Introduction to Research Methods in Sociology

Quantitative vs Qualitative Research

  • Research in sociology falls under two umbrellas: quantitative and qualitative.
  • Quantitative research = numbers, statistics; qualitative research = words, meanings.
  • Key distinctions:
    • Quantitative: tells us the scope and scale of an issue, often seeks cause-and-effect relationships, uses numerical data.
    • Qualitative: seeks to understand meaning, experience, and how people perceive social phenomena; uses non-numerical data such as interview transcripts.
  • Some studies use mixed methods to get a fuller picture of what’s going on, why it’s happening, and how people experience it.

The Scientific Method in Original Research

  • Original research = you collect your own data.
  • Steps of the scientific method (as applied to sociology):
    • Step 1: Identify a research question. Examples:
    • How many students are enrolled in each major at OCC, broken down by gender?
    • What are the most popular majors for male vs. female students?
    • Step 2: Conduct a literature review. What has already been written about this topic? Are there gaps in knowledge you could fill?
    • Step 3: Form a hypothesis. A hypothesis is your best educated guess about the answer to the research question.
    • Step 4: Develop a research design. Choose methods (quantitative, qualitative, or mixed) that will best answer the question.
    • Step 5: Collect data using the chosen methods.
    • Step 6: Analyze the data.
    • Quantitative analysis involves mathematical calculations and statistics.
    • Qualitative analysis involves coding transcripts or notes to identify themes and patterns.
    • Step 7: Report findings. Share results through articles, presentations, infographics, social media, etc., depending on the audience.
  • Mixed-methods note: Some studies combine quantitative and qualitative approaches to provide a fuller picture of what’s happening, why, and how people experience it.

Qualitative Methods

  • Ethnography: studying people in their natural environment to understand what things mean to them and why people behave as they do.
    • Example: Study student life on campus by observing classes, sports events, clubs, and the student center.
    • Field notes: contemporaneous notes taken during observation.
  • Participant observation: researcher becomes a participant in the setting (e.g., enrolls as a student) and studies from inside the group.
  • Interviews: one-on-one conversations with a researcher-guided list of questions; aims to capture personal experiences and meanings.
  • Focus groups: group interviews where participants discuss topics together.
    • Advantages: dynamic responses, new ideas sparked by others; disadvantages: some participants may dominate, others may not share as much.
  • Data analysis in qualitative research: coding transcripts or field notes to identify themes, keywords, and patterns across interviews or observations.
  • Practical notes: choice of method depends on whether you want to understand experiences and meanings (qualitative) or measure quantities and relationships (quantitative).

Quantitative Methods

  • Surveys: the most common quantitative method.
    • Open-ended questions: allow free responses.
    • Closed-ended questions: provide predefined response options (true/false; Likert scales such as 1–5).
    • Numerical data enables statistical analysis and calculations such as percentages and averages.
    • Example calculations:
    • Percentage of respondents answering a given option: P(option)=N<em>optionN</em>total×100P(option) = \frac{N<em>{option}}{N</em>{total}} \times 100
    • Average score on a 1–5 scale: xˉ=1n∑<em>i=1nx</em>i\bar{x} = \frac{1}{n} \sum<em>{i=1}^{n} x</em>i
  • Big data and secondary data analysis:
    • Researchers may reuse existing large datasets collected by governments or universities and apply new analyses to them.
  • Geographic Information Systems (GIS) mapping: a newer method that maps where phenomena occur locally and geographically; used for disease prevalence, resource distribution, etc.
  • Experiments (in the social world): randomized or quasi-experimental designs.
    • Key idea: separation into a test (intervention) group and a control (no intervention) group to isolate effects of the treatment.
    • Example: A Department of Labor program for noncustodial parents randomized participants to receive temporary jobs versus no job offer.
    • Outcome measures include employment status and child support payments over time.
    • The goal is to determine whether the intervention has a lasting impact beyond the program period.
  • Important caveat about experiments:
    • Random assignment helps isolate causal effects, but real-world contexts may introduce confounding variables.

Variables and Causation vs Correlation

  • What is a variable?
    • Any piece of social information that can take different values (e.g., participation in a program, employment status).
  • Relationship types:
    • Correlation: two variables move together (e.g., ice cream sales and shootings peak in August).
    • Causation: one variable causes a change in another.
    • Correlation does not imply causation; a third variable may influence both.
  • Example to illustrate correlation vs causation:
    • Ice cream sales and shootings are correlated in summer due to warm weather increasing outdoor activity, not because ice cream causes shootings.
  • Illustration of a controversial claim and critical thinking:
    • Some claims may incorrectly attribute causation (e.g., vaccines and infant mortality in Japan) without accounting for other important variables.
    • Important to consider: universal health care, prenatal/postnatal care, home visits, parental leave, and broader health policies when interpreting infant mortality rates.

Case Studies and Applications

  • Illustrative case from the lecture:
    • A real-world program for noncustodial parents: random assignment to a temporary-jobs program vs no program; evaluation over time showed short-term benefits but no long-term gains compared to non-participants.
    • This demonstrates how experiments can test causal effects of interventions.
  • The role of critical thinking in evaluating statistics:
    • Always ask: Is there a plausible causal mechanism? Are there confounding variables? Could bias or framing influence conclusions?
    • Example discussed: a video claiming Japan stopped all vaccines in 1994 and infant mortality fell; the claim omitted context and other variables that explain infant mortality trends.

Ethics in Research and Responsible Practice

  • Be aware of who benefits from research. Howard Becker's question: "Whose side are we on?"
    • Researchers should acknowledge biases and recognize that neutrality is difficult or impossible.
    • Consider whether research serves the interests of powerful groups or the marginalized and oppressed.
  • Informed consent:
    • Participants must be informed about the study, its risks, and give voluntary written consent to participate.
    • This ethical requirement arose from historical abuses and the Tuskegee syphilis study.
  • Tuskegee syphilis study (historical example):
    • African American men with syphilis were studied without their informed consent and without providing penicillin once it became available; the study pursued knowledge at the expense of participants’ health.
    • Modern research ethics require informed consent and protection of participants' rights and welfare.

Dissemination and Audience Considerations

  • Findings can be shared in multiple formats depending on the audience:
    • Academic articles and conferences
    • Public presentations and policy briefs
    • Infographics for social media
    • Short videos or TikTok-style content to reach broader audiences
  • The dissemination choice influences question framing, emphasis, and the level of technical detail provided.

Key Takeaways for Exam Preparation

  • Distinguish between quantitative and qualitative research and recognize when a mixed-methods approach is advantageous.
  • Understand the steps of the scientific method in sociology and how they connect to original research goals.
  • Be able to describe qualitative methods (ethnography, field notes, participant observation, interviews, focus groups) and the trade-offs of each.
  • Be able to describe quantitative methods (surveys, big data, GIS mapping, experiments) and how they enable numerical analysis and causal inference.
  • Grasp the concept of variables, correlation vs causation, and the role of potential confounding variables.
  • Recognize the ethical foundations of research (informed consent, stripped neutrality, and the responsibility to marginalized groups).
  • Appreciate the importance of literature reviews for identifying gaps and situating a new study within existing knowledge.
  • Be able to discuss how to design an original study (research question, literature review, hypothesis, design, data collection, analysis, dissemination).

Quick Review Questions (to test your understanding)

  • What is the primary difference between quantitative and qualitative research?
  • Name three qualitative methods and one advantage and one limitation for each.
  • What are the steps of the scientific method as applied to sociology?
  • How does a randomized controlled trial isolate the effect of an intervention?
  • Why is correlation not the same as causation? Give a real-world example.
  • What is informed consent, and why is it essential in social research?
  • Who were the key figures or concepts discussed related to ethics and bias (e.g., Howard Becker, Tuskegee)?
  • How can findings be disseminated to different audiences, and why does this matter for research impact?