Deductive vs Inductive Theory Construction Deductive Reasoning Construction general particular 1 start off with a topic 2 take an inventory of already known knowledge of the topic preliminary research beware overlooking the value of introspection 3 specify the range of phenomena your theory addresses 4 identify and specify your major concepts and variables 5 find out what is known about the relationships among those variables 6 reason logically from those propositions to the specific topic you re examining Inductive Theory Construction particular general Example in Babbie Why do people smoke marijuana by Takeuchi looking at the specifics of the outcomes of his research ended up changing his explanation from why some students smoