Chapter 1 Notes — The Research Process

What Is the Research Process?

• A set of activities social scientists engage in to answer questions, examine ideas, or test theories.
• Five stages of the research process (as presented in the lecture):
  • Develop a research design
  • Examine a social relationship and study relevant literature
  • Asking the Research Question
  • Formulating the Hypotheses
  • Collecting Data
  • Analyzing Data
  • Evaluating the Hypotheses
  • THEORY: Contribute new evidence to literature and begin again
  • Positionality is acknowledged as part of the research process
• The process is iterative: theories inform questions and data collection, which in turn refine theory.

The Research Question and Its Foundations

• Empirical Research definition: research based on information verifiable by data and direct experience.
• Quantitative emphasis: factors that can be measured and quantified; produces numerical results suitable for statistical analysis.
• Limits of non-empirical reasoning: cannot rely solely on moral judgment, speculation, or subjective preference for empirical questions.
• Example research question: Do average math scores differ across racial groups in elementary school?
  • Shows a cross-group comparison using numerical data (e.g., STAAR 3rd Grade Math in Texas – Grade Level Proficiency).
• Contextual factors that might explain differences in math scores:
  • Socioeconomic status and related resources
  • Teacher expectations and school funding
  • Family resources and parental support
  • Neighborhood effects, redlining, access to learning materials
  • Test prep, curriculum quality, and school district resources
• The goal is to move beyond race as a sole predictor to understand contextual factors that explain variation in outcomes.

Contextual Factors Shaping Outcomes (Social Determinants Context)

• Socioeconomic Context: income, neighborhood resources, access to high-quality schools influence outcomes, not race per se.
• School and Classroom Environment: teacher expectations, curriculum quality, school funding vary across schools serving different populations.
• Family and Community Context: parental education, tutoring availability, cultural attitudes toward math.
• Historical and Structural Inequalities: segregation, underinvestment, systemic racism shaping opportunities.
• Student Positionality and Identity: discrimination experiences, stereotype threat, sense of belonging in math classes.

Deficit vs Strengths-Based Approaches

• Deficit-based approach asks: What is missing? What is wrong?
  • Can lead to dysfunction, helplessness, and blame on groups.
• Strengths-based approach asks: What assets exist that can be leveraged?
  • Promotes innovation and addressing community needs.
• Case study: The Moynihan Report (1965) framed disparities as family weaknesses rather than structural racism and economic inequality.
• Deficit Lens: The Moynihan Report asserted issues like the “Negro Family” structure as the root cause of disparities, shaping policy for decades and often ignoring systemic barriers.
• Slide guidance for reframing questions (Small Groups):
  1) Identify deficit assumptions in the question
  2) Consider contextual factors
  3) Reframe to focus on understanding and supporting the group, not deficits
• Strengths-based Research examples: strong kinship bonds, work orientation, achievement orientation, adaptability of family roles, religious orientation.
• Contextual example data point: 1960 Black family structure shows stark differences in incarceration rates and family organization data used to illustrate the risks of deficit framing.
• In analyzing questions about inequalities, the lens used matters; examining inequalities is not inherently deficit-based if framed to understand structure and contexts.

Historical Context and Ethical Considerations

• 1960 Black Family Structure: White male incarceration rate ~262 per 100,000; Black male rate ~1,313 per 100,000 (about 5x higher).
• Moynihan’s framing had policy implications and contributed to long-term harmful narratives about Black families.
• Ethical implication: researchers must avoid blaming individuals or groups and instead examine structural factors and systemic conditions.

The Role of Theory in Research

• Theory: a set of assumptions and propositions used to explain, predict, and understand phenomena.
• Purpose of theory: links observed data to a broader conceptual understanding of why something is happening.
• Where theory comes from: experiential knowledge (practice) and theoretical knowledge (book learning).
• The role of theory in guiding research questions, data collection, analysis, and interpretation.
• Important caveat: be critical of the theory selected; theories carry assumptions that shape what is asked and what is observed.

Examples in Practice: Social Determinants of Health (SDOH) and Health Disparities

• SDOH definition (WHO framing in lecture): the conditions in environments where people are born, live, learn, work, play, worship, and age that affect health outcomes and risks.
• Five SDOH domains:
  • Economic Stability
  • Education Access and Quality
  • Health Care Access and Quality
  • Neighborhood and Built Environment
  • Social and Community Context
• Example data visual: Age-adjusted COVID-19 deaths per 100,000 Americans through Aug 18, 2020 by race:
  • Black: 118.8
  • Indigenous: 111.8
  • Pacific Islander: 105.6
  • Latinx: 99.2
  • Asian: 44.8
  • White: 33.3
• Additional health disparities context: COVID-19 infections by race explained by multiple factors including social determinants, discrimination, comorbidities, access to healthcare, location, beliefs about care, type of work.
• World Health Organization framing: health disparities are differences closely linked with social, economic, and environmental disadvantage, driven by social conditions.

