Chapter 2: Examining Our Social World

Chapter 2: Examining Our Social World

This chapter introduces sociological research and explains how systematic investigation helps us understand the social world. It focuses on how people develop knowledge, why sociological research matters in everyday life, and the role of the scientific method in studying society.

Chapter objectives

By the end of the chapter, students should be able to:

1. Compare knowledge based on tradition, authority, and research.

2. Explain why sociological research is important in everyday life.

3. Describe the scientific method.

4. Describe the basic steps of the research process.

5. Compare the five most common sociological data-collection methods, including their strengths and limitations.

6. Explain why ethics are important in scientific research.

How Do We Know What We Know?

People gain knowledge in several different ways. Sociology distinguishes between knowledge based on traditions, knowledge accepted from authorities, and knowledge developed through systematic research.

Sources of knowledge

These sources differ in how knowledge is justified. Tradition relies on what has been passed down, authority relies on an accepted source or figure, and research relies on systematic evidence-gathering.

Applying the three sources

The following examples illustrate the difference:

1. “Sam believes it is true because the Bible says so.”
   This is classified as tradition in the knowledge check.

2. “Nine out of 10 people who answered the survey said that they were in favor of the change.”
   This is classified as science because it is based on survey data.

3. “A good spanking always worked for my children.”
   This is classified as authority in the knowledge check because the statement is presented as an accepted personal claim rather than the result of systematic research.

Why Sociological Research Matters in Everyday Life

Sociological research examines patterns of social behaviour, relationships, institutions, and social problems. It is useful not only to researchers but also to people making decisions in everyday life.

Sociological research:

• Counteracts misinformation by testing claims against evidence.

• Exposes myths by showing when commonly held beliefs are inaccurate or unsupported.

• Helps explain why people behave as they do by examining social influences and patterns.

• Affects social policies by providing evidence that can guide decisions about social programmes and public issues.

• Sharpens critical-thinking skills by encouraging people to question claims, evaluate evidence, and consider how research was conducted.

Research reported in the news should therefore not be accepted uncritically. Readers should consider how the research was conducted and whether the evidence supports the conclusions.

The Scientific Method

The chapter introduces the scientific method as a systematic approach to developing knowledge. The scientific method is connected to the broader research process and provides a framework for investigating social questions using evidence.

The supplied material identifies the scientific method as a major topic, but does not provide its specific stages or procedures. Later sections of the chapter are intended to cover:

• The scientific method.

• The basic steps of the research process.

• Common sociological data-collection methods.

• The strengths and limitations of different research methods.

• The ethical principles that guide scientific research.

Scientific Method

The scientific method is a body of objective and systematic techniques used to investigate phenomena, acquire knowledge, and test hypotheses and theories. In sociology, it provides a structured way to study social behavior and relationships using evidence rather than personal assumptions or opinions.

Elements of the Scientific Method

Concepts

A concept is an abstract idea, mental image, or general notion that represents some aspect of the world. Examples of sociological concepts include social class, prejudice, education, family, and inequality.

Concepts help researchers identify and describe the subject they want to study. Because concepts can be broad or abstract, researchers often need to define them carefully before collecting data.

Variables

A variable is something that changes in value or magnitude under different conditions. Researchers examine variables to determine whether they are related to one another.

The main types of variables are:

• Independent variable: The variable that is expected to influence or cause a change in another variable.

• Dependent variable: The variable that is expected to be affected or changed by the independent variable.

• Control variable: A factor that researchers hold constant or take into account so that it does not distort the relationship between the independent and dependent variables.

For example, a researcher might ask whether the amount of time students spend studying affects their exam performance.

• Independent variable: study time

• Dependent variable: exam performance

• Possible control variables: age, prior academic achievement, course difficulty, or access to learning resources

Hypotheses

Research begins with a hypothesis. A hypothesis is a statement of the expected relationship between two or more variables.

A hypothesis should be clear enough to be tested through the collection and analysis of data. For example: Students who spend more time studying will achieve higher exam scores.

