Posted by Brian C. Ventura on June 25, 2010
C. Generating the Hypothesis
Overview-the main task here is to identify the primary relationships that your theory predict will occur given a set of conditions outlined in your theory; testable assumptions.
a. must refer to at least concept as independent variable (predictor)
b. must refer to one concept that represent a dependent variable
c. must be as parsimonious as possible so it is clear to the reader
d. refer to concepts, not variables
Ex. 1. (Political Dynasty Hypothesis) The presence of a (mayoral) (congressional/district rep) candidate from long entrenched political family increases the number of gun related incidence during election period.
2. (Incumbency hypothesis) the absence of incumbent candidate increases the number of harassment reports.
3. (Conflict stalemate hypothesis 1) The longer the duration of an enduring rivalry, the more likely mediation is to take place.
4. (Conflict stalemate hypothesis 2) Mediation is more likely to take place when a civil war is ongoing within at least one of the enduring rivals.
5. (Disputant characteristics hypothesis 1.) Mediation is less likely to occur in rivalries involving a major power.
e. there must be sufficient knowledge to posit a relationship from existing studies and existing theories.
D. Identifying Variables (for June 28, 2010 Thursday lecture)
a. Overview
Variables are empirical phenomenon that takes on different values or intensities.
Constant-any phenomenon that always take the same value
Ex. Demographic variables- age, sex, marital status,
Behavioral variables – emotional, attitudinal, cognitive, party loyalty, trust towards political institutions.
b. Conceptual or nominal definition of variables
i. Conceptual/nominal definition of variables- dictionary meaning/ definition by classic work, main theorist/seminal work – use familiar terms to describe words or phrases that are unfamiliar
ii. operational definition of variables- specifies in empirical terms precisely what should be done to observe variations in some variables
Ex. Long entrenched political clans- a political family (nuclear/extended) holding two or more elected office in the municipal and district or provincial level simultaneously for at least two consecutive three-term cycle (18 yrs)
Wealth- self-reported income(statement of assets, liabilities and net worth), materials used for house construction.
c. level of measurement/levels of calibration- from simple to complex
i. nominal- naming the calibrating units,
ex. Sex, religious preference, party affiliation, college major, elected, appointed, NGO-Non NGO,
ii. ordinal- involves arranging the calibrating units into a logical order or rank,
ex. Academic ranking (1st/2nd); class of municipalities, 1st class, second class; country income, high middle income, middle middle income, lower middle income,
iii. interval level measurement-involves not only having an order to the calibrating units (ordinal) but also specifies the distance (usually an equal distance) between each successive units.
Ex. Self-rated degree of loneliness from 1-10, sense of political efficacy from 1-5
iv. ratio measurement-not only specifies the distance between each successive variable but is also assumes that there is a point at which the variables literally does not exist and designates this as point zero
ex. Population density and suicide rate; urban population and political participation; economic development and democracy
d. discrete and continuous variables
i. discrete variables- exist in two or more segment regardless of how political scientists decide to measure it, can be nominal or ordinal
ex. Sex (male and female); religious affiliation (Catholic; Protestant; Muslim; none); family type (extended, nuclear)
ii. continuous- does not exist in segment but instead vary from low to high, strong to weak
ex. Age, academic ability, IQ, level of violence, economic growth, population growht