Sociological Research Methods
Chapter 2: Methods of Research
2.1 Types of Data, Methods, and Research Design
Primary Data
Definition: Information collected personally by a researcher.
Methods: Questionnaires, interviews, observational studies.
Strengths:
Complete control over data collection (how, by whom, for what purpose).
Greater control over reliability, validity, and representativeness.
Limitations:
Time-consuming to design, construct, and carry out.
Can be expensive.
Difficulty gaining access to the target group.
Potential respondents may refuse to participate or are deceased (in historical research).
Secondary Data
Definition: Data that already exists (documents, reports, statistics, previous research).
Strengths:
Saves time, money, and effort.
May be the only available resource (e.g., researching suicide).
Useful for historical and comparative purposes.
Official statistics may be highly reliable due to consistent collection methods.
Limitations:
Not always produced with the needs of sociologists in mind; definitions may differ.
Sources (e.g., personal documents) can be unreliable.
Historical documents may reflect individual views rather than wider opinions.
Quantitative Data
Definition: Information expressed numerically.
Forms:
Raw number.
Percentage (people per 100 in a population).
Rate (people per 1,000 in a population).
Strengths:
Useful for comparing numbers and summarizing information.
Statistical comparisons and correlations can test hypotheses.
Longitudinal studies can track changes over time.
More reliable due to easier replication.
Easier for researchers to remain objective.
Limitations:
May place respondents in an artificial social setting, affecting responses.
Captures a narrow range of information (the 'who, what, when, and where').
May lack depth and not reveal the reasons for behavior; often seen as surface level and superficial.
Qualitative Data
Definition: Aims to capture the quality of people’s behavior by exploring the ‘why’ rather than the ‘what, when and where’.
Examples:
Venkatesh (2009) studied a young gang from the viewpoint of its members.
Goffman (1961) examined the experiences of patients in a mental institution.
Strengths:
Allows for capturing complex reasons for behavior.
Researchers can study people in their 'normal' settings.
More likely to show how people really behave and what they really believe.
Limitations:
Focuses on small groups, limiting wider application; groups may not be representative.
Difficult to compare data across time and location.
Difficult to replicate, resulting in lower reliability than quantitative research.
Strengths and Limitations of Secondary Data Sources
Official Statistics:
Strengths:
May be the only available source covering a particular area (e.g., suicide).
Readily available and often based on large samples.
Some statistics ('hard statistics') have high accuracy (e.g., divorce rates).
Limitations:
Lack depth or detail.
Validity issues due to government inclusions/exclusions.
May only give a partial picture.
Statistical data may not reveal the reasons for people’s behavior.
Significance must be interpreted by researchers.
Changes in definitions can bring reliability into question.
Personal Documents, Digital Content, and Media Sources:
Strengths:
Access to data that would cost a lot of money, time and effort to collect personally.
Provide qualitative data of great depth and detail.
Can be used for comparative purposes.
Reveal hidden meanings and insights into the hopes, fears, and beliefs of those who produced them.
Limitations:
Practical limitations: can be difficult to find available sources and to ascertain sources.
Paper documents can be faked.
May be incomplete, inaccurate, or unrepresentative.
Digital sources can be subject to change.
Quantitative Research Methods
Questionnaires
Types:
Postal questionnaires (completed in private).
Researcher-administered questionnaires (structured interviews).
Question Types:
Closed-ended/pre-coded questions.
Example:
Do you own a sociology textbook?
Yes[1] No[2] Don't Know[3]
Open-ended questions.
Strengths:
Pre-coded questions make data easier to quantify.
Efficient for contacting large numbers of people quickly.
Can result in highly reliable data due to standardized questions.
Anonymity can improve validity.
Limitations:
Low response rate can lead to unrepresentative samples.
Difficult to examine complex issues.
Researcher decides significance at the start.
Researchers can't verify the understanding or honesty of responses.
Potential for biased questions.
Structured Interviews
Definition: Researcher asks questions to respondents in person, with the same questions asked in the same order each time.
Strengths:
Potential reliability problems can be fixed by the researcher.
Avoids the problem of unrepresentative samples, response rates will be 100\%.
Limitations:
Involve assumptions/pre-judgements about people's behavior.
Can contain biased questions.
Lack of anonymity can lead to the interview effect (respondents trying to please the researcher).
Researcher effect (relationship between researcher and participant may bias responses).
Experiments
Definition: Testing the relationship between different variables under controlled conditions. The researcher changes (manipulates) independent variables to see whether they produce a change in dependent variables that are not changed by the researcher.
Variables:
Independent variables: Manipulated by the researcher.
Dependent variables: Measured for changes.
Relationships:
Correlations: Things that happen at roughly the same time.
