Lecture: Sociological Research Methods
Lecture script · about 15–20 minutes spoken · Week 2
Opening
Picture three people reading three claims. A neighborhood group wants to know whether its new program helped. A small-business owner is staring at a customer poll. A health worker is reading a headline about a survey. They need the same skill. How was the evidence gathered? Whom does it speak for? And does it show a cause, or only a pattern?
In Week 1, you learned to see a private trouble as part of a public pattern, and to ask four questions of any statistic. In this lecture I want to show you where the answers to those questions come from. We'll walk the research cycle, match questions to methods, check our measuring sticks, decide whom a sample can speak for, and earn the word "cause." Then we'll meet three research traditions, the ethics rules that protect people, and one real survey figure read four ways. Next week, in Week 3, the course turns to culture, and you'll read every claim about culture with these tools in hand.
Let's start with a hunch.
1. Turn a hunch into a study
Imagine a vacant lot that neighbors turn into a community garden. A year later, a gardener tells you everybody on the block knows each other now. Maybe that's true. But "knows each other" is fuzzy, and one gardener is one point of view.
Sociologists check a hunch by running it through the research cycle: ask a question, state a hypothesis, define the variables so they can be measured, choose a method, collect data, and analyze them to see whether the hypothesis holds. A hypothesis is a testable statement of the relationship a researcher expects to find between two or more variables, written before the data come in.
The step people skip is the one that matters most. You have to operationalize your variables, which means defining each one as an exact, observable measurement. "Knows each other" becomes the number of neighbors a resident can name on a short questionnaire. Now two researchers can measure it the same way.
A hypothesis usually links two kinds of variable. The independent variable is the presumed cause. The dependent variable is the presumed effect, the thing you measure to see whether it changes. In the garden study, the hypothesis says garden members can name more neighbors than non-members. So garden membership is the independent variable, and the number of neighbors a resident can name is the dependent variable.
Pause and try this. In the garden study, which variable would change if the hypothesis is right?
Say it to yourself before I go on.
Here's the answer. The number of neighbors a resident can name. That's the outcome you measure, so it's the dependent variable, and it depends on garden membership.
2. Match the question to a method
Think about a toolbox. A hammer and a tape measure are both good tools, but you wouldn't measure a wall with a hammer. Research methods work the same way. Each one is built for a different kind of question.
A survey gathers data by asking many people the same questions, through a questionnaire or an interview. It buys you breadth. An experiment changes one variable for an experimental group, leaves it unchanged for a control group, and compares the two. It's built for cause. Field research collects firsthand data in the natural settings where people live and work, and ethnography is the long version, where the researcher joins a group's daily life over months. That buys depth. And secondary data analysis works with data someone else already collected, such as Census tables. That buys reach across time and place.
Secondary data is as old as the discipline. Émile Durkheim's book Suicide, from 1897, worked from official statistics that already existed. He showed that suicide rates differed in regular ways between social groups, and he read those rates as facts about how strongly groups bind their members.
Here's a quick way to hold the control group. It isn't the group nothing happens to. It's the comparison. Its members are handled the same way in every other respect, so a difference between the groups can be traced to the treatment.
Pause and try this. A library wants to know whether a text-message reminder makes people return books on time. Which method fits?
Ask yourself what kind of claim the library wants.
Here's the answer. An experiment. The library wants a cause, so one group gets the reminder, the control group doesn't, and you compare on-time returns.
3. Check the measuring stick
Step on a bathroom scale three times. Same number each time. You trust it. Now suppose that scale has always read two pounds heavy. It's consistent, and it's wrong every single time.
That scale pulls apart two words people blur. Reliability is the consistency of a measure: a reliable measure gives the same result each time it's used on the same thing under the same conditions. Validity is whether a measure captures what it's meant to capture. The heavy scale is reliable but not valid.
Here's a sociology version. Say a survey wants to measure how involved people are in their neighborhood, and it counts the clubs a resident has joined. Ask the same resident twice and you'll get the same count, so it's reliable. But a resident who joined no club might check on an older neighbor every day or organize a block cleanup. The club count misses that, so its validity is weak.
I want you to hold one caveat. A measure that jumps around can't be valid. But a measure that stays steady can still be aimed at the wrong thing. Reliability is needed for validity, and it is never enough.
Pause and try this. A kitchen thermometer always reads a little too warm, by the same amount every time you use it. Is it reliable, valid, both, or neither?
Run it through the two questions: same reading twice, and the right thing?
Here's the answer. Reliable but not valid. It gives the same reading each time, so it's consistent, but it misses the true temperature every time.
4. Decide whom you can speak for
Picture a sports team's fan page posting a poll: should the city build a new stadium? By evening there is a huge pile of votes, and most say yes. Does that tell you what the city thinks?
You need two words. The population is the whole group you want to draw conclusions about. The sample is the smaller set of people who actually take part. The fan page claims to speak for the city, but its sample is people who follow the page, saw the post, and chose to vote. That's a self-selected sample, and it represents only the kind of person who opts in.
The fix is a random sample, chosen so every member of the population has a known, equal chance of selection. A much smaller group of residents drawn at random from a list of every household will include people who never visit the fan page and people who don't care about sports.
The fan-page problem has a name: sampling bias, a systematic difference between a sample and its population caused by how the sample was chosen. And here's the part that surprises people. Collecting more responses the same way doesn't remove it. A bigger pile of fan votes only repeats the bias.
Pause and try this. Which speaks for the city better: a huge pile of fan-page votes, or a much smaller group of residents chosen at random?
