Introduction to Social Research

Course Description

Introduction to Social Research is an introduction to the foundational principles, tools, and techniques of research in the social science disciplines, aimed at a working knowledge of research and its application to the real world.

Within the SCNS taxonomy, ISS is the Social Sciences interdisciplinary prefix. Daytona State publishes this at 3 credits, offered spring, giving approximately 45 contact hours.

The phrase "application to the real world" is what makes this course broadly worth taking rather than only useful to future sociologists. Claims derived from social research reach everyone constantly — in news coverage, in political argument, in workplace decisions, in advertising, and in the endless supply of confident statements about what studies show. Most people have no framework for telling a well-designed study from a badly designed one, and this course supplies it. That is a general-purpose skill, and it is the reason a methods course belongs in a general education pathway rather than only in a major.

Learning Outcomes

Required Outcomes

• Describe the logic of scientific inquiry as applied to social phenomena.
• Distinguish inductive and deductive reasoning in research design.
• Describe the relationship between theory, hypothesis, and evidence.
• Formulate a researchable question and a testable hypothesis.
• Conduct and synthesize a literature review.
• Distinguish conceptualization, operationalization, and measurement.
• Define and identify variables, including independent, dependent, and control variables.
• Describe levels of measurement and their implications for analysis.
• Assess reliability and validity, including their principal types.
• Describe probability and non-probability sampling methods and their trade-offs.
• Describe sampling error, sample size, and generalizability.
• Describe experimental design and the logic of causal inference.
• Distinguish correlation from causation and identify spurious relationships.
• Design and evaluate survey instruments and question wording.
• Describe interviewing methods, including structured and semi-structured approaches.
• Describe field research, ethnography, and participant observation.
• Describe content analysis and the use of existing data and secondary sources.
• Distinguish quantitative and qualitative approaches and describe mixed methods.
• Apply descriptive statistics and interpret measures of central tendency and dispersion.
• Interpret basic inferential results, including statistical significance and its limits.
• Read, interpret, and critique tables, graphs, and figures.
• Apply research ethics, including informed consent and institutional review requirements.
• Evaluate a published study's design and identify its limitations.
• Write a research proposal or report in a discipline-appropriate format.

Optional Outcomes

• Use statistical software to analyze a dataset.
• Code and analyze qualitative data.
• Describe evaluation research and applied policy research.
• Describe visualization principles and misleading graphics.
• Conduct a small original study.
• Describe big data, computational social science, and their ethical issues.

Major Topics

Required Topics

• The logic of social inquiry
• Induction, deduction, and theory
• Research questions and hypotheses
• The literature review
• Conceptualization and operationalization
• Variables and levels of measurement
• Reliability and validity
• Sampling methods
• Sampling error and generalizability
• Experimental design and causal inference
• Correlation, causation, and spuriousness
• Survey design and question wording
• Interviewing
• Field research and ethnography
• Content analysis and secondary data
• Quantitative, qualitative, and mixed methods
• Descriptive statistics
• Inferential statistics and significance
• Reading tables and graphs
• Research ethics and the IRB
• Critiquing published research
• Proposal and report writing

Optional Topics

• Statistical software
• Qualitative coding and analysis
• Evaluation and policy research
• Data visualization and misleading graphics
• Conducting an original study
• Big data and computational methods

Resources & Tools

• The Practice of Social Research (Earl Babbie) — the dominant text in this field by a wide margin, and clear enough to read rather than merely consult.
• Research Methods in the Social Sciences (Frankfort-Nachmias) or Social Research Methods (Bryman) — the alternatives; Bryman is particularly good on qualitative methods.
• How to Lie with Statistics (Darrell Huff) — old, short, and still the fastest inoculation against misleading numbers.
• The Craft of Research (Booth, Colomb & Williams) — the best general guide to research as an activity.
• Pew Research Center (pewresearch.org) — free, and unusually transparent about methodology; their methods write-ups are teaching material in themselves.
• US Census Bureau, American Community Survey, and data.gov — free datasets, including Florida-specific data at county level.
• General Social Survey (gss.norc.org) and ICPSR — free access to major social science datasets, and the GSS Data Explorer runs in a browser.
• Florida Department of Health FLHealthCHARTS and Florida's OEDR — free state and county data, useful for a locally grounded project.
• Google Scholar and your library's databases — and remember that library database access ends when you leave the institution.
• R with RStudio, or Python, or JASP — all free; JASP is the gentlest introduction and produces APA-formatted output. SPSS is common in social science but expensive outside a campus licence.
• Your college librarians — free, expert at exactly the literature review task, and badly underused.