Causation: One action always follows another.
Separating Correlation from Causality:
Test and retest a relationship.
Use different groups with exactly the same characteristics: an experimental group whose behaviour is manipulated; a control group whose behaviour is not manipulated.
Types:
Laboratory Experiments:
Closed environment with controlled conditions.
Unusual in sociology because findings unlikely to apply to the 'real' world.
Participants are aware that research is taking place and this will affect how they behave.
Raise ethical issues.
Field Experiments:
Not conducted in a closed, controlled environment.
Difficult to control possible independent variables, so tend to establish correlations rather than causation.
Researchers use dependent and independent variables to test a hypothesis or answer a research question.
Strengths:
Laboratory experiments easier to replicate.
Standardized research conditions give experiments a high level of reliability.
Field experiments can be used to manipulate situations in the real world to understand behavior.
Limitations:
It can be difficult to control all possible influences on behavior.
Awareness of being studied may introduce an uncontrolled variable (Hawthorne effect).
Ethical issues regarding permission/consent.
Hawthorne Effect: Changes in people’s behavior directly resulting from their knowledge of being studied.
Content Analysis
Definition: Study of texts (data sources such as television, written documents, etc), which can be approached quantitatively or qualitatively.
Quantitative Analysis:
Uses statistical techniques to categorize and count the frequency of behaviors.
Strengths:
Can identify underlying themes and patterns of behavior that may not be immediately apparent.
Allows researchers to draw complex conclusions from simple data-collection techniques.
Standardized frameworks allow for data to be checked and replicated.
Limitations:
Reliability may be limited because researchers must make subjective judgements about what they are counting.
Does not tell us very much about how or why audiences receive, understand, accept or ignore themes and patterns discovered by the research.
Qualitative Research Methods
Unstructured Interviews
Definition: Respondents are encouraged to talk freely about the things they feel are important.
Strengths:
Data reflects the interests of the respondent.
Avoids the problem of the researcher pre-judging what makes important or irrelevant data.
Sensitive issues can be explored in depth.
Can take place somewhere the respondent will feel at ease.
Limitations:
Requires considerable skill.
Researcher has little control over the direction of the interview.
Time-consuming and so are analysing and interpreting all the data they generate.
Non-standardized format makes the interview impossible to replicate.
Biased by interview effects.
Semi-structured Interviews
Definition: Combines the strengths of unstructured and structured interviews. There is a structure or ‘interview schedule’ – the areas the interviewer wants to focus on – but there is no list of specific questions and Researchers are free to ask questions in any order and to phrase the questions as they think best.
Strengths:
Less risk of the researcher deciding (predetermining) what will be discussed.
Possible to pick up ideas and information that may not have occurred to the interviewer or of which they had no previous knowledge.
Focus on issues that the respondent considers important results in a much greater depth of information.
Limitations:
Demands certain skills in the researcher.
Not only time-consuming but the large amounts of information they produce must also be analysed and interpreted.
Respondents must remember and describe past events, and this creates problems for both researcher and respondent.
Lack of standardization makes analyzing data and generalizing difficult.
Group Interviews
Definition: Respondents gathering to discuss a topic decided in advance by the researcher.
Success depends on:
Clear guidelines for the participants.
Advance (predetermined) questions.
Interaction within the group.
Skill of the researcher.
Strengths:
Reflect how people naturally share and discuss ideas.
Researchers can help the discussion by controlling the pace; planning a schedule and asking questions.
Limitations:
Researcher must control the behavior of the group.
May have problems with representativeness if participants don't show up.
Risk of interview effect ('Groupthink').
Run the risk of simply reflecting a ‘group consensus’ rather than revealing what individuals really believe.
Observation
Definition: Data are more valid if they are gathered by seeing how people behave, rather than taking on trust that people do what they say they do. Can be participant or non-participant.
Participant Observation:
Definition: The researcher takes part in the behavior being studied.
Forms:
Overt: People know they are being studied.
Covert: People are unaware they are being studied.
Strengths (Overt):
Recording data is relatively easy.
Can gain access to all levels.
Ability to ask questions, observe individual behaviors, and experience the day-to-day life of respondents helps researchers to build up a highly detailed picture of the lives they are describing.
Reduces the chance of researchers becoming so involved in a group that they stop observing and simply become participants.
Limitations (Overt):
Research cannot be carried out if a group refuses the researcher permission to observe it.
Requires substantial amounts of time, effort and money.
Observer/Hawthorne effect.
Researcher's level of involvement.
Overt participant observation is impossible to replicate.
Strengths (Covert):
May be the only way to study certain people (Criminal or deviant groups; Closed groups; Defensive groups).
Avoids the observer effect.
Through personal experience, the researcher gains valuable understanding (insights) of the meanings, motivations and relationships within a group.