Applying Variables to a Sociological Research Question

A useful research question should identify a social issue or behavior that can be studied systematically. Researchers should then determine which variables are involved.

A research activity can be completed through the following process:

1. Work with a partner to create a sociological research question of interest.

2. Identify the independent variable or variables.

3. Identify the dependent variable or variables.

4. Identify important control variables.

5. Share the example with the class for discussion and comparison.

This process helps researchers translate a general topic into a question that can be measured and investigated.

Deductive and Inductive Reasoning

Researchers use two main forms of reasoning to connect observations, data, hypotheses, and theories.

Deductive Reasoning

Deductive reasoning begins with a theory, prediction, or general principle. Researchers then collect data to test whether the prediction is supported.

The general pattern is:

1. Begin with a theory or general principle.

2. Develop a specific prediction or hypothesis.

3. Collect relevant data.

4. Test the hypothesis.

5. Decide whether the evidence supports the prediction.

Deductive reasoning moves from the general to the specific. For example, a theory about social inequality might lead a researcher to predict that students from higher-income families have greater access to educational resources.

Inductive Reasoning

Inductive reasoning begins with a specific observation. Researchers then collect data, identify patterns or irregularities, and formulate hypotheses that may contribute to the construction of a theory.

The general pattern is:

1. Make a specific observation.

2. Collect additional data.

3. Identify patterns or irregularities.

4. Formulate a hypothesis.

5. Develop or contribute to a broader theory.

Inductive reasoning moves from the specific to the general. For example, after observing repeated differences in educational opportunities among social groups, a researcher might develop a hypothesis about the relationship between social class and access to education.

Reliability and Validity

Researchers must evaluate the quality of the measures they use to collect data.

Reliability

Reliability is the consistency with which the same measure produces similar results over time.

A reliable measure gives stable results when it is used repeatedly under similar conditions. For example, a survey designed to measure political attitudes should produce broadly similar results if the same person completes it again shortly afterward, assuming their attitudes have not changed.

Validity

Validity is the degree to which a measure is accurate and actually measures what it claims to measure.

For example, a measure intended to assess academic ability has validity if it genuinely measures academic ability rather than reading speed, test anxiety, or familiarity with a particular testing format.

Reliability and validity are related but different:

• A measure can be reliable without being valid if it produces consistent results but measures the wrong thing.

• A strong research measure should be both reliable and valid.

An analogy is a clock that is always ten minutes slow. It is reliable because it gives the same incorrect time consistently, but it is not valid because it does not accurately measure the actual time.

Sampling

Researchers usually cannot study every member of the group they are interested in. Instead, they select a sample.

Population

A population is a well-defined group of people or things that researchers want to know something about.

For example, the population in a study of university students’ attitudes could be all students enrolled at a particular university.

Sample

A sample is a group selected to represent the population that researchers intend to study.

The quality of a sample affects whether findings can be generalized to the wider population. A sample that accurately reflects the population is more useful than one that overrepresents particular groups.

Probability Samples

In a probability sample, each person or thing in the population has an equal chance of being selected because selection is random.

Probability sampling can produce a more representative sample and reduce the risk that the researcher’s preferences or convenience will influence who is included.

Nonprobability Samples

In a nonprobability sample, there is little or no attempt to obtain a representative cross-section of the population.

These samples may be easier, faster, or less expensive to obtain, but they can produce biased results. For example, surveying only people who are easy to reach may exclude important sections of the population.

Time Dimension Studies

Researchers can compare variables across time using either cross-sectional or longitudinal studies.

Cross-Sectional Studies

A cross-sectional study collects data at one point in time. It provides a snapshot of behavior, attitudes, or social conditions during a particular period.

Cross-sectional studies can provide valuable information and are often relatively quick and less expensive to conduct. However, because they examine data at only one point in time, they cannot show how behavior or attitudes change over time as effectively as longitudinal studies.

