Presentation on theme: "Sociology 282. Searching for [causal] relationships."— Presentation transcript:
Searching for [causal] relationships
Cause Laypersons definition: when one thing affects another an explanation involving the belief that variation in the independent variable will be followed by variation in the dependent variable, when all other things are equal (ceteris paribus)
Hypothesis A statement of a presumed causal relationship [between 2 or more variables] Schutt: a tentative statement about empirical reality involving a relationship between two or more variables
Nomothetic Usually where the quantitative folks are Often is probabilistic
example As more of kid’s play is in the street more injuries will occur The more often a women goes to her car alone out of the mall the more likely her purse will be stolen As you increase your partying before exams your grades go down
Probabilistic? Note it probably will happen, but not to everyone…or not every time
Ideographic Usually the place the qualitative folks are located the finding that a series of events following an initial set of conditions leads in a progressive manner to a particular event or outcome Narrative reasoning High concern for context Holistic explanations Deterministic
example Stephen King’s broken leg was caused by walking on a narrow shoulder of the road, on a slope where there was poor vision and a speeding, drunken driver swerved off the road
cause When changes in the [Causal thing] X independent variable lead to changes in the [effect] dependent variable Y
example Study [X] Performance [Y] The more you study, the better you'll do in this class
example 2 Fire fightersFire damage
Spurious relationship? Size of fire fire fighters damage
spurious relationship Schutt: a relationship that appears to be connected but is not An apparent relationship is really caused by a prior variable Z which only makes X and Y look as if they're related
Major Problem: What looks like a causal relationship may be spurious This is a major issue for research design/data collection and will be considered below in the “Internal validity section
Types* of social science research * note, I think that Schutt’s calling these “types” of research rather than “goals” is confusing
Types* of social science research Descriptive Research Exploratory Research Predictive Research [not mentioned in Schutt] Explanatory Research Evaluation
Descriptive Research Research that defines and describes social phenomena (e.g., National Geographic “Survey 2000” that described Internet users around the world and identified differences between countries) Research that is not searching for causes or reasons why things happen … stuff like taking a census or poll taking for political purposes
Exploratory Research investigation of social phenomena without expectations (e.g., electronic diabetes newsgroups were found to also be support and information networks, a place where information could be assimilated to inform choices) Perhaps research that looks for causes when we know very little about our topic of study
Predictive Research Research that establishes relationships that will let us guess how folks will do in the future (e.g. using information like SATs or ACTs to predict future college performance when we don’t presume that performance on these tests causes later college performance)
Explanatory Research research that identifies causes and effects of social phenomena (e.g., research that suggests that Internet use hurts or helps other forms of social interaction.)
Evaluation research that determines the effects of a social program or other type of intervention (e.g., in the Toronto, Ont. Suburb that was wired with the Internet, universal Internet access increased relations between residents) Research that assesses how well you do your job …how well a program works for your kids Whether your job will be cut Research that will affect YOU!
Evaluation Inputs: Programs and goals; Equipment and facilities Processes: Program and its delivery Products: Short run, immediate accomplishments Outputs: Program goals accomplished Outcomes; Long run result of program goals
Quantitative and Qualitative Orientation These days, there are two major philosophical orientations towards data collection and interpretation. One is “soft” and the other is “hard.” One advocates that “science will save us.” …and the other has advocates criticizing science as the subjective perspective imposed on us by powerful, Eurocentric, white Protestant males. One advocates “objective truth” and sees the other as “any subjectivity goes.” Be prepared to consider these perspectives throughout the semester.
Orientations? Note that I see these as extremes on a continuum and I see that all the issues and criteria relevant for one situation, relevant for others. In particular, we’ll study experiments, surveys and participant observation [or qualitative research] I think that what applies to one applies to all.
–Quantitative methods: data collection methods such as surveys and experiments that record variation in social life in terms of categories that vary in amount Data are numbers OR attributes that can be ordered in terms of magnitude Most often used for explanation, description, and evaluation
–Qualitative methods: data collection methods such as participant observation, intensive interviewing, and focus groups that are designed to capture social life as participants experience it rather than in categories predetermined by the researcher Data are mostly written or spoken words or observations Data do not have a direct numerical interpretation Exploration is the most often motive for using qualitative methods
Goals of Social Research Note: this is Schutt’s title I prefer types of validity or Four reasons science must be tentative
Measurement validity: exists when a measure measures what we think it measures Operational definition matches conceptual definition
Schutt says “Generalizability: exists when a conclusion holds true for the population, group, setting, or event that we say it does, given the conditions that we specify.” External Validity involves generalizing from a study about 1. types of subjects, 2. settings, 3. measures employed, 4.the timing of delivering the causal variable and the timing of when the effect is observed
External validity Some causal influences work under; 1) some conditions [winter or summer and SAD] 2) for certain types of people [children vs. adults] 3) with specific types of causes in the same category [e.g. therapy] or with different timing [studying for this class all at once or in small doses]
Generalization is based on representation. If you haven’t looked at it you can’t talk about how your results apply to it.
Statistical interaction If the X-Y relationship works differently under some conditions, etc. than other this is refereed in the social sciences as statistical interaction.
example Performance –High o –o –Low – o –No audienceAudience –Red dots are for experienced performers
Causal (internal) validity: exists when a conclusion that A leads to or results in B is correct Causal validity…more later…means that we tentatively trust that the cause is responsible for the effect…and that we’re not being fooled by other things going on Internal validity is concerned with eliminating spuriousness and certain types of statistical interaction
Authenticity: When the understanding of a social process or social setting is one that reflects fairly the various perspectives of participants in that setting (i.e., a resolution of whether an objective social reality exists independent of actors’ interpretations ) Note that I don't see a resolution here. I see that colleagues want to ensure that what we as researchers think is going on is cool with participants
However Sometimes the critics of “objective research” feel that the “scientist” is speaking for the participant and that leads to false research…yet such critics have no qualms about becoming “the voices of the disenfranchised” even to the point of figuring out what people feel for them…saying what the participant is saying even when the participant isn’t aware of this…aren’t we back to square 1?
What we’re about: Major emphasis on validating cause-effect relationships …in experiments surveys and so called qualitative studies Examining measurement, casual & generalizing validity and authenticity Using learning research tools and Learning new vocabulary! Becoming an informed consumer and critic or research