Research Methods: Sampling, Data & Analysis

• Sampling should accurately represent the population to ensure research findings are generalizable, using methods like random, stratified, or purposive sampling based on the research design.
• Operationalization involves defining and measuring key concepts or variables, converting abstract concepts into measurable factors for consistent data collection.
• Data collection methods include surveys, interviews, observations, or experiments, depending on the research design.
• Data analysis identifies patterns, trends, or relationships through statistical or thematic analysis to address research questions.
• Methodology outlines the research design, sampling method, data collection, and analysis process, explaining how the study ensures validity and reliability, including ethical considerations.

Problem

• Problems may involve estimating the level of a particular behavior.
• Identifying prevalence of a particular behavior.
• Finding patterns or relationships.
• Identifying existing trends.
• Testing hypotheses.
• Establishing cause and effect relationships through interventions.

Questions

• Types of questions include determining the level of something.
• Finding the prevalence.
• Determining differences.
• Identifying trends; finding correlates.
• Establishing associated factors.
• Identifying predictors; and examining effects.

Components

• Sampling refers to selecting participants or items from a larger population which should represent the population accurately so the research findings are generalizable.

Module 0: Quantitative Research

• Quantitative research involves collecting and analyzing numerical data to identify patterns, test hypotheses, and make predictions.
• This approach is objective, systematic, and replicable, used to quantify behaviors, attitudes, opinions, or other variables.
• Characteristics of Quantitative Research
• Objective Measurement: Focuses on quantifying data through statistical analysis, including descriptive statistics like means, SD, and frequencies.
• Structured Methodology: Uses predetermined tools like surveys, tests, or existing data.
• Large Sample Size: Ensures generalizability and statistical significance.
• Replicability: Allows repetition by other researchers for verification.
• Statistical Analysis: Employs statistical models (e.g., descriptive and inferential statistics) to interpret data.
• Types of Quantitative Research
• Descriptive research: Describes characteristics or functions (e.g., surveys, observational studies).
• Correlational research: Explores relationships (e.g., study hours and test scores).
• Quasi-experimental research: Examines cause-effect relationships without full experimental control (e.g., field studies).
• Experimental research: Tests causal relationships using manipulated variables and control groups (e.g., clinical trials).
• Data Collection Methods
• Surveys/Questionnaires: Collects large amounts of data through closed-ended questions.
• Experiments: Involves manipulation of variables in controlled settings.
• Observations: Systematically records behaviors under specified conditions.
• Secondary Data: Uses existing datasets (e.g., government statistics, company records).
• Variables in Quantitative Research
• Independent Variable (IV): The manipulated cause or factor.
• Dependent Variable (DV): The measured effect or outcome.
• Control Variables: Factors kept constant to prevent influence on the results.
• Extraneous Variables: Uncontrolled factors that might affect what is being tested.

Sampling Techniques

• Probability Sampling: Each population member has a known chance of selection through simple random, stratified, or systematic sampling.
• Non-Probability Sampling: Not all members have an equal chance of selection, including convenience sampling, purposive sampling, and quota sampling.

Data Analysis Techniques

• Descriptive Statistics: Summarize data using mean, median, mode, and standard deviation.
• Inferential Statistics: Draw conclusions and generalize findings using T-tests, ANOVA, regression analysis, and chi-square tests.

Validity and Reliability

• Validity: Accuracy in measuring what the study intends to.
• Internal Validity: Ensuring causal relationship is real.
• External Validity: Generalizability of findings.
• Reliability: Consistency of results across time and contexts.
• Test-Retest Reliability: Consistency over time.
• Inter-Rater Reliability: Agreement between observers.
• Strengths: Produces objective and generalizable results, allows for replication and statistical testing, and is efficient for large-scale data analysis.
• Limitations: May overlook contextual or qualitative insights, rigid structure may not capture evolving phenomena, and requires statistical knowledge for accurate interpretation.
• Ethical Considerations: Requires informed consent where participants are aware and agree to the study.
• Maintaining confidentiality and protecting participants' data privacy, ensuring non-maleficence by avoiding harm, and ensuring transparency by providing honest reporting and disclosure of findings.

Module 1: Study Designs

• Research Design

Census

• A complete enumeration of all members, used for policy planning, budgeting, and population projection.
• Advantage: large data, no sampling bias.
• Disadvantage: high cost, double counting.

Observational Design

• Cross-Sectional
• Data Collection occurs one time, capturing a snapshot.
• Advantage: shows what happens on one particular date. May be easy to do (short timeframe).
• Disadvantage: cannot see the trends or make any comparisons.

Repeated Cross-Sectional

• Multiple data collection (interview or face-to-face).
• Advantages: Can see the trends (changes) and see how it changes in the society.
• Disadvantage: Hard to do since mahaba or matagal (period of time required).

Longitudinal

• Observing characteristics (interview or face-to-face).
• Tracking of same individuals throughout.
• Advantage: Can see how a change occurs in the same environment (e.g., child development).
• Disadvantage: Time-consuming.

Prospective Cohort

• Similar to longitudinal, studies have the same individuals, but these are categorized as exposed and unexposed.
• No manipulation and no randomization.
• Advantage: Results are observable.
• Experimental Design
• Quasi-Experimental
• Manipulate the variables with intervention with no randomization
• Behavior change should occur in the experimental group, no changes in the control group.

Randomized Controlled Trial

• Experimental, manipulate variables with randomization to remove bias.
• Gold standard for causality: ensures equal chances to participate across groups.
• High standards and cause and effect.

Research Tips

• One can use quotation marks for exact phrase searches and 'pdf: or docs:' for file type searches.
• Use AND to only include or exclude certain results

Types of Literature

• Includes blogs, communities and news articles
• Journal Articles are more reliable as they are peer-reviewed.
• Reliable sources are published, have volumes, and are peer reviewed.

Analysis of Literature

• Can use a table to organise study/design, samples, and methods, with the goal to see the themes,
• This can help provide any conclusion or limitations

Research Gap

• Identifies unexplored study areas, population or methodological gaps, knowledge gaps or contradictory findings.

Citations

• Used for referencing in text or in the bibliography

Module 3: Theoretical and Conceptual Frameworks

• Help to have academic research skills where you systematicly collect information for a particular phenomenon
• This could uncover facts for mass report or determine if something is feasible.
• You can derive research hypotheses, or develop theoretical background

Theory

• Includes grand theory, middle-range theory, helping to understand the methodology. variable.
• Theories can have things like:
  • Plan Behavior: which looks at intention, attitude,
  • Social Determininant Theory, which looks at autonomy, competence, and relatedness

Conceptual Framework

• Can be more illustrate, and be tailored to specific problems

Hypothesis

• HO (null): no difference / independent
• HA (alternative): with application/association

Module 4: Variables

• Measurements are key where you should understand the realtionship between dependent variables
• One should be able to control some variablies to make the independent and dependent more accurate

Qualitative vs Quantitative

Variables (Categorical): Describe qualities or characteristics (Numerical): Represent measurable quantities

• Data can be measured from nominal to relational which helps understand the data even more

Operationalization

• Key since it needs to be defined how things will be measured. Example: Measuring stress with the stress scale