Research Methods

Carl Sagan

advocate of science; prevent backslide into supersitition and unintellectualism, resist impulse to give up

Where does social science fit in with the other sciences?

it usually doesn’t really fit in; hierarchy based on complexity (deterministic → propable science: problems we are trying to solve becoming more complex) (physics → chem → bio → psych → sociology)

Physics

first in hierarchy (deterministic); simpler world outlook than sociolpgy; gravity does not change

Chemistry

2nd in line; conclusions reached become more propable; more likely vs less likely outcomes

Biology

3rd in line up; cells are so much more ocmplex than subatomic particles; can’t say how a specific kind of single cell amoeba will react to a stimuli with certainity

Psychology

4th in ilne; complex/more propable because human behavior

Sociology - Sciences Hierarchy

human beings as a whole (more coplicated phenomena); never tell you with certainity what will happen (tendancies and patterns) → when adding human behavior, things go non-linear

What are we typically interested in in science?

what causes certain things to happen

Variable

any measureable thing that tends to have different values across cases; ex. GNP (gorss national product), height, weight, age, etc.

Case

whatever unit of anaylsis we are doing; ex. countries, people, etc.

Research Question

start with this; a proposed relationship between 2 or more variables; answers aren’t usually a yes/no; usually 2 variables (independet and dependent); need to conclude with adequate confidence that 1 thing (x) causes another (y)

Confirmatory Research Question

imply a hypothesis; can conclude with a yes or no; ex. Do high levels of political polarization lead to civil war? Does a country’s industrialization decrease economic inequality among its citizens?

Exploratory Research Question

not designed for a yes or no; designed to have an open-ended answer; ex. How do ideas of masculinity influence Pakistani men’s decisions to join terrorist groups?

Independent Variable

variable you are changing; suppose it is influencing the other variable; ex. political polarization

Dependent Variable

out come; effect trying to explain; ex. civil war

3 Primary Goals of a Scientific Inquiry

validity, reliability, and generalizability

Validity

1 of 3 goals of scientific inquiry; refers to how we “operationalize” our variables; are we truly measuring what we claim to be studying?

Operationalize

to find something measureable that stands in for the thing we want to learn about

Reliability

1 of 3 goals of scientific inquiry; concerns the consistency of measurements (across time and observers)

Generalizability

1 of 3 goals of scientific inquiry; means the observations and findings of a specific study can be applied to other, diff. situations; what you observe in 1 context, can be obswerved in another; ex. population, ecological, and conceptual generalizability

Population Generalizability

to what extent does a sample represent the universe of cases?; hopefully smaple represents the population

Ecological Generalizability

to what extent are the findings from 1 setting also found in different settings? (lab → real world or 1 place/time → another); one setting to aonther; ex. conclusions about terrorist groups in 1 country would transfer to a diff. country/region

Conceptual Generalizability

aka robustness; to what extent would the findings be similar, if the variables were operationalized/measured differently; how would findings be different if variables were operationized differently

4 Causal Hurdles

1. Is there covariation betrween X and Y?

2. Is there a credible mechanism that connects X to Y?

3. Can we rule out the possibility that Y could cause X?

4. Have we controlled for all confounding variables (“Z” variables)

Is there a covariation between X and Y?

1 of 4 causal hurdles; i.e. correlation; is there some pattern between values of X and corressponding values of Y

Is there a credble causal mechanism that connects X to Y?