Possible limitations include:

• They may show that variables are associated without demonstrating how the relationship develops.

• They cannot easily establish whether a change occurred before or after another factor.

• They may be affected by unusual events occurring at the time the data were collected.

Longitudinal Studies

A longitudinal study collects data at two or more points in time. Researchers may use the same sample repeatedly or use different samples at each time point.

Longitudinal studies are used to examine trends in behavior or attitudes. They can show whether social conditions remain stable, change gradually, or change rapidly.

Advantages include:

• Tracking changes over time.

• Identifying trends in behavior or attitudes.

• Showing the sequence in which changes occur.

• Providing a clearer picture of development and long-term relationships between variables.

Possible limitations include:

• They usually require more time and resources.

• Participants may drop out before the study is complete.

• Repeated questioning may influence participants’ responses.

• Social conditions may change in ways that make comparisons more difficult.

• If different samples are used at different times, changes may reflect differences between the samples rather than genuine changes in the population.

Key Study Comparisons

Several pairs of concepts are especially important:

• Independent variable versus dependent variable: The independent variable is expected to influence the dependent variable.

• Reliability versus validity: Reliability concerns consistency, while validity concerns accuracy.

• Population versus sample: The population is the entire group of interest, while the sample is the smaller group studied.

• Probability versus nonprobability sample: Probability sampling uses random selection, while nonprobability sampling makes little or no attempt to create a representative cross-section.

• Cross-sectional versus longitudinal study: Cross-sectional studies provide a snapshot, while longitudinal studies examine change across time.

• Deductive versus inductive reasoning: Deductive reasoning tests predictions based on existing theories, while inductive reasoning develops broader conclusions from specific observations.

Basic Steps in the Research Process

Sociological research follows a structured process for developing questions, collecting evidence, analyzing information, and communicating conclusions. Researchers may use qualitative or quantitative approaches, depending on the type of information needed.

Research approaches

Qualitative research examines and interprets nonnumerical material. It is useful for understanding meanings, experiences, opinions, and social behavior in depth.

Quantitative research focuses on the numerical analysis of people’s responses or specific characteristics. It is useful for identifying patterns, comparing groups, and measuring relationships between variables.

Causation and Correlation

Causation

Causation is a relationship in which one variable is the direct consequence of another. In a causal relationship, a change in one variable produces a change in the other.

Causation is difficult to prove because other variables may influence both factors. Researchers must therefore consider alternative explanations and determine whether the proposed cause actually produces the outcome.

Correlation

Correlation is a relationship between two or more variables. When variables are correlated, they change in relation to one another, but this does not necessarily mean that one causes the other.

A correlation may occur because:

• One variable causes the other.

• The variables are both influenced by a third variable.

• The relationship is coincidental.

• The direction of influence is unclear.

Key principle: Correlation does not automatically demonstrate causation.

Independent and dependent variables

In a proposed causal relationship:

• The independent variable, or determinant, is the factor thought to influence another variable.

• The dependent variable, or outcome, is the factor that may be affected.

For example, alcohol misuse may be proposed as an independent variable related to domestic violence as a dependent variable. However, this relationship may also be influenced by other factors, such as:

• Unemployment

• Stress

• Depression

• Low income

• Poverty

• Growing up in an abusive household

• Other substance misuse

The presence of these additional factors means that a relationship between alcohol misuse and domestic violence should not automatically be interpreted as proof that alcohol misuse directly causes domestic violence.

Examples of correlation

Firetrucks and fires

Cities with more firetrucks have a significantly higher average number of fires each year. This is a correlation, but it does not mean that having more firetrucks causes more fires.

A more likely explanation is that larger cities have more people, buildings, and potential fire risks. They may therefore experience more fires and require more firetrucks. The number of fires and the number of firetrucks are both related to the size and needs of the city.

Ice cream sales and murders

When ice cream sales increase, the number of murders also significantly increases. This does not mean that buying or eating ice cream causes murder.