Career Pathways

• Research assistant — universities, survey organizations, and think tanks; the direct entry route.
• Market research analyst — SOC 13-1161; one of the larger and better-paid destinations for these skills.
• Survey researcher — SOC 19-3022.
• Programme evaluator — non-profits, government agencies, and grant-funded programmes must demonstrate outcomes, and people who can design and report an evaluation are scarce.
• Data analyst — with additional statistical and software skills, a substantially higher ceiling.
• Public health research support — county health departments and Florida's academic medical centres.
• Policy analyst — state and local government, and Florida's legislative and agency research offices.
• Social work, criminal justice, and human services — where evidence-based practice requires reading research critically.
• Journalism — data journalism specifically; see this repository's DIG2151 guide.
• User experience research — a well-paid technology role built on interviewing and observational methods.
• Continuation to a bachelor's degree in sociology, psychology, criminology, political science, public administration, or social work, where a methods course is a required foundation.

Special Information

⚠ Correlation is not causation — and knowing the slogan is not the same as using it

• Nearly everyone can recite this and very few apply it, including in professional contexts. The course's value is in making the reasoning habitual rather than quotable.
• Three conditions are required for a causal claim: the variables must covary, the cause must precede the effect, and alternative explanations must be ruled out. The third is where almost everything fails.
• Spurious relationships are produced by a third variable causing both. Ice cream sales and drowning rise together because both rise with temperature — and most real examples are less obvious than that.
• Selection effects are the most common trap in social research. If people choose whether to be in a group, differences between groups may reflect who chose rather than what the group did. Nearly every claim about a programme's effectiveness must confront this.
• Randomization is what makes experiments powerful, because random assignment balances unmeasured differences. Much social research cannot randomize for ethical or practical reasons, which is exactly why its causal claims must be hedged.
• Ask "what else could produce this pattern?" every time. That single habit is the most transferable thing in the course.
• Reverse causation is easy to miss. Does the association run the direction the author assumes? Frequently it could run either way, or both.
• Watch the language in reporting. "Linked to," "associated with," and "may contribute to" describe correlations; headlines routinely translate them into causal claims the study did not make.

⚠ Sampling determines what a study can say — and it is where most of them break

• The sample determines the population you can generalize to, and nothing else about the study can repair a bad sample.
• Probability sampling is what licenses inference to a population. Convenience samples — students in one class, volunteers, people who answered an online poll — describe themselves and no one else, however large they are.
• Size does not fix bias. A large biased sample is a precisely wrong answer, and this is the single most counterintuitive fact in the course. The 1936 Literary Digest poll had over two million responses and got the election badly wrong because of who was in it.
• Non-response bias is the modern version of the same problem. Response rates to telephone and mail surveys have fallen dramatically, and whether non-responders differ systematically from responders is the live methodological question in survey research today.
• Online opt-in panels are not probability samples regardless of how they are weighted, and the weighting assumptions do real work that should be disclosed.
• Margin of error describes sampling error only. It says nothing about question wording, coverage, non-response, or measurement error — which are often larger.
• Subgroup results have much larger error than the headline figure, and reporting on small subgroups is where poll coverage most often misleads.
• Ask who was excluded. People without phones, without internet, in institutions, not fluent in the survey language, or unhoused are systematically underrepresented — and they are frequently the people a study is nominally about.

⚠ Question wording is measurement — and it is easy to do badly on purpose

• How a question is asked changes the answer, sometimes dramatically. Two surveys can produce opposite results on the same topic through wording alone, and both can be technically honest.
• Leading questions embed the answer. "Do you agree that the wasteful programme should be cut?" is not a measurement instrument.
• Double-barrelled questions ask two things at once and cannot be answered cleanly — "Do you support increased funding and stricter oversight?" has four possible positions and two response options.
• Social desirability bias makes people overreport voting, exercise, and charitable giving, and underreport drinking, prejudice, and income difficulties. Anonymity and careful wording reduce it; nothing eliminates it.
• Response options constrain the answer. The scale offered, whether a middle option exists, whether "don't know" is available, and the order of options all shift results measurably.
• Question order matters. Earlier questions prime later ones, and this is a well-documented effect rather than a curiosity.
• Pretest your instrument. Every survey you write will contain a question that means something different to respondents than it does to you, and only a pretest reveals it.
• Read the actual question wording in any poll you cite. Reputable organizations publish it; the ones that do not are telling you something.
• Advocacy polling exists — surveys designed to produce a number for a press release. Knowing how to spot one is a practical civic skill.

⚠⚠ Research ethics: consent, harm, and the IRB

• Social research can harm people, and the field's ethical framework exists because it did. The Tuskegee syphilis study, Milgram's obedience experiments, and the Stanford prison experiment are the standard cases, and they are taught because the harms were real.
• The Belmont Report principles — respect for persons, beneficence, and justice — are the framework, and the Common Rule is the federal regulation.
• Institutional Review Board approval is required before data collection begins for research involving human subjects, including surveys and interviews for a class project where the institution requires it. Approval cannot be obtained retroactively, and data collected without it generally cannot be used.
• Ask your instructor about your college's requirements at the start of the term, not the week before you plan to collect data.
• Informed consent is a process, not a form — participants must understand the purpose, the risks, that participation is voluntary, and that they may stop.
• Vulnerable populations require additional protection, including minors, prisoners, and people with impaired capacity.
• Confidentiality obligations are real and continue after the project ends. Think about how identifiable data will be stored and destroyed before you collect it.
• Deception requires justification and debriefing, and is subject to specific review.
• Digital and social media data raise unsettled questions. Publicly visible does not mean ethically available, and scraping platform data has its own legal and ethical dimensions. Rule 11 applies — this area is developing.
• Report honestly. Suppressing inconvenient findings, hunting through analyses until something reaches significance, and presenting exploratory results as confirmatory are research misconduct, not technique.

⚠ Statistical significance is widely misunderstood — including by people who cite it

• A p-value is not the probability that the hypothesis is true, and it is not the probability the result occurred by chance. It is the probability of observing data at least this extreme if the null hypothesis were true — a narrower and less satisfying statement.
• "Statistically significant" does not mean "important." With a large enough sample, trivially small differences reach significance. Always ask about effect size, which is the question that actually matters.
• "Not significant" does not mean "no effect." It frequently means the study was too small to detect one, and absence of evidence is being confused with evidence of absence.
• The 0.05 threshold is a convention, not a law of nature, and treating it as a bright line between real and unreal is a known problem in the literature.
• P-hacking and the replication crisis are part of the honest picture. Substantial portions of published findings in psychology and adjacent fields have failed to replicate, and a methods course that does not mention this is teaching an outdated confidence.
• Preregistration, open data, and replication are the field's response, and they are worth knowing about as the direction the discipline is moving.
• Confidence intervals communicate more than p-values because they show magnitude and precision together. Prefer them when you can.
• Graphs mislead easily — truncated axes, inconsistent scales, and misleading area comparisons are common. Look at the axes first, every time.

How Florida course levels affect transfer

The first digit of an SCNS number denotes the year of offering, not transferability. Courses at the 1000 and 2000 levels transfer transparently between Florida public institutions, and 3000 to 4000 is unproblematic since both are upper division. The boundary that actually matters is 2000 to 3000, where lower-division credit generally cannot satisfy an upper-division requirement.

ISS1301 is 3 credits and approximately 45 contact hours, offered spring. Expect a research proposal or a small original study as the major deliverable, along with critiques of published research — which is the most transferable exercise in the course.

As a 1000-level course it transfers on the ordinary lower-division basis between Florida public institutions, and it is a common requirement or prerequisite in sociology, psychology, criminology, and social work pathways. Confirm how it applies at your intended receiving institution, since some programmes require their own discipline-specific methods course rather than an interdisciplinary one.