Why Theory Is Important

• Theory shapes the research question and data collection/analysis plans.
• It influences potential solutions and predictions/hypotheses.
• Researchers should remain critical of the theory’s assumptions and limitations.

Theoretical Frameworks and Child Development (Ecological Systems)

• Bronfenbrenner’s Ecological Systems Theory (adapted to focus on Black youth development):
  • Microsystem: immediate environment (family, peers, school)
  • Mesosystem: interactions between microsystems (e.g., family-school relationships)
  • Exosystem: indirect environment (parents' workplaces, community resources)
  • Macrosystem: broader cultural and societal context (laws, norms, policies)
  • Chronosystem: life-course events and transitions over time
• Adapted integrative model (R³ISE) foregrounds racism, resilience, and resistance in the childhood ecosystem, including components like family, mentors, health systems, community, and policy systems.
• The model emphasizes interactions across levels and the intergenerational and systemic nature of development in Black youth.

Formulating the Hypotheses

• Definition: a statement predicting the relationship between two or more observable attributes/variables.
• Key features:
  • Testable statements
  • Indicates direction of relationship
• Types of relationships:
  • Positive relationship: When one variable is high, the other is high; when one is low, the other is low. r > 0
  • Negative (inverse) relationship: When one variable is high, the other is low; when one is low, the other is high. r < 0
• Core components to specify in hypotheses:
  1) Variable: values that can be measured or manipulated
  2) Unit of Analysis: the entity being studied (e.g., individual students, classrooms, schools)
  3) Independent and Dependent Variable: cause–effect relationships
• Example (in schools): I hypothesize that regardless of race, as a teacher’s equity mindset increases, children’s acceptance of differences increases.
  • Variables: Teacher Equity Mindset (independent), Children’s Acceptance of Differences (dependent)
  • Presentation note: Independent variable is typically plotted on the x-axis; dependent variable on the y-axis: X = ext{Independent Variable}, \ Y = ext{Dependent Variable}
• Another example (caregiver socialization): I hypothesize that higher caregiver racial socialization is related to children’s decreased acceptance of others (depending on theory), with variables: Caregiver Racial Socialization (IV), Children’s Acceptance of Others (DV).
• Key terms: Unit of Analysis, Descriptive vs Inferential focus, Measurement of variables, and the potential role of theoretical knowledge in informing hypotheses.

Collecting Data

• Data collection techniques:
  • Observations (classrooms, customers, wildlife)
  • Surveys/Questionnaires (online, paper, telephone)
  • Interviews (face-to-face, phone, video)
  • Focus Groups
  • Experiments (lab, field)
  • Case Studies (individual, group, event, situation)
  • Secondary Data Analysis (census data, prior research)
  • Content Analysis (newspapers, social media, policy documents)
  • Ethnography (immersive study of communities)
  • Longitudinal Studies (long-term effects, e.g., health trajectories)
• Practical question for final project: what type of data will you explore? (choose techniques accordingly)

Levels of Measurement and Data Coding

• Levels of measurement describe how variables are measured or classified:
  • Categorical vs Numeric
  • Categorical: Nominal and Ordinal
  • Numeric: Interval and Ratio
  • Continuous vs Discrete (within Numeric)
• Variable hierarchy (quick guide):
  • Categorical → Nominal / Ordinal
  • Numeric → Continuous / Discrete
  • Further distinction: Interval vs Ratio (with true zero in Ratio)
• Nominal (categorical, no intrinsic order): example categories like Census Race categories with numeric codes (1, 2, 3, …).
• Ordinal (categorical with order): categories have a meaningful order but not equal spacing (e.g., Likert scales).
• Interval (numeric with equal spacing but no true zero): examples include temperature scales; IQ scores have no true zero.
• Ratio (numeric with meaningful zero): examples include annual income, height, weight; has true zero.
• Detailed table of levels (Nominal, Ordinal, Interval, Ratio) includes characteristics like mutually exclusive categories, equal spacing, and true zero.
• Discrete vs Continuous within Numeric:
  • Continuous: can take any value within a range (e.g., height, time)\
  • Discrete: takes only specific values (e.g., number of children, test items correct)
• Why coding is necessary:
  • Enables processing by statistical software
  • Optimizes storage and supports multivariate models
  • Ensures consistency across dataset and aids visualization
  • Example: data coding uses a codebook and can be implemented in Excel or statistical software
• Numerical coding example: Nominal Race categories coded as 1–6 (American Indian/Alaska Native, Asian, Black, Native Hawaiian/Other Pacific Islander, White, Multiracial)
• Dichotomous variables: two possible values; consider the implications of dichotomizing experiences like racial discrimination (loss of nuance and information)

