Study Guide Unit 1 Epistemology What do we know and how do we know it Traditionally we know through recognized authorities precedent intuition common sense and observation Inference Inference requires quantitative and qualitative judgments How we make inferences We cannot observe causation We don t see x and y happen We observe correlation We see x and y move together We come up with a causal story to explain why the correlation takes place Deterministic Relationships Cause X always leads to effect Y When X occurs Y will also occur with CERTAINTY Probabilistic Relationship Cause X usually leads to effect Y When X occurs Y will tend to occur but NOT with certainty Previous Research and theories When examining research that has already been done keep these questions in mind What if any other causes of the dependent variable did the previous researchers miss Can their theory be applied elsewhere How might this theory work at different levels of aggregation micro macro If we believe their findings are there further implications Induction Empirics Theory Observe something and come up with a theory Example 90 of humans are right handed Joe is human Therefore the probability the Joe is right handed is 90 Deduction Theory Empirics Build a theory on assumptions then observe Example All men are mortal Socrates is a man Therefore Socrates is mortal Independent Variable the suspected cause Dependent Variable the something because the effect of the outcome Independent X causes dependent Y Conceptual Definitions the meaning we assign to terms concepts area a share understanding Concepts are our perception of things with common characteristics Simple concepts are tools used to describe the world Correlation two variables seems to move together Causation changes in one variable leads the changes in another variable Correlation does not cause causation The Fallacy of affirming the consequent We have evidence and reason in our faovr therefore we have provided our theory All we can say is we have yet to falsify our theory Time Series in which the spatial dimension is the same for all cases and the dependent variable is measured at multiple points in time Cross sectional in which the time dimension is the same for all cases and the dependent variable is measured for multiple spatial units Starting with a puzzle what is the general phenomenon you are seeking to explain The puzzle is the two cases to have similar outcomes but they have different outcomes We begin with a puzzle instead of a question because a puzzle requires some knowledge of events and puzzles make us look for non obvious causes Means of evaluating theories Generalization Make your nouns and verbs more general President Obama wants to be reelected Presidents want to be reelected Politicians want to be reelected Falsifiable a theory Ability of hypothesis to fail to be falsified Helps to discern the worthiness of Falsify the null hypothesis Your theory is safe for now Parsimony Simple stupid To make a theory more general we often have to give up parsimony and to make a theory more parsimonious we often have to give up generality If we can make our theory more general or more parsimonious and without sacrifice we should do so Formal theory Begin with assumptions and use math and game theory to develop theoretical models Rational Choice Theory you know what you prefer and your preferences are consistent If I prefer A over B and I prefer B over C then I prefer A over C Reliability how consistent will the process by which we measure these concepts be is it consistent Validity how well does the operational definition match the theoretical definition are we measuring what we want to measure Hypotheses The hypothesis is a testable statement derived from theory that indicates a cause and effect between two concepts Requirements for hypothesis Must offer a cause X and effect Y Must specify the direction of that relationship positive or negative Must specify the type of relationship linear or curvilinear Must specify units of variation of interest Units individuals or aggregate Variation overtime Between units Must be falsifiable were not wrong yet Linear graph Curvilinear graph Testing hypotheses We must operationalize our concepts First we need solid theoretical definition Operational definition explains what the concept will look like and how it is to be tapped or measured in the empirical world Null hypothesis the expected relationship if our theory is wrong implicit in all hypothesis When we test our hypothesis what we are doing is attempting to reject the null hypotheses Bivariate Relationships ONE independent variable causes ONE dependent variable Endogenous Relationships X causes Y but Y can cause X too U S Soviet military build up X U S spending Y Soviet spending U S increases spending causes Soviet Union to increase spending X Y But how does the U S respond to increasing Soviet spending Y X Spurious Relationships is there another factor that you are not considering X Z Y
View Full Document
Unlocking...