Chapter 3 Notes: Doing Sociological Research
Tools and aims of sociological research
Sociological research is a true scientific endeavor, whether quantitative or qualitative, rigorously applying systematic methods to understand social phenomena.
Aim: To systematically observe, describe, explain, predict, and ultimately understand patterns of human behavior in social contexts.
This involves developing and testing theories about how societies function and change, moving beyond common sense or anecdotal evidence to rely on empirical data and structured analysis. It seeks not just to identify what happens, but why and how.
Systematic Methods: These include carefully designed procedures for data collection (e.g., surveys, interviews, observations, experiments) and rigorous methods for data analysis (e.g., statistical tests, thematic analysis, content analysis), ensuring that findings are reliable and valid.
The steps in research design are critically linked to the sociological question being asked, dictating the most appropriate methodology.
Sociologists utilize a diverse array of research tools, each possessing distinct advantages and disadvantages suitable for different types of questions.
Quantitative tools (e.g., surveys, statistical analysis of large datasets) are often used to identify broad patterns, correlations, and generalizable trends across populations.
Qualitative tools (e.g., in-depth interviews, ethnography, participant observation, focus groups) are employed to explore nuanced meanings, lived experiences, and contextual understandings within specific groups or situations.
Professional ethics, including informed consent, confidentiality, and avoiding harm, play a pivotal role throughout the entire research process, ensuring responsible and respectable inquiry.
Researchers must ensure participants fully understand the nature of the study and voluntarily agree to participate (informed consent), protect their anonymity and privacy (confidentiality), and take all possible measures to prevent any physical, psychological, or social harm. Ethical review boards (Institutional Review Boards - IRBs) are often mandated to oversee this process.
Introduction: scope and questions in sociological research
Sociologists investigate a vast and complex range of social phenomena (e.g., cultural practices, the pervasive issue of homelessness, patterns of immigration, social inequality, digital citizenship, social movements, family structures, gender roles, urban development, crime and deviance, globalization, education systems, and health disparities).
This broad scope reflects the intertwined nature of human interactions and societal structures, aiming to uncover the underlying social forces that shape individual and collective lives.
Research questions are the foundational guides that shape the choice of method and overall design; they define what needs to be studied and how.
Well-formulated research questions are clear, focused, feasible, and sociologically relevant, moving beyond simple factual queries to explore relationships, causes, and meanings.
Example questions about homelessness illustrate this specificity:
What is the daily lived experience for homeless people, encompassing their routines, challenges, and coping mechanisms? (Suggests qualitative methods like ethnography).
How dangerous is it to be homeless, considering physical threats, health risks, and lack of safety nets? (Might involve both qualitative data on experiences and quantitative data on crime rates or health outcomes).
Where are homeless people typically found, and what are the geographic and structural factors influencing their presence in certain areas? (Could involve spatial analysis, census data, and policy review).
Do they interact with each other, and what forms do these social networks or communities take, if any? (Best explored through observational studies and in-depth interviews).
Do they engage in work, and if so, what kind of work and under what conditions, including formal, informal, or survival strategies? (Requires detailed qualitative inquiry and possibly economic data).
All research methods, regardless of their specific approach, ultimately aim for a deeper, evidence-based understanding of how society operates, its structures, processes, and implications for individuals.
Sociological research serves as an essential tool to answer specific questions and rigorously test hypotheses about social life.
The scientific method in sociology
Sociology, as a science, systematically adheres to the scientific method—a structured approach involving observation, hypothesis formulation, data collection, empirical testing, data analysis, and drawing conclusions—as originally described by Francis Bacon.
This method provides a logical framework for inquiry, ensuring objectivity and reducing bias in the pursuit of knowledge.
The steps in the research process flow logically from the initial research question, forming an iterative cycle (refer to Figure 3.1 in the text for a visual representation):
Observation: Identifying patterns or anomalies in social life that spark curiosity or signal a research problem. This can be informal (everyday experiences) or formal (reviewing existing literature, preliminary data scans).
Hypothesis formulation/Question refining: Developing a specific, testable statement (hypothesis) about the relationship between variables or a clear, focused research question if an exploratory approach is taken. This step is guided by existing theory or prior observations.
Research Design and Data Collection: Selecting appropriate methods (e.g., survey, experiment, interview, content analysis) and gathering relevant data systematically to test the hypothesis or answer the research question. This involves defining the population of interest, sampling strategy, and measurement tools.
Analysis of data: Systematically processing and interpreting collected data, using appropriate statistical (for quantitative data) or qualitative (for qualitative data) techniques to identify patterns, themes, or relationships.
Drawing conclusions: Interpreting the findings in relation to the initial hypothesis and broader theoretical frameworks, discussing their implications, limitations, and suggesting avenues for future research. This often leads back to new observations or refined hypotheses, continuing the iterative cycle.
Deductive reasoning (top-down approach): This process starts with a general theory, principle, or existing knowledge to formulate a specific, testable research question or hypothesis.
Mechanism: Researchers use established theories or widely accepted principles to predict what they should observe in specific cases. If the observations align with the predictions, the theory is strengthened; if not, the theory may need revision.
Example: If Catholic doctrine strongly forbids abortion (general principle founded in religious theory), then it can be hypothesized that Catholics would, on average, be less likely to support abortion rights (specific testable question). Researchers would then test this hypothesis, e.g., with a survey measuring religious affiliation and abortion attitudes. Interestingly, findings may sometimes contradict the initial expectation, leading to theory refinement or re-examination of assumptions (e.g., some Catholics may be more likely to support abortion rights due to other social factors like political alignment or personal experiences, indicating the general principle doesn't fully capture social reality).
Represented conceptually as: moving from a broad general principle to a focused, empirically testable question.