Analyzing Data

• Core questions in analysis:
  • What is the Unit of Analysis? (individuals, groups, organizations, geographic locations)
  • What is the Population vs Sample? (definitions and implications for generalizability)
  • Descriptive vs Inferential Statistics
  • Measurement Quality (reliability and validity)
• Population vs Sample definitions:
  • Population (N): total set of cases researchers are interested in
  • Sample (n): a subset drawn from the population
• Descriptive statistics: summarizes data from a sample or population (e.g., means, frequencies, standard deviations)
• Inferential statistics: makes inferences about population characteristics from sample data (e.g., hypothesis tests, confidence intervals, p-values)
• Measurement quality concepts:
  • Reliability: consistency of measurements across time and raters
  • Validity: whether the measurement actually captures the intended construct
• Example of measurement quality in practice: Fostering racial empathy through virtual reality (VR)
  • Reliability example: consistent scores across trials but possibly not valid for measuring racial empathy
  • Validity example: instrument content aligns with intended construct but scoring inconsistent across graders or sessions
• Conceptual takeaway: a tool can be reliable but not valid; or valid but not reliable.

Evaluating Hypotheses

• Definition: rigorously assess hypotheses using systematic data analysis to validate, refine, or reject original assumptions.
• Process is part of the iterative cycle: after data collection and analysis, hypotheses are evaluated and the study informs the next round of theory and design.

Looking Ahead & Takeaways

• Key takeaways:
  • Research questions are driven by personal experience and educational theory.
  • Variables have different properties that have implications for analysis (levels of measurement, type, etc.).
• Looking ahead: upcoming schedule and readings (Chapter 2 on Distributions & Graphic Presentation) and subsequent course activities.

Quick Summary of Theoretical and Practical Connections

• Theory guides questions, data collection, and interpretation; empirical data tests and refines theory.
• Context matters: avoid simplistic, deficit-laden interpretations of race; consider structural and contextual factors.
• Ethical practice requires awareness of positionality and a critical stance toward theory and data sources.
• The research process is cyclical: theory informs questions, questions guide data collection, data analysis updates theory, and the cycle repeats to contribute to literature.

Appendix: Key Definitions and Formulas

• Empirical Research: research grounded in observed and verifiable data.
• Positive Relationship: r > 0
• Negative Relationship: r < 0
• Independent Variable (IV): the cause or predictor (often plotted on the x-axis)
• Dependent Variable (DV): the effect or outcome (often plotted on the y-axis)
• Unit of Analysis: the entity studied (e.g., individual, school, group)
• Population: the entire set of cases of interest (N)
• Sample: a subset of the population (n)
• Levels of Measurement: Nominal, Ordinal, Interval, Ratio
• Dichotomous Variable: a variable with two possible values
• Reliability vs Validity: consistency vs accuracy of measurement
• Bronfenbrenner’s Ecological Systems Theory: Microsystem, Mesosystem, Exosystem, Macrosystem, Chronosystem
• SDOH: Economic Stability, Education Access & Quality, Health Care Access & Quality, Neighborhood & Built Environment, Social & Community Context
• COVID-19 Race-Based Deaths (illustrative data):
  • Black 118.8, Indigenous 111.8, Pacific Islander 105.6, Latinx 99.2, Asian 44.8, White 33.3 per 100,000 through Aug 18, 2020
• The Moynihan Report (1965) as an example of deficit framing and its historical impact on policy and discourse
• Deficit Lens vs Strengths-Based Lens: implications for research questions and policy recommendations
• Data Coding: purpose, methods, and why clear codebooks matter
• Data Collection Techniques: a spectrum from qualitative (interviews, focus groups, ethnography) to quantitative (surveys, experiments, secondary data)
• Hypothesis Examples (from lecture):
  • Example 1: I hypothesize that regardless of race, as teacher equity mindset increases, children’s acceptance of differences increases.
  • Example 2: I hypothesize that higher caregiver racial socialization is related to children’s increased awareness of group membership (context-dependent hypothesis; requires specification of direction and measurement).

End of Chapter 1 Notes