The Research Process

Wallace's Wheel of Research

Terminology

Units of analysis- can never be a variable; who or what we are studying / what level we are studying (ex: individuals, neighborhoods, counties, cities, states, countries); units of analyses can be at the micro, meso, or macro level

Cases - the number of the units of analysis (ex: 600 individuals)

Positivism - belief that observations are the best way to seek knowledge.

Hypothesis - a testable prediction/statement of the relationship between 2 or more variables

Research question - a question

Variable - different types of variables/ change; must be mutually exclusive and exhaustive

     Mutually exclusive -  can only fit one category ( correct ex: age 0-20, 21-40, 41- 60, 60+); in other words, values do not overlap (ex: age 0-20, 20-30, 30-40 would not be mutually exclusive since someone who is 20 could choose either option 1 or option 2)

     Mutually exhaustive - is able to fit into a category (ex: age 21-40, 41-60, 61 would not be mutually exhaustive because  if someone is 19, where would they fit??)

Constant - if attribute is the same, you have a constant

Dependent variable - depends on independent variable (y)

Independent variable - stands alone (x)

   EXAMPLE
   Independent variable: studying
   Independent variable 2: IQ
   Independent variable 3: Employment
   (You can have many other variables)
   Dependent variable: Performance
   (^how do you measure performance though, more later on that!)

For practice identifying independent versus dependent variables go here: http://mathbench.umd.edu/modules/visualization_graph/page02.htm


Ecological fallacy- occurs when conclusion are drawn from aggregate data; you can't draw inferences about a lower level from a higher level of study (ex:finding the U.S. has many welfare programs then concluding Oklahoma has many welfare programs would be an ecological fallacy

(Here is a great video explaining mutually exclusiveness, mutually exhaustiveness, and ecological fallacies: https://www.youtube.com/watch?v=wj3ByhQr0Bg)

Cross sectional - a "snapshot" of data; data gathered once
Longitudinal - "snapshots" over time; data gathered over time

Operationalization (how you measure something)

Nominal measurement:
Ex: yes / no (dichotomous variable); gender; city you were born in

Ordinal
Ex: sometimes often always; 0-20, 21-40, 41-60, 61+

Interval/Ratio
Ex: income last year?, what is your age?

*interval and ratio also called continuous variables or metrics