Notation:
\text{General principle: a theory or belief} \rightarrow \text{Testable question: derived from that principle} \rightarrow \text{Empirical observation/Data} \rightarrow \text{Confirm or modify principle}
Inductive reasoning (bottom-up approach): This process begins with specific empirical observations or data points and, through analysis, derives broader generalizations or general conclusions.
Mechanism: Researchers gather data without a preconceived hypothesis, identify patterns or themes within that data, and then formulate a theory or general principle to explain those patterns. It's often associated with qualitative research and theory generation.
Example: Observing specific interactions within a homeless community (specific observations like sharing food, guarding possessions, creating informal rules) might lead to a broader sociological insight about the formation of informal social support networks and subcultures among marginalized groups (general conclusion, potentially leading to a new theory about social cohesion under duress).
Critical for generating new theories or refining existing ones based on real-world evidence, allowing for the discovery of unexpected patterns.
Both deductive and inductive approaches are valid and often complementary parts of sociological research, guiding different stages of inquiry. A researcher might use induction to generate a theory from observations and then deduction to test that theory in another context.
Conclusion: Sociological research can initiate from diverse starting points, including established theory, findings from prior studies, or direct empirical observations. The subsequent design and data collection procedures are directly determined by the nature of the chosen research question and approach.
Research design and questions
Research design constitutes the systematic and strategic organization of a study’s questions, methods, and procedures to effectively investigate social phenomena, ensuring logical and efficient data collection.
It serves as a blueprint for the study, guiding decisions on what data to collect, from whom, how, and when. A well-constructed design maximizes the reliability and validity of findings while minimizing potential biases.
The first and most crucial step is to develop a clear, precise, and researchable question.
Sources include: reviewing past research and identifying gaps (e.g., studies that only focus on one gender or ethnicity), puzzling observations in society (e.g., a sudden rise in a particular social issue), scrutinizing conclusions from prior research that might be oversimplified or misinterpreted, or considering current social problems and public debates.
Researchers must critically scrutinize how previous conclusions were drawn, including methodologies, data sources, and potential biases, to refine or challenge existing knowledge. This involves asking: Who conducted the research? What methods were used? What assumptions were made?
Hypotheses are specific, testable statements derived from a research question, proposing an expected relationship between two or more variables. They are not proofs of fact but educated guesses that guide empirical investigation.
Characteristics of a good hypothesis: It must be clear, concise, testable through empirical observation, and state an expected relationship between variables.
Example: If a person’s parents hold strong racially prejudiced views (independent variable), then that person will, on average, exhibit more prejudiced attitudes themselves (dependent variable). This statement is a hypothesis, requiring empirical testing, not an an established fact. Researchers often formulate a null hypothesis (H0 - no relationship between variables) and an alternative hypothesis (H1 - a specific relationship exists).
Key research concepts:
Variables: Quantities or attributes that vary across individuals or cases and can be measured. Variables can be categorized by their level of measurement, which determines the appropriate statistical analyses:
Nominal: Categorical variables with no inherent order (e.g., gender (male, female, non-binary), race (Asian, Black, White, etc.), marital status (single, married, divorced)). Numbers assigned to these categories are purely for labeling.
Ordinal: Categorical variables with a meaningful order, but the intervals between categories are not necessarily equal or quantifiable (e.g., social class (lower, middle, upper), educational attainment (high school, college, graduate), Likert scale responses (strongly disagree, disagree, neutral, agree, strongly agree)). We know 'upper' is higher than 'middle,' but not by how much.
Interval: Numerical variables with equal intervals between values, but no true or meaningful zero point (e.g., temperature in Celsius/Fahrenheit – 0°C does not mean an absence of temperature; IQ scores). Ratios are not meaningful with interval data.
Ratio: Numerical variables with equal intervals and a true zero point, meaning that zero signifies the absence of the quantity (e.g., age, income, number of children, height, weight). Ratio data allows for meaningful ratios (e.g., someone earning 20,000 is earning half of someone earning 40,000).
Independent variable (X): The presumed cause or explanatory variable that is manipulated or observed to have an effect on another variable. It is the variable that the researcher changes or observes changes in, to see if it influences the dependent variable. Notation: X.
Dependent variable (Y): The outcome or effect variable, whose changes are presumed to be caused by the independent variable. It is the variable that is measured and is expected to change in response to the independent variable. Notation: Y.
Intervening variable (Z): A variable that explains the causal link or mechanism between the independent variable (X) and the dependent variable (Y). It is affected by X and in turn affects Y. It clarifies how X influences Y. For example, if X is education and Y is income, Z could be job skills acquired through education; education (X) leads to job skills (Z), which then leads to higher income (Y).
Concepts: Abstract theoretical constructs or attributes that are not directly observable (e.g., social class, social power, quality of life, religious fundamentalism, anomie, cultural capital, social cohesion). They require careful conceptual definition to be understood within a theoretical framework.
Indicators: Observable and measurable phenomena that reflect or represent an abstract concept. They serve as the operational definition for a concept, allowing it to be empirically studied (e.g., education level, income, and life expectancy are indicators for the concept of quality of life; attendance at religious services or frequency of prayer could be indicators for religious fundamentalism). Multiple indicators are often used to capture the complexity of a single concept.
Example: United Nations Human Development Index (HDI)
The HDI is a composite index designed to measure a country's average achievements in three basic dimensions of human development: health, knowledge, and standard of living. It serves as a classic example of using multiple indicators to measure a complex concept like human development.
Health is indicated by life expectancy at birth.
Knowledge is indicated by mean years of schooling and expected years of schooling.
Standard of living is indicated by gross national income (GNI) per capita.
By combining these distinct, measurable indicators, the HDI provides a robust numerical representation of a multidimensional and abstract concept, allowing for comparisons between countries and over time.