A third variable, such as warmer weather, may explain the correlation. Hot weather can increase ice cream sales, and it may also lead to more people spending time outdoors, creating more opportunities for interaction and crime. The correlation therefore does not establish a direct causal relationship between ice cream sales and murders.

These examples show why researchers must look beyond a statistical relationship and investigate possible third variables and alternative explanations.

The basic research process

Sociological research generally involves seven connected steps.

1. Choose a topic to study
   Select a social issue or phenomenon that can be investigated through systematic research.

2. Summarize the related research
   Review existing studies and scholarship about the topic. This helps researchers understand what is already known and identify gaps or unanswered questions.

3. Formulate a hypothesis or ask a research question
   A hypothesis is a testable statement about the expected relationship between variables. Alternatively, researchers may develop an open-ended research question when they are exploring a topic.

4. Describe the data collection method or methods
   Explain how information will be gathered. The methods should fit the research question and may use qualitative, quantitative, or combined approaches.

5. Collect the data
   Gather the relevant information from participants, observations, documents, surveys, or other sources.

6. Analyze the data
   Examine the collected information to identify patterns, relationships, and evidence that address the hypothesis or research question.

7. Present and explain the results
   Communicate the findings and explain what they mean. Researchers should discuss whether the evidence supports the hypothesis, answer the research question, and acknowledge possible limitations or alternative explanations.

Key takeaway

Research provides a systematic way to study society. Qualitative research interprets nonnumerical information, whereas quantitative research analyzes numerical information. Researchers must distinguish carefully between correlation, which shows that variables are related, and causation, which means that one variable directly produces a change in another.

Some Major Data Collection Methods

Sociologists use several methods to collect data about social life. The main methods are:

• Surveys

• Field research

• Content analysis

• Experiments

• Secondary analysis of existing data

• Evaluation research

Each method has different strengths and limitations. The most appropriate method depends on the research question, the type of information needed, the resources available, and whether the researcher needs to study people’s behavior, attitudes, communication, existing records, or the effects of a social intervention.

Surveys

A survey is a data collection method that gathers information through questionnaires, interviews, or a combination of both.

Sampling

An important part of survey research is selecting a sample. A sample is a smaller group selected from a larger population. Researchers generally prefer random samples because they are more likely to represent the larger population. This allows findings from the sample to be generalized to the population as a whole.

Questionnaires and interviews

A questionnaire is a series of written questions that asks respondents for information. An interview involves directly asking a person a series of questions. Interviews may be conducted face to face or by telephone.

Surveys and interviews can also be combined. For example, researchers might use a written questionnaire to collect standardized information and then conduct interviews to explore some responses in greater depth.

Strengths and limitations

Surveys are generally:

• Inexpensive

• Simple to administer

• Fast to complete and analyze

However, different forms of survey research have particular disadvantages:

• Face-to-face interviews can be very expensive.

• Mailed questionnaires often have low response rates.

• Respondents may provide answers they believe are socially acceptable rather than completely accurate answers. This is known as social desirability bias.

Surveys are especially useful when researchers want comparable information from many people, such as opinions about a public policy, patterns of media use, or levels of satisfaction with a service. Interviews are more useful when the research topic requires detailed explanations, personal experiences, or follow-up questions.

Field Research

Field research collects data by observing people in their natural surroundings rather than in an artificial research setting. Researchers study people as they behave in everyday environments, such as workplaces, neighborhoods, schools, or community organizations.

Types of observation

Field research may use:

• Participant observation, in which the researcher takes part in the activities of the group being studied.

• Nonparticipant observation, in which the researcher observes without becoming an active member of the group.

Field research can involve short-term observation or longer-term ethnographies. An ethnography is an in-depth study of a group or community based on extended observation and, often, interaction with group members.

Strengths and limitations

The main strength of field research is that it provides detailed and valid information about how people behave in real social settings. It can reveal aspects of social life that people might not remember, explain, or report accurately in a questionnaire.

