Practical Research 2 – Kinds of Quantitative Research Design

Descriptive Research Design

• Definition & Purpose

  • Aims to accurately describe the characteristics, behaviors, or phenomena of a particular population, situation, or event.
  • Focuses on answering the “what,” “who,” “where,” and “how” questions—not the “why.”
  • Does not manipulate variables; therefore, it does not establish cause-and-effect relationships.

• Typical Methods Used

  • Surveys (self-administered, interviewer-administered, online, paper-based).
  • Direct observations (structured or unstructured).
  • Case studies or case series.
  • Secondary statistical analysis of existing datasets (e.g., census, school records).

• Key Features

  • Variables remain unchanged; researcher merely records or categorizes what exists.
  • Provides a snapshot at one point in time (cross-sectional) or across time (longitudinal descriptive).
  • Generates baseline data often used for further correlational or experimental studies.

• Example Research Topics Mentioned in the Transcript

  • Measuring students’ study habits.
  • Observational study describing traffic patterns in a city.
  • Reporting demographic characteristics of a community.

• Variable Examples for “Students’ Study Habits” Survey

  • Time Management
  • Study Techniques / Strategies
  • Study Environment (noise level, location)
  • Motivation & Self-Regulation
  • Resource Utilization: frequency of textbooks, online resources, tutoring, etc.

• Sample Survey Item (Likert-type)

  • “How often do you review your notes after each class?” (1 = Never … 5 = Always)

• Sample Frequency Item

  • “On average, how many hours do you spend studying for all your courses per week?” Options: 0{-}5,\;6{-}10,\;11{-}15,\;16{-}20,\;>20\text{ hours}

• Ethical / Practical Considerations

  • Ensure anonymity/confidentiality when collecting self-reported behaviors.
  • Use clear, non-leading questions to minimize response bias.
  • Pilot-test questionnaires for reliability and validity before full deployment.

• Real-world Relevance

  • Descriptive data often guides school policy, public health interventions, urban planning, and resource allocation.

• Assignment 1 Prompt (Transcript)

  • “What variables are you going to measure in an observational study describing traffic patterns in a city?”
  • Potential variables: traffic volume per intersection, vehicle type counts, peak/off-peak speed, stop-light compliance rate, pedestrian crossings, weather conditions.

Correlational Research Design

• Definition & Purpose

  • A non-experimental design that examines the statistical relationship or association between two or more naturally occurring variables.
  • Answers questions about whether variables move together (covary) and, if so, in what direction (positive/negative) and strength.

• Key Statistical Tools

  • Pearson’s Correlation Coefficient ($r$)
  • Measures the linear relationship between two continuous, normally distributed variables.
  • Values range from $-1$ (perfect negative) through $0$ (no correlation) to $+1$ (perfect positive).
  • Spearman’s Rank Correlation Coefficient ($\rho$ or $r_s$)
  • Non-parametric; assesses monotonic relationships using ranked data or when assumptions of Pearson’s $r$ are violated.

• Interpreting Correlation Strength (Common Benchmarks)

  • $|r| \approx 0.00{-}0.19$: very weak or none.
  • $|r| \approx 0.20{-}0.39$: weak.
  • $|r| \approx 0.40{-}0.59$: moderate.
  • $|r| \approx 0.60{-}0.79$: strong.
  • $|r| \approx 0.80{-}1.00$: very strong / near-perfect.

• Vital Caveat: “Correlation ≠ Causation”

  • Even a perfect correlation cannot confirm that variable A causes variable B; confounding variables or reverse causality may be at play.

• Example Study #1: Screen Time vs. Physical Activity in Children

  • Research Question: “Is there an association between daily screen time and daily minutes of moderate-to-vigorous physical activity among primary school children?”
  • Variable 1 (Screen Time): Parent-reported log or questionnaire (e.g., total minutes per day over one week).
  • Variable 2 (Physical Activity): Child wears an accelerometer for a week; device records daily minutes of MVPA.
  • Possible Outcomes: Negative correlation expected (more screen time → less physical activity).

• Example Study #2: Sleep Duration vs. Academic Stress in University Students

  • Research Question: “Is there a relationship between students’ average nightly sleep hours and their self-reported academic stress levels?”
  • Variable 1 (Sleep Duration): Self-reported hours of sleep per night over the last week.
  • Variable 2 (Academic Stress): Standardized stress scale (Likert-type items such as “I feel overwhelmed by my coursework”).
  • Anticipated Direction: Negative correlation (less sleep → higher stress), yet causality cannot be assumed.

• Broader Applications

  • Early detection of risk factors (e.g., correlation between smoking frequency and lung capacity).
  • Hypothesis generation for later experimental research.

• Ethical / Practical Considerations

  • Obtain informed consent when collecting personal or health data.
  • Protect participant privacy, particularly with wearable sensors (e.g., accelerometers).
  • Disclose limitations: avoid over-interpreting correlational findings in media reports, policy briefs, or academic papers.

Measurement Instruments & Scales Mentioned

• Likert Scale

  • Ordinal rating scale to measure agreement, frequency, importance, etc.
  • Typical format: 5- or 7-point options (e.g., 1 = Strongly Disagree … 5 = Strongly Agree).
  • Allows transformation into numerical scores for descriptive or correlational analysis.

• Frequency / Range Questions

  • Capture count or duration data via discrete categories (e.g., study hours 0-5, 6-10).
  • Advantage: easy for respondents; produces ordinal/interval data for descriptive statistics.

• Objective Devices

  • Accelerometers in physical-activity studies provide minute-by-minute motion counts, enabling precise MVPA minutes.

Assignments & Review Prompts

• Assignment 1 (Descriptive Research)

  • Identify and list variables for an observational traffic study.

• Assignment 2 (Correlational Research)

  1. Define Pearson’s Correlation Coefficient and give an example (e.g., height vs. weight in adults).
  2. Define Spearman’s Correlation Coefficient and give an example (e.g., ranking of students’ class standing vs. ranking of leadership scores).
  3. Define Likert Scale and give an example item (e.g., “I feel confident in mathematics”—1 = Strongly Disagree … 5 = Strongly Agree).

Connections to Previous / Future Content

• Descriptive designs often precede correlational studies by establishing baseline distributions and identifying potential predictor variables.
• Correlational findings can lead to experimental or quasi-experimental studies to test causality.

Practical & Philosophical Implications

• Policymaking: Descriptive data may justify resource allocation (e.g., more streetlights at high-traffic times).
• Public Awareness: Correlational evidence on screen time and health can inform parenting guidelines.
• Ethical Communication: Researchers must emphasize the non-causal nature of correlations to avoid misinformation.