Limitations (Covert):
Getting in can be a problem.
Researcher may not have access to all areas and has to quickly learn the culture and dynamics of a group to participate fully.
It can be difficult to separate the roles of participant and observer.
Certain ethical questions may arise.
Research cannot be replicated.
Recording data is frequently difficult.
Non-participant Observation:
Definition: Observing behavior from a distance so that the research subjects do not know they are being observed. The researcher does not become personally involved in the behavior they are studying.
Strengths:
Access is one practical advantage.
Respondents can be objectively studied in a natural setting.
Limitations:
Cannot be easily or exactly replicated.
Raises ethical questions because people are being observed without their permission.
2.2 Approaches to Sociological Research
Case Studies
Definition: Studies the characteristics of a particular group or ‘case’.
Strengths:
Focus on a single group studied over time provides great depth and detail of information that has greater validity than simple quantitative studies.
Help to uncover the meanings that people give to everyday behavior.
Can be used as pilot studies to allow a researcher to develop hypotheses, test data-collection methods and identify potential problems in preparation for a larger study.
Limitations:
Large- scale, in-depth studies can take a lot of time, effort and money.
Difficult to generalize from case studies.
Social Surveys
Definition: Designed to produce a ‘snapshot’ of behavior at any given time.
Types:
Cross-Sectional Survey:
Definition: Types can be qualitative(descriptive) or quantitative(analytic).
Longitudinal Survey
Strengths:
Can make generalisations about behaviour.
Limitations:
focus on identifying groups that share broad similarities, such as income, education and gender, instead of focusing on the individual.
Ethnography
Definition: A way of researching which tries to achieve a detailed, in- depth understanding of a group of people or of a social situation.
Longitudinal Studies
Definition: Tracking changes among a representative sample over time.
Strengths:
Allow the researcher to identify and track personal and social changes over long periods, revealing trends that would otherwise remain hidden.
Usually based on large representative samples and so can be used to suggest correlations and causal relationships.
Limitations:
Sample attrition reduces the representativeness of the sample over time.
Lacking depth and validity.
Methodological Pluralism
Definition: Some research projects involve more than one method, even combining methods that produce quantitative and qualitative data.
Triangulation
Definition: The ways through which methodological pluralism is put into practice that aims to improve research reliability and validity. Can be methodological, researcher and/or data triangulation.
2.3 Research Issues
Positivism
Definition: States it is both possible and desirable to study social behavior using similar methods to those used when studying the natural world; in other words, sociology can be like a science.
Interpretivism
Definition: States social reality is formed through the interaction of people who have consciousness.
Theoretical Considerations
Factor involving researchers making initial decision on what counts as data. Should the data be statistical or descriptive? Topic Choice:
Based on personal or institutional factors. Personal- extent to which the researcher is interested in the topic. Institutional- universities and governments are important sources of research funding and can affect the choice of topic.
Choice of Method:Beliefs about the reliability and validity of particular methods will play a part: what they choose is influenced by researcher's perspective.
Practical Considerations
Access to research subjects:
* Researchers may choose to carry out covert research as access may be denied.
lack of Cooperation may be solved by sponsorship:
* Problems can also also explain why a lot of sociological research focuses on the activities of the powerless, people who cannot say no, rather than the powerful, who can and often do resist being studied.
choice of method is also affected by practical considerations: Time is a factor
Ethical Considerations
Definition: Moral principles that determine whether actions are right or wrong.
Legal Considerations: Important when research involves observing or participating in illegal behavior.
Moral obligation to obtain informed consent- participants must be aware of all aspects of the research and be able to make a free and informed decision about participating.
Validity, Reliability, Objectivity, Representativeness, and Ethics
*Validity: Methods and data are only useful if they actually measure or describe what they claim to measure or describe.
Ethical validity: refers to the extent to which research methods reflect the world being studied.
Laborotory experiments have very low ecological validity because laboratories are artificial situations in which people do not behave as they would in other situations: Covert participant observation, on the other hand, has much higher validity, because people are observed acting in ‘real life’.
*Reliability: Refers to how effective a research approach is at collecting consistent data; it is about whether the accuracy of the data can be checked by repeating or replicating the research.
*Objectivity: The ability to make objective statements about behavior. Is a particularly significant factor- researcher has no personal stake in the truth or falsity of the behavior they are testing or describing and try to avoid unfairly influencing that behavior.
*Representativeness: the extent to which the research findings apply to a larger population. Representativeness is the ability to generalize observations made about one relatively small sample group to the much larger target population it represents.
*Ethics: There exist both deontological ethics (rules for conduct) and consequentialist ethics (evaluates whether research is morally acceptable by looking at the end result).