Quantitative Research Fundamentals: Null Hypothesis, Variables & Core Concerns

Null Hypothesis ( $H_0$ )

• Implicit, complementary statement to the research (alternative) hypothesis ($H_a$).
• Claims no relationship or difference exists between the variables under study.
• Presumed true until empirical evidence supports the research hypothesis.
• Researchers attempt to reject $H<em>0$ to lend support to $H</em>a$.
• Example scenario
  • Research claim: “Students work better on Monday morning than on Friday afternoon.”
  • Independent Variable (IV): Day (Monday vs. Friday)
  • Dependent Variable (DV): Standard of work / amount recalled
  • $H_0$: “There is no significant difference in the amount recalled on a Monday morning compared with a Friday afternoon.”

Research Hypotheses & Research Questions

• Hypotheses are always tentative; they guide but do not predetermine the outcome.
• Research (alternative) hypothesis—not the null—is the primary focus and is what is ultimately presented in the research report.
• Research Questions (RQs) are preferred when:
  • Little is known about the phenomenon.
  • Previous studies show conflicting results.
  • The goal is to describe rather than predict or explain.

Variables

• A variable is any property or characteristic of people or things that varies in quality or magnitude.
  • Must have two or more levels (categories, values).
  • Every variable in a hypothesis/RQ is identified as independent or dependent.

Independent Variable (IV)

• Manipulation or natural variation of the IV is hypothesized to cause changes in other variables.
• Technically, the term “independent variable” is reserved for experimental studies where manipulation occurs.
• Synonyms: antecedent variable, experimental variable, treatment variable, causal variable, predictor variable.
• Example: Want to know whether plants grow better in hot vs. cold areas.
  • IV = Temperature (hot vs. cold)

Dependent Variable (DV)

• The variable of primary interest; the research aims to describe, explain, or predict its changes.
• DV changes as a function of the IV.
• In non-experimental research, DV is also called criterion variable or outcome variable.
• Example (continued):
  • DV = Plant growth (amount of growth)

Applied Example: Stress & Heart Rate

• Research interest: “How does stress affect human heart rate?”
• IV = Stress level (manipulated or measured)
• DV = Heart rate

Relationship Considerations

• One cannot specify an IV without specifying its corresponding DV(s).
• The number of IVs and DVs depends on study complexity.
• Variable structure dictates the appropriate statistical test (e.g., $t$-test, ANOVA, regression).

Four Preoccupations of Quantitative Research

1. Measurement

• Goal: assign numbers to observations in a consistent manner so they stand for empirical events.
• Two core quality criteria:
  • Reliability – Consistency or stability of a measure.
  • If repeated measurement under identical conditions gives the same result, the instrument is reliable.
  • Validity – Extent to which an indicator really measures the intended concept.
  • Example: An alarm clock set for 6:30 a.m. but ringing at 7:00 a.m.
    • Highly reliable (rings at exactly 7:00 every day) but invalid for waking you at 6:30.

2. Causality

• Seeks to explain why things happen—not just to describe that they do.
• Requires:
  • Temporal precedence (cause precedes effect).
  • Covariation (IV & DV change together).
  • Non-spuriousness (no plausible alternative explanations).

3. Generalization

• Aims for findings that extend beyond the context in which data were collected.
• Relies on:
  • Representative sampling.
  • Clear specification of the population.

4. Replication

• Ability to repeat the essential components of a study.
• Standardized procedures enhance credibility and allow independent verification.

Strengths and Weaknesses of Quantitative Research

Strengths

• Findings can often be generalized to the target population when sampling is appropriate.
• Researchers can select samples of individuals, communities, or organizations in ways that ensure representativeness.
• Standardized instruments allow replication across locations and time, producing comparable results.
• Additional advantages:
  • Replicated findings across many populations/sub-populations strengthen generalizability.
  • Provides precise, numerical data amenable to statistical analysis.
  • Data analysis is relatively less time-consuming once data are collected.
  • Well-suited for studying large numbers of participants.

Weaknesses

• Some groups (e.g., sex workers, drug users, illegal immigrants, squatters, ethnic minorities) are hard to reach, making representative quantitative sampling difficult.
• Large-scale surveys/experiments can be expensive and time-consuming.
• Research methods are often inflexible once data collection begins; instruments cannot easily be modified mid-study.
• The researcher’s pre-defined categories or theories may not match local participants’ understandings.
• Knowledge produced might be too abstract or generalized for direct application to specific local contexts or individuals.