Think about how each group was chosen, not how big it is.
Here's the answer. The random sample. Every household had a fair chance of being picked, so it looks like the city in miniature. Representative beats big.
5. Earn the word "cause"
Every summer in a beach town, two numbers climb together: ice-cream sales and drownings. Does ice cream make people drown? No. Something else is going on.
A correlation is an association between two variables that change together. Causation is the stronger claim that a change in one produces a change in the other. The beach town shows the classic trap. Heat sends more people to the ice-cream stand. Heat also sends more people into the water. Heat is the third variable, an outside factor that affects both, and the ice-cream link is a spurious correlation, an association created by that third variable.
So how does a study earn "cause"? With random assignment: placing each person into the experimental group or the control group by chance. If library patrons get the reminder by a coin flip, then busy, careful, and forgetful patrons land in both groups. A difference in returns can then be traced to the reminder. Without random assignment, or a study built to find a cause, you say "tends to come with," not "causes."
Pause and try this. A headline says neighborhoods with more coffee shops have lower crime. Correlation or cause, and what third variable might be at work?
Look for random assignment first.
Here's the answer. A correlation. Nobody assigned coffee shops at random. Income could be a third variable, because wealthier neighborhoods may attract more coffee shops and also have lower crime for other reasons.
6. Name the three traditions
Imagine two researchers at one neighborhood food pantry. One counts the households that visit each week and charts the totals over a year. The other sits in the waiting line and asks visitors what it feels like to come. Both are doing sociology, in different traditions.
Positivism studies society with the methods of the natural sciences, measuring observable facts and looking for general patterns. Auguste Comte, who gave the discipline its name, argued for a positive, meaning scientific, study of society. Interpretivism seeks to understand the meanings people give to their own actions. Max Weber called this verstehen, a German word for interpretive understanding of the meaning an action has for the actor.
The traditions tend to collect different kinds of data. Quantitative research works with numbers, such as survey shares or rates. Qualitative research works with words, observations, and meanings, such as interviews and field notes. The pantry counts show how big the need is and when it rises. The conversations show why some households wait until a crisis to come. Each answers a question the other can't.
Pause and try this. Which pantry researcher is doing qualitative research in the interpretivist tradition?
Ask which one is collecting words and meanings.
Here's the answer. The one in the waiting line. Conversations about what a visit feels like are words and meanings, which is qualitative work aimed at understanding.
7. Protect the people you study
Imagine a student planning a small study of how people juggle work and school, by interviewing coworkers at a part-time job. Before the first question, that student owes every coworker three protections.
Informed consent is a person's voluntary agreement to take part after being told the study's purpose, what it involves, its risks, and the right to stop at any time. Confidentiality means information about individual participants is kept private and never reported in a way that identifies them. And a review board is a committee that examines a proposed study involving people and must approve it as ethical before any data are collected.
Why do these rules exist? The Tuskegee syphilis study is named as the reason they exist. The Readings and Resources page links the public-health agency's own account of what happened in that study.
One mix-up comes up a lot. Confidentiality does not mean the results stay secret. Results are meant to be shared. Confidentiality protects the people, so no reader can tell who said what.
Pause and try this. In the coworker study, which protection is broken if the write-up names the one coworker with a rare job title?
Think about whether a reader could tell who said it.
Here's the answer. Confidentiality. A rare job title lets a reader identify the person, even without a name.
8. Read a real survey figure four ways
Now back to a real number. In its Mobile Fact Sheet, Pew Research Center reports that 91% of U.S. adults said they own a smartphone, and 98% said they own a cellphone of some kind, in its 2025 survey. Let's read it with the four questions from Week 1.
What's measured? Self-reported ownership. Over whom? The population is U.S. adults, and the sample is 5,022 adults surveyed in one window in 2025. What does it show? Smartphone ownership was 35% when Pew first measured it in 2011 and 91% in 2025. What doesn't it show? Why ownership rose, or what a smartphone does to anyone. Correlation or cause? Neither. It's a description of one population at two points in time.
Now the trap. The change from 35% to 91% is 56 percentage points, the unit for the difference between two percentages. As a percent change, the difference divided by the starting share and multiplied by 100, it's 160 percent. One change, two units. Write the unit every time.
Counts and rates come back too. Two illustrative towns: Town A has 12 events among 40,000 residents, which is 30 per 100,000. Town B has 15 events among 30,000, which is 50 per 100,000. You compare the rates, never the counts.
Pause and try this. Someone says smartphone ownership "rose 56 percent." What's wrong with that sentence?
Check which unit the 56 is in.
Here's the answer. The 56 is a difference in percentage points. As a percent of the 2011 share, the rise is 160 percent, so the sentence mixes up the units.
Closing
Let me pull it together. The research cycle turns a hunch into a study, and operationalizing makes a fuzzy idea measurable. Surveys buy breadth, experiments buy cause, field research buys depth, and secondary data reuses what exists. A reliable measure is consistent, and a valid one hits its target. A representative random sample beats a big self-selected one. A correlation is a pattern, and only random assignment or a design built to find a cause earns "cause." Positivism counts patterns, and interpretivism seeks meaning. Consent, confidentiality, and review protect people. And every figure gets read four ways, with its unit written out.
Next in this module come the slide deck, the Readings & Resources page, the lecture tutorial, the practice set, and the quiz, then the discussion, the assignment, and the Sociology in Action Workshop. Next week, in Week 3, the course turns to culture: the shared values, norms, and symbols that shape everyday life.
Built with The Course Maker · samples.thecoursemaker.com