Variables, Measurement & Quantitative Research
Kinds of Variables
- Independent: manipulated cause/treatment (e.g., peace\;loving\;environment \;\rightarrow\; X)
- Dependent: observed effect/response (e.g., test\;anxiety \;\rightarrow\; Y)
- Confounding/Extraneous: uncontrolled factors influencing Y; must be minimized (e.g., family background)
- Discrete: countable, whole-number values (e.g., # of votes, respondents)
- Continuous: measurable on a scale (e.g., height, temperature, income)
- Quantitative: convey amount or frequency (e.g., popularity votes)
- Qualitative: describe kinds/types; coded by names or labels (e.g., yes/no)
- Categorical
• Nominal – mutually exclusive, no order (civil status)
• Ordinal – ordered categories (small/medium/large)
Levels of Measurement
- Nominal: names/categories; no quantity
- Ordinal: ranked order; unequal intervals acceptable
- Interval: equal intervals, no true zero (e.g., temperature)
- Ratio: equal intervals with true 0 (e.g., height, weight)
Research Approaches
- Qualitative
• Describes qualities & meanings through narratives
• Advantages: rich, real-life detail; flexible probing
• Disadvantages: small n, subjective, limited generalizability - Quantitative
• Tests hypotheses with numerical data & stats
• Advantages: objectivity, large samples, replicable, generalizable
• Disadvantages: context-thin, artificial settings possible - Mixed Method
• Combines both; boosts validity & depth
• Drawback: time-consuming, possible result conflicts
Quantitative Approach Overview
- Structured tools (questionnaires, tests)
- Larger, representative samples
- Pre-planned design
- Data in numbers/statistics
- Enables generalization, prediction, causal testing
Strengths
- Broader scope via bigger n
- Objective, accurate; bias minimized
- Results replicable; comparison across studies/time
- Summarizes vast info; cross-category comparison
Limitations
- Numerical focus may miss context
- Controlled/artificial settings may limit realism
Types of Quantitative Research
- Descriptive: numerically portray variables (surveys, observations)
• Correlational: gauges strength/direction of relation between \ge 2 variables; no causation inferred - Causal-Comparative: explores cause–effect using non-manipulated groups (e.g., gender ➔ performance)
- Experimental: manipulates X, random assignment to test effect on Y
• True Experimental
• Quasi-Experimental: lacks full control/randomization; examines natural events
Hypotheses
- Null (H_0): predicts no difference/relationship (e.g., "There is no significant relationship between gender and performance")
- Alternative (H_a)
• Directional: predicts specific change ("Group1 > Group2")
• Non-Directional: relationship/difference expected but direction unspecified - Writing Tips
• One issue at a time; consistent variable order (independent ➔ dependent)
• State type/direction for correlational hypotheses (positive/negative)
• Align with literature & theory; clearly measurable variables
Writing Research Titles
- Aim for (10\text{–}12) words
- Summarize main issue & primary variables
- State relationship and/or population
- Omit redundant phrases (e.g., "A Study of")
- Include only essential, informative words
Literature Review & Topic Selection
- Source Types: General (books, reports), Primary (journals), Secondary (textbooks)
- Steps to Pick Topic
• Reflect on class discussions, observed issues
• Conduct preliminary library/web search; narrow focus
• Draft brief descriptions & possible research questions for shortlisted topics - Reviewing Literature
• Target recent (\le 5-year-old) works; \approx (20\text{–}50) sources for undergraduate projects
• Evaluate each work:
– Clearly defined, significant problem? Variables/theory aligned?
– Contrasting or supporting findings noted?
– Sound design, sampling, measures, analysis?
– Relevance to your variables, context, locale? - Purpose: identify gaps, justify study significance, refine variables & methods
Defining Variables
- Lexical/Conceptual Definition: authoritative dictionary or scholarly meaning
- Operational Definition: specifies how variable is measured/observed in study
- Importance
• Sets precise scope & parameters (who, what, where, how data collected)
• Guides instrument choice & statistical tests
• May evolve during literature review but must be finalized before measurement
Key Reminders
- Measure each variable separately, matching its level (nominal, ordinal, etc.)
- Ensure coherence among theory, variables, hypotheses, and methods
- Use instruments for indirectly observable constructs (e.g., attitude scale)