Practical Research 2: Nature of Inquiry and Quantitative Research Designs

Overview of Quantitative Research

• Quantitative research is an objective, systematic, and empirical investigation of observable phenomena through the use of computational, mathematical, or statistical techniques.
• It focuses on numerical analysis and the relationship between numbers and specific events.
• The central goal is to provide a clear and unbiased understanding of data that can be generalized to a larger population.

Characteristics of Quantitative Research

• Objective: The researcher seeks to remain impartial and unbiased, ensuring that the results are based on data rather than personal opinions.
• Clearly Defined Research Questions: These questions specify what needs to be answered, often focusing on "Who," "What," "Where," "When," "Why," and "How."
• Structured Research Instruments: Experiments, surveys, and questionnaires are structured to gather standardized data from all respondents.
• Numerical Data: Findings are presented in the form of scales, percentages, proportions, or other statistical measurements.
• Large Sample Sizes: To ensure the reliability of the findings and their generalizability to the overall population, a significant number of participants ($N$) is required.
• Replicability: Studies must be designed in a way that other researchers can repeat the experiment or survey to verify the original findings.
• Future Outcomes: The data allows for the prediction of future trends and outcomes based on current statistical evidence.

Strengths and Weaknesses of Quantitative Research

Strengths

• Objectivity: Findings are based on concrete data and statistical evidence, reducing researcher bias.
• Sophisticated Analysis: Statistical techniques allow researchers to analyze complex data sets and identify trends quickly.
• Real and Unbiased: Because it relies on numbers, the results are considered objective and factual.
• Efficiency: Numerical data can be processed and analyzed rapidly using software.
• Testing Results: It is highly effective for testing theories or results obtained from preliminary qualitative studies to reach a final, solid conclusion.

Weaknesses

• Large Sample Requirements: Gathering enough respondents ($n$) can be difficult and time-consuming.
• Cost: Large-scale quantitative studies can be expensive due to the logistical requirements of reaching many participants.
• Lack of Context: Numerical data often ignores the contextual factors or the "human side" of the data which might explain variations or motivations.
• Instrument Difficulty: Designing a perfect research instrument is hard; if tools are not constructed correctly, the data may be incomplete or inaccurate.
• Rigid Data Collection: If not done seriously, nuances and deeper meanings are lost because questions are strictly structured.

Classification of Quantitative Research Designs

Quantitative research is broadly divided into Experimental and Non-Experimental designs.

Experimental Research

Experimental research establishes causality (cause-and-effect relationships) and involves the manipulation of independent variables ($IV$) to see their effect on dependent variables ($DV$).

1. Pre-Experimental Design

This design has the least internal validity and provides the researcher with very little control over extraneous variables.

• One-Shot Case Study: A single group is observed after an intervention. There is no control group or pre-test for comparison.
  • Example: A researcher measures the stress levels of a group after they attend a one-day workshop. Because there is no baseline (pre-test) or comparison group, the change cannot be definitively linked to the workshop.
• One-Group Pretest-Posttest Design: A single group is measured before and after a treatment.
  • Example: A teacher gives a pre-test, applies a new teaching method, and then gives a post-test. Improvements are noted, but without a control group, other factors like student motivation could have influenced the result.

2. Quasi-Experimental Design

Involves manipulation of variables but lacks either randomization or a control group.

• Non-Equivalent Groups Design: Similar to a true experiment but participants are not randomly assigned.
  • Example (Education): Comparing Two schools (School $A$ and School $B$) where School $A$ gets a new method and School $B$ stays the same. Pre-existing differences between schools (e.g., socioeconomic status) may bias the result.
  • Example (Health): Comparing smokers who voluntarily enroll in a cessation program with those who do not.

3. True-Experimental Design

The most rigorous design involving random assignment, a control group, and researcher control over variables.

• Pretest-Posttest Controlled Group: Participants are randomly assigned to experimental and control groups. Both groups take a pre-test and a post-test.
  • Example: Testing a new blood pressure medication by randomly assigning patients to a drug group or a placebo group and measuring their blood pressure before and after the treatment period.
• Posttest-Only Controlled Group: Participants are randomly assigned, but no pre-test is given to prevent "sensitization" or practice effects.
  • Example: Evaluating an online learning platform vs. traditional classrooms where students take a standardized test only at the end of the semester.
• Solomon Four-Group Design: A complex design with four groups (two experimental, two control). Two groups receive pre-tests; two do not. This helps determine if the act of pre-testing itself affects the subjects' performance.

Non-Experimental Research

Focuses on associations, connections, or descriptions without manipulating variables.

1. Descriptive Research

Aims to observe and document natural occurrences to serve as a starting point for theories.

• Survey Research: Collecting data from samples of a population.
  • Cross-Sectional: Data is gathered at a single point in time (e.g., a political poll before an election).
  • Longitudinal: Data is gathered from the same group over an extended period (e.g., a cohort study following infants until adulthood).

2. Correlational Research

Determines the direction and strength of associations between variables.

• Bivariate Correlational: Relationship between two variables (e.g., Study Time vs. $GPA$).
• Prediction Studies: Using existing data to forecast future outcomes (e.g., atmospheric data for weather forecasting).
• Multiple Regression Prediction: Using several predictor variables to estimate one outcome (e.g., using study hours, attendance, and prior $GPA$ to predict final grades).

3. Ex-Post Facto (Causal-Comparative)

Examines relationships after the events have already occurred. The researcher does not intervene.

• Example: Analyzing health records of current smokers vs. non-smokers to find an association with lung cancer, rather than forcing people to smoke in a controlled experiment.

4. Other Non-Experimental Types

• Comparative: Compares two or more samples on specific variables (e.g., comparing stress levels of urban vs. rural adolescents).
• Evaluative: Determines the impact or success of a program or institution (e.g., assessing a government job training program).
• Methodological: Focuses on the implementation of varying methodologies to integrate data from different disciplines.

The "MR.C" Rule for Design Identification

To identify the research design, ask three questions:

1. Manipulation (M): Is there manipulation of variables? If NO, it is Non-Experimental. If YES, proceed.
2. Randomization (R): Is there randomization in samples? If NO, it could be Quasi-Experimental or Pre-Experimental.
3. Control (C): Is there a Control Group? If YES (along with M and R), it is True Experimental. If NO (but there is manipulation), it is Pre-Experimental.

Practical Application: Research Matters

• Students are encouraged to use quantitative research to advocate for a sustainable planet, following the Bedan mission of "Ora et Labora" (Pray and Work).
• Numerical data can be used to track laboratory hours, social media usage habits, exercise frequency, and sleep patterns ($8$ hours typical goal) to improve student well-being and academic success.
• Key biblical guidance for researchers: "For the Lord gives wisdom; from his mouth come knowledge and understanding." - Proverbs $2:6$"