COMPONENTS OF A THEORY

[Examples from international relations "realism"]

Vocabulary, variables and concepts - what is important?

Military power and geographical location are very important;
economic relations less so; "Power" and "national interest" are concepts

Relationships between variables

"Nations seeks to maximize their power"

Explanation for why these relationships hold

"Humans are fundamentally evil and greedy, therefore nations will be so"

Historical archetypes which illustrate the operation of the theory and provide empirical support for it

Europe from 1815 to around 1900
Greek city-state system 400 BCE Formal analysis: Correlates of War Project

Theory in Practice:

Identifies what is important in the world; provides a "filter" for information

Method of making predictions

Provides "missing values" when there is no information contradicting the theory; becomes "common sense"

Provides information on individual motivations Ñ what are they trying to achieve?

Normative theory answers issues of "should" and "ought"; positive theory only addresses "what is".

"There is nothing more practical than a good theory" 

HOW ARE THEORIES EVALUATED?

Formally:

Multiple systematic statistical tests

Informally

Generalizations from world history

Reactions to recent events [Often a reaction to the failure of an earlier theory, for example WWI and WWII]

Individual experience 

Characteristics of a good formal theory

Logically consistent

Testable/falsifiable

Communicable and unambiguous

GeneralÑexplains more than one case

Parsimonious -- "Occam's Razor"

Interesting(!) -- why should we care?

---

Stages of Social Research

Stages of Social Research

1. Selection of the research problem. This also involves

a. doing the appropriate literature review

b. figuring out what is possible to study

2. Formal statement of hypothesis or more usually multiple hypotheses

3. Operationalization and measurement of the variables. Also see if the variables you need have already been measured

4. Decide on the sample of the units (this may be the entire population).

5. Collect the data

6. Analyze the data, usually statistically. At this point one usually does a lot of exploratory analysis to see what form of the hypothesis works best.

7. Interpret the results.

---

How to Read a Scientific Study

1. What is the audience? (frequently determined by the journal)

2. What is the question/thesis/hypothesis. What is the underlying theory?

3. What does the existing literature say (typically: it is confused)

4. What is the data set?

Population (in time and space)

Unit of analysis

Variables

5. What is the statistical model?

crosstabulation

regression

logit

6. What are the results?

7. How do the results change our understanding of the theory?