the significance section in Chapter 1?
Model Answer [3 Marks]: Explaining who benefits from the research outcomes (e.g. university admins, developers, industry).
P1-Q59Medium
Scenario: Testing network latency in a local lab vs cloud. Which has higher internal validity?
Model Answer [3 Marks]: Local lab—because background internet traffic noise can be strictly eliminated.
P1-Q60Medium
Scenario: Which testing setup has higher external validity?
Model Answer [3 Marks]: Cloud—because it mirrors real-world production network conditions.
Part 3: Hard / Defense & Synthesis Questions (Questions 61 - 80)
P1-Q61Hard
DEC 2024 EXAM ESSAY DRILL: Synthesize the relationship between Philosophical Assumptions, Epistemology, and Ontology for a quantitative kernel driver benchmark study. [15 Marks]
NUST Model Answer [15 Marks]: 1. Paradigm: Positivism—assuming system behavior follows objective physical and logical laws. 2. Ontology: Realist—the CPU cycles, memory allocations, and register states exist objectively independent of the observer. 3. Epistemology: Objectivist—knowledge is acquired by recording raw automated logs without researcher interference, ensuring value-free empirical evidence.
P1-Q62Hard
DEFENSE: Why cannot Positivism alone effectively evaluate the success of an educational behavioral firewall? [10 Marks]
NUST Model Answer [10 Marks]: Positivism measures quantitative metrics (packet drop rates, latency spikes). However, firewall adoption in schools also depends on qualitative human factors (admin usability, student privacy concerns, policy compliance), which require an Interpretivist or Pragmatist approach.
P1-Q63Hard
Evaluate how an un-justified sample size threatens both internal and external validity.
Model Answer [5 Marks]: Small sample sizes lead to low statistical power (Type II errors) internally and render findings non-representative externally.
P1-Q64Hard
Defend the use of a 1-sentence Aim over a multi-paragraph statement.
Model Answer [4 Marks]: A 1-sentence aim provides razor-sharp focus on the ultimate research destination without confusing scope or methods.
P1-Q65Hard
Compare Positivist and Interpretivist approaches to data anonymization.
Model Answer [4 Marks]: Positivist: Hashing IDs and IP addresses in logs. Interpretivist: Redacting identifying quotes and pseudonymizing participant names.
P1-Q66Hard
Explain how a Pragmatist researcher resolves conflicting quantitative and qualitative data.
Model Answer [5 Marks]: By conducting secondary explanatory analysis to determine if qualitative context reveals unmeasured quantitative variables.
P1-Q67Hard
Critique this hypothesis: "Our new app is better than existing apps."
Model Answer [5 Marks]: Flawed: "Better" is un-operationalized. Must specify exact metrics (e.g. "reduces latency by 20%").
P1-Q68Hard
Differentiate between Research Scope and Delimitations.
Model Answer [4 Marks]: Scope: The extent of topics covered. Delimitations: Conscious choices by the researcher to exclude specific areas.
P1-Q69Hard
Why is a software feature list invalid as a set of research objectives?
Model Answer [5 Marks]: Feature lists describe software construction; research objectives must describe investigation, evaluation, and empirical comparison.
P1-Q70Hard
Explain how hardware thermal throttling introduces confounding variables.
Model Answer [5 Marks]: It reduces CPU clock speeds during long benchmark runs, making later test iterations artificially slower.
P1-Q71Hard
Contrast Subjectivist Epistemology with Objectivist Epistemology in AI ethics research.
Model Answer [5 Marks]: Objectivist measures demographic classification accuracy; Subjectivist explores user perceptions of algorithmic fairness.
P1-Q72Hard
Explain how a well-formulated problem statement simplifies methodology design.
Model Answer [4 Marks]: "A well-formulated problem is half-solved"—it directly dictates the required variables, data collection tools, and test suites.
P1-Q73Hard
Why must hypotheses be directional in quantitative experimental research?
Model Answer [4 Marks]: To allow one-tailed statistical significance testing ($p < 0.05$) against a specific predicted outcome.
P1-Q74Hard
Evaluate the risk of using non-standard performance metrics in CS research.
