– If it is viewed as an arts discipline.
– often designed to get at the rich meanings that people attribute to their lives.
– Usually used to understand the topic or population being studied. This is done to make an accurate argument concerning group. (Observing people over and over again to draw conclusions)
– Is not absolute and many sociologists prefer NOT to use it.
– If sociology is viewed as a science
– Sociological research is often designed to measure and quantify social life
– Although complete objectivity is impossible, due to the fact that no person can completely remove themselves from their past. Closest form of objectivity in Sociology can be obtained if proper methods are used to record, collect and analyze data.
– Scientific rules are never definite, but if correct research is done, they make extremely good guidelines to life.
– Research must be closely connected with theories
– Theories are abstract ideas
– Try to help understand societal problems better (crime, environment, etc).
– Models/Maps of how the world works.
– Most can be grouped into: structural functionalism, conflict theory, symbolic interactionism, and feminism.
– THEORIES MUST BE TRANSLATED INTO OBSERVABLE IDEAS.
THEORIES (compiled of) > CONCEPTS
then physically observed through –
HYPOTHESIS (compiled of) > VARIABLES
– THEORIES MUST BE TRANSLATED INTO OBSERVABLE IDEAS
– The process of translating theories and concepts into hypothesis and variables.
– Basically the theory of capitalism, was formed off the concepts ‘exploitation’, ‘class’ and ‘alienation’. Which could have been formed off the hypothesis that “School is only successful option for the middle class” which would be based on variables like ‘yearly income’ (dependent), ‘type of job’ (independent).
– Usually theories explain how two or more concepts are related to each other
– Example: Karl Marx used concepts of “alienation” “exploitation” and “class” to form his theory of capitalism.
– Must contain a minimum of two types of variables.
– Must be observable, and take on a range of different values (i.e. age, salary, etc).
– Independent: Something that doesn’t change. Like gender. That the dependent variable will rely on.
– Dependent: Something like salary. Will differ based on age and gender in some cases. Therefore salary depends on those two classifications, it is DEPENDENT.
External Validity – the extent to which the results of research can be generalized to the larger population.
^Accurate sample sizing, as well as unbiased participant results are crucial in determining external validity.
Internal Validity – the degree to which the conclusions of a study support the data and methods that were used.
^ This is threatened if a portion of the evaluation process could be influenced by more than one accounted for variable. FOR EXAMPLE if your testing for math tests and the reason students did well is not just because they had a class in it (unaccounted for previous studying, etc.)
– For example if the idea or concept is amount of education ‘Years of Schooling’ would be an extremely valid variable. However years of schooling would not be valid in relation to ethnicity as the two subjects have little in common.
– A reliable measurement process is one that produces the same results of the same phenomenon over and over.
– Harder to determine in a social environment
Bias: Systematic inaccuracies in our data or analysis. Usually due to respondent bias though. People checking off whatever they want, or selecting based on a surveyor being present. Have larger effect on survey conclusions.
– Experiments: A controlled environment in which research can be done, and specific factors can be manipulated to determine their outcome. Good for comparing variables.
– Factors of people (age, sex) are not as easily adjustable, and circumstances may differ in the real world (real threats vs. fake threats), so sociologists tend NOT TO experiment.
– Excellent way to gather data on a large population, and generalize results.
1. Pseudo Surveys: usually in the form of a questionnaire that just asks specific questions, to specific people (for example, to form sales pitches). Surveys usually are specifically designed for a wide range of people to gather information. Surveys collect rather than manipulate.
2. Constructing Survey Questions: Should be relevant, unbiased, clear, brief, specific, useful and focused. Avoid double barrelled questions (having multiple focuses). Be non-threatening, non-invasive.
3. Random Sampling, Sample Size and Response Rates: Sample’s should be well representative of the population. They should be completely random, as to not show biased. Simple random sampling is often used where people are put on a list and randomly chosen. Population must always be accounted for. Response rate is also an important factor, a certain group of people may be more inclined to respond.
– Pseudo Surveys are done for research in social science rather than specific
– often reffered to as quasi experiments
– Spend time getting to know subjects in order to capture a world view.
– Much easier to make changes to and reassess than surveys. Can provide unexpected opportunities, and usually only done until the researcher can not learn any more about the specific group being studied.
– In many studies more than one of the following techniques is used:
1. Ethnographic or Participant observation Research:
– Observing the daily activities of research subjects. This may include accompanying them on their daily activities, or even living with them. During such activities researchers take field notes. For example Ben Killings worth’s’s study of Mom’s and Tot’s and observations made on natural conversation.
2. In Depth Interviews:
– Usually extensive interviews that are often tape-recorded and later transcribed into text.
– Can be highly structured or unstructured. Usually semi-structured with a decent amount of relative questions allow interviews to be flexible, yet still knowledge gathering.
– In some studies, mostly in organizations, people have access to documents which they can study to learn more about a group of people (such as a religious group, etc.)
– Usually already formed statistics, or surveys done by other researchers.
1. Secondary Data Analysis:
– One of the most common forms of research, usually in pre-made graphs or charts.
– Advantages are that it is helpful as the collectable data has already been done for you, and now a days previously created research is easily accessible (StatsCan for example).
– Disadvantages is researchers can’t test specific ideas using this population of data.
2. Historical Research and Content Analysis:
– Historical Sociology relies heavily on historical documents to conduct research.
– Often times content analysis will be done on historical docs to determine, patterns, manifested ideas, and analyze what documents meant during that social period.
– They can often times contain unknown biases, or be of bad quality because so much of history is still unknown/they’re old.
– Exploratory: or descriptive, the goal is to find out more about a group of people, or a topic.
– Explanatory: Usually these studies test different theories against each other, to determine which theory provides the best explanation for the phenomenon (or from past research). For example – If public or catholic school students do better because they have social capital.
– Some researchers want to empower the group of people they are studying (such as feminists). Participatory Action Research: When researchers are guided as much by their desires as they are by their interests.
– Such as voluntary participation, consent (unless it’s a study about consent), knowledge of what is being taken part in, confidentiality, or anonymity (not ever meeting the person), and lastly research should not involve ANY harm to participants.
– Some problems require more observation, and some require more gathered data.