Its main limitations are that it can be expensive and time-consuming. Extended observation requires considerable researcher effort, and studying a small group in depth may make it difficult to generalize findings to a much larger population.

Content Analysis

Content analysis systematically examines some form of communication. It can be used with written or oral material, including documents, speeches, news reports, advertisements, interviews, and other forms of communication.

Main steps

Content analysis generally involves the following process:

1. Identify the communication or material to be studied.

2. Develop categories for coding the material.

3. Sort the material according to those categories.

4. Analyze the coded data.

5. Draw conclusions from the results.

For example, researchers could examine news articles to study how different social groups are represented. They would create coding categories, classify relevant parts of the articles, and then analyze patterns in the results.

Strengths and limitations

Content analysis is:

• Inexpensive

• Less time-consuming than some other methods

• Unobtrusive, because researchers do not need to interact directly with the people who produced the material

• Relatively easy to correct if coding errors are discovered, since the material can be recoded

However, content analysis can also be labor intensive, particularly when large amounts of communication must be coded. Coding may be subjective because researchers must decide how to classify material. The categories and interpretations may also reflect social class bias.

Applying content analysis

To use content analysis, researchers should begin with a sociological research question and then identify communication that could provide evidence for answering it. For example, a researcher interested in gender stereotypes might examine advertisements, television programs, or social media posts and code the roles assigned to different genders.

Experiments

An experiment takes place in a controlled, artificial situation. Researchers manipulate an independent variable to determine whether it produces a change in a dependent variable.

Experimental and control groups

Experiments commonly compare two groups:

• Experimental group: Participants are exposed to the independent variable.

• Control group: Participants are not exposed to the independent variable.

The two groups are then compared to determine whether exposure to the independent variable produced a difference in the dependent variable.

Basic experimental design

A basic experimental design may follow these stages:

1. Measure the dependent variable in both groups during a pretest.

2. Expose the experimental group to the independent variable.

3. Do not expose the control group to the independent variable.

4. Measure the dependent variable again during a posttest.

5. Compare the groups to determine whether their results differ.

For example, a researcher might test the hypothesis that watching a film about racial discrimination reduces prejudice. The experimental group would watch the film, while the control group would not. Researchers would measure prejudice before and after the film and compare the results.

Strengths and limitations

Experiments are useful because they can identify cause-and-effect relationships. They can also:

• Involve volunteers

• Often be conducted without substantial payment

• Be replicated with participants from diverse backgrounds

The major limitation is that laboratory experiments are artificial. Participants may behave differently in a laboratory than they would in everyday life. As a result, findings may not generalize to larger populations. Laboratory experiments are also usually unsuitable for studying very large groups.

Secondary Analysis of Existing Data

Secondary analysis examines information collected by someone else. Instead of gathering new data, the researcher uses existing materials to answer a new research question or to investigate the original question in a different way.

Sources of secondary data

Secondary data may include:

• Historical materials

• Personal documents

• Public records

• Official statistics

Examples include government population statistics, archived letters, census records, court documents, organizational records, and previously conducted surveys.

Strengths and limitations

Secondary analysis is:

• Accessible

• Convenient

• Inexpensive

It allows researchers to study large amounts of information without having to collect the data themselves.

However, researchers have limited control over the data. The exact information needed may not be available, and the original researchers may have used different measures or definitions for the variables. Researchers also cannot change how particular variables were measured or add new questions to an already completed study.

Evaluation Research

Evaluation research determines whether a social intervention has produced its intended result. A social intervention may be a policy, program, service, or activity designed to change a social problem or improve people’s lives.

Evaluation research focuses more on the purpose of the research than on one particular method. It may use surveys, interviews, field research, experiments, secondary data, qualitative approaches, quantitative approaches, or a combination of methods.

Purpose and uses

Evaluation research is intended to have a real-world effect. Its results can help organizations and policymakers:

• Determine whether a policy or program is effective

• Improve the efficiency of a program or group

• Improve the effectiveness of an intervention

• Decide how funding should be allocated

For example, researchers might evaluate whether a community health program improves access to medical care or whether an educational intervention improves student outcomes.