Model Answer [4 Marks]: Prevents direct comparison against published baselines, destroying external comparability.
P1-Q75Hard
Synthesize how Axiology governs data reporting integrity.
Model Answer [4 Marks]: Mandates full disclosure of all data runs, including failed runs or negative results.
P1-Q76Hard
Differentiate between Research Problem and Research Topic.
Model Answer [4 Marks]: Topic is broad subject area (e.g. Network Security); Problem is specific un-solved gap within that topic.
P1-Q77Hard
Why is generalizability lower in qualitative CS studies?
Model Answer [4 Marks]: Because sample sizes are small and contextualized to specific human groups.
P1-Q78Hard
Explain how warm-up runs eliminate initial cache miss noise in benchmarks.
Model Answer [4 Marks]: They ensure CPU caches and memory pages are populated so measured iterations reflect steady-state performance.
P1-Q79Hard
Contrast Relativist Ontology with Realist Ontology in automated grading systems.
Model Answer [4 Marks]: Realist measures exact code execution pass/fail; Relativist evaluates subjective code readability and style.
P1-Q80Hard
Critique the use of passive verbs in research objectives.
Model Answer [4 Marks]: Passive verbs (e.g. "to learn") are un-assessable because they describe internal mental states rather than observable research deliverables.
TRICK: Can a research study hold a Realist Ontology and a Subjectivist Epistemology simultaneously? Explain with a CS example.
NUST Model Answer [5 Marks]: YES! Example: Human-Computer Interaction (HCI). The physical computer hardware and screen pixels exist objectively (Realist Ontology), but the user's perception of usability and satisfaction is constructed subjectively through interaction (Subjectivist Epistemology).
P1-Q82EXAMINER TRICK
TRICK: A student claims: "My hypothesis was proven 100% true with absolute certainty." What is the examiner flaw in this statement?
NUST Model Answer [5 Marks]: FLAW! Hypotheses in empirical science are never proven with 100% certainty; they are only supported or rejected at a specific statistical confidence level (e.g. $p < 0.05$).
P1-Q83EXAMINER TRICK
TRICK: Does building a faster algorithm automatically make a project valid CS research?
Model Answer [5 Marks]: NO! Building code is engineering. It becomes research only when accompanied by controlled evaluation, baseline comparison, and generalizable formal analysis.
P1-Q84EXAMINER TRICK
TRICK: Can an Objective be identical to the Research Aim?
Model Answer [4 Marks]: NO! The Aim is the single overall destination; Objectives are discrete, sequential action steps required to reach that destination.
P1-Q85EXAMINER TRICK
TRICK: Is Positivism applicable to qualitative interview studies?
Model Answer [4 Marks]: Generally NO. Qualitative interviews align with Interpretivism or Constructivism. Positivism requires quantitative, value-free numerical measurement.
P1-Q86EXAMINER TRICK
TRICK: What is wrong with writing: "Our aim is to build a Flutter app"?
Model Answer [5 Marks]: It frames a software engineering task as the research goal. Research aims must focus on designing, evaluating, or comparing a scientific solution to a knowledge gap.
P1-Q87EXAMINER TRICK
TRICK: Can a research question be answered with a simple "Yes" or "No"?
Model Answer [4 Marks]: NO! Research questions must be open-ended (How, What, Why, To what extent) to encourage deep empirical inquiry.
P1-Q88EXAMINER TRICK
TRICK: Is an uncontrolled internet speed test a valid benchmark for network latency research?
Model Answer [5 Marks]: NO! Internet speed tests suffer from uncontrolled ISP throttling, background traffic, and routing variance, destroying internal validity.
P1-Q89EXAMINER TRICK
TRICK: If a researcher's value bias is present in a study, which assumption is violated?
Model Answer [4 Marks]: Axiological (or Value-Free Objectivist) assumption.
P1-Q90EXAMINER TRICK
TRICK: What is the single biggest trap students make in Chapter 1 of a CS research report?
Model Answer [5 Marks]: Confusing a commercial software development project with scientific research by failing to state an empirical problem gap or research questions.