Chapter 2

primary data

Information collected personally by a researcher

secondary data

data that already exists; data not personally generated by the researcher

official statistics

A government- generated secondary source of data on areas such as crime, marriage, and employment

quantitative data

information expressed numerically that captures the 'who, what, when, and where' of behavior

correlation

a statistical relationship between two or more variables that expresses a level of probability. A high (positive) correlation suggests the strong probability of a relationship. A low (negative) correlation suggests the probability of little or no relationship.

hypothesis

a statement put forward by the researcher which is then tested to see to see whether it is correct.

qualitative data

non- numeric data that expresses the quality of a relationship

participant observation

a research method that involves the researcher participating, openly (overtly) or secretly (covertly) in the behavior they are studying.

comparative analysis

collecting data on one group or society and then comparing them with others to identify conditions that may be causing social events.

representativeness

extent to which the characteristics of a sample population accurately reflect those of the target population

personal documents

secondary source of data covering areas such as personal letters, diaries, oral (verbal) histories, websites, social networking sites and photographs.

questionnaire

research method consisting of a list of written questions

structured interview

Set of standard questions asked by the researcher of the respondent. It is similar to a questionnaire, but is delivered by the researcher rather than completed by a respondent.

researcher effect

Also called the interviewer effect, this refers to how the relationship between researcher and respondent may bias responses and lead to invalid data

variables

factors that can be changed (manipulated) by the researcher to understand their effect on behavior.

correlations

when research shows a relationship or connection between two or more variables

causation

the idea that when one action occurs, another always follows because the latter is caused by the former

experimental group

the subjects of an experiment. The researcher changes different variables to test their effect on behavior. There is often also a control group, for whom the variable is not changed. This allows comparisons to be made between the control and experimental groups.

control group

in experiments, a group for which the researcher does not change the variables, so that it can be compared with groups for whom variables have been changed.

experiment

a research method which follows a set procedure to test a hypothesis. Laboratory experiments take place in a closed environment where conditions can be precisely monitored and controlled. Field experiments take place in the 'real world.'

Hawthorne (observer) effect

changes in people's behavior that result from their knowledge of being observed

content analysis

Research method used for the systematic analysis of media texts and communications

unstructured interviews

free-form interview method where the objective is to get the respondent to talk, without help (prompting) or interruption, about whatever they feel is important about a topic

Semi-structured interviews

a research method in which a respondent is encouraged to talk at length about a particular subject. Also called focused interviews because the topic is decided by the researcher and is the focus of their questions.

Group interview

also called focus groups, these involve respondents discussing a topic as a group rather than individually.

Verstehen

research strategy central to participant observation, which takes advantage of the researcher's ability to see things from the subject's viewpoint.

Overt participant observation

when those being studied are aware they are being researched. contrasted with covert participant observation, which is when those being studied are unaware they are part of a research project.

Non-participant observation

when the researcher observes behavior without participating in that behavior.

sample

involves a small number of subjects drawn from a much larger (target)population. sociologists use a variety of random and non-random sampling techniques.

pilot study

a 'mini version' of full-scale study designed to test its feasibility.

operationalization

converting a concept, such as social class or cultural capital, into something that can be researched and measured.

case study

an in-depth, qualitative, study of a particular group of 'case.'

cross-sectional study

research method focused on identifying groups that share broad similarities, such as level of education, and measuring differences in a single variable; whether, for example, people with a high level of education have higher rates of suicide than those with a lower level of education.

ethnography

the in-depth study of a group or culture, usually involving participant observation and often other methods as well.

longitudinal study

a form of comparative analysis that involves tracking changes among a representative sample over time.

methodological pluralism

combining research methodologies in ways that allow each to complement the other to improve research reliability and validity.

positivism

the approach within sociology which promotes (advocates) scientific methods to discover facts about the social world.

interpretivism

the approach within sociology which advocated methods that enable the sociologist to understand the meanings that people attach to their actions.

triangulation

the use of two or more research methods where the weaknesses of one method, such as a quantitative interview, can be balanced (offset) by the strengths of another, such as qualitative participant observation, to improve overall research reliability and validity.

value-free

the ability of researchers to prevent their own values (such as personal, political or religious values) from influencing their research.

ethical issues

'ethics' refers to the morality of doing something. ethical questions relating to sociological research involve beliefs about what a researcher should or should not do before, during and after their research.

validity

the extent to which a research method describes or measures what it claims to describe or measure.

reliability

the effectiveness of the research approach in generating consistent data. a researcher can check the reliability of their research by the repeating (replicating) the research to see whether they get the same, or very similar, results.

generalize

when findings of research on a sample can be said to apply to a larger population.