Strengths and limitations

Evaluation research:

• Helps policymakers make informed decisions about funding

• Can combine qualitative and quantitative approaches

• Addresses practical, real-life problems

Its limitation is that it can examine only a limited number of factors affecting behavior. Human behavior is influenced by many social, economic, cultural, and personal conditions, so it may be difficult to determine all the reasons why an intervention succeeds or fails.

Comparing Data Collection Methods

Key Study Points

• Surveys ask people questions, usually through questionnaires or interviews.

• Random samples are preferred because they make it more reasonable to generalize findings to a larger population.

• Field research studies people in natural settings through participant or nonparticipant observation.

• Content analysis studies communication by coding and analyzing material systematically.

• Experiments compare an experimental group with a control group to investigate cause and effect.

• Secondary analysis uses data previously collected by another person or organization.

• Evaluation research assesses whether a policy, program, or intervention achieved its intended outcome.

• No single method is best for every research question. Researchers choose methods according to the type of evidence required and the practical limitations of the study.

Ethics and Social Research

Ethics are principles that guide researchers in conducting studies responsibly, protecting participants, maintaining scientific integrity, and presenting findings honestly. Ethical research requires consideration of both the people involved in a study and the reliability of the knowledge produced.

Basic Principles of Ethical Sociological Research

Protecting participants

Researchers must obtain participants’ consent before involving them in a study. Participants should also give permission before researchers quote their responses or comments.

Researchers must not exploit participants or research assistants for personal benefit. Participation should never involve physical or psychological harm, humiliation, abuse, or coercion.

Researchers must respect any promises made about participants’ privacy, anonymity, and confidentiality:

• Privacy concerns participants’ control over personal information and involvement in the research.

• Anonymity means that participants’ identities are not revealed.

• Confidentiality means that information provided by participants is protected and not disclosed improperly.

Maintaining research quality and honesty

Researchers should use the highest methodological standards possible and aim to be accurate. This includes collecting and analysing evidence carefully rather than allowing personal interests or outside influences to affect the results.

Published research reports should explain the study’s limitations and shortcomings. Acknowledging weaknesses allows readers to judge how confidently the findings can be applied and prevents researchers from presenting conclusions as more certain than the evidence allows.

Researchers must also identify the sponsors or organisations that funded the study. Disclosing funding sources helps readers recognise possible conflicts of interest and evaluate the research more critically.

The contributions of research assistants must be acknowledged. This gives proper credit to people who helped conduct the research project and promotes fairness in academic work.

Scientific Misconduct

Scientific misconduct occurs when researchers deliberately violate standards of honesty, accuracy, or professional responsibility. Examples include:

• Changing research results to satisfy the corporation or organisation that sponsored the research.

• Endorsing particular drugs for financial gain even though the drugs do not reduce health problems.

• Allowing drug manufacturers to ghost-write articles that are then published in medical journals under another person’s name.

• Falsifying data or fabricating information.

These actions can mislead the public, damage scientific credibility, influence medical decisions unfairly, and undermine trust in research.

External Pressure on Researchers

Researchers may face pressure from outside groups, institutions, sponsors, or the public. This pressure is especially likely when research examines sensitive or controversial issues.

Research findings may challenge personal attitudes, values, and beliefs that people strongly cherish. As a result, individuals or organisations may reject, criticise, or attempt to discredit findings simply because they disagree with them.

Scientists also often do not participate directly in public policy debates or explain research findings to the wider public. When researchers do not communicate their findings clearly, public discussions may be shaped by misunderstandings, personal opinions, or selective interpretations of the evidence.

Key idea

Ethical sociological research requires researchers to protect participants, report methods and limitations honestly, disclose funding, recognise collaborators, and resist pressures that could distort findings. These practices help preserve both the rights of research participants and the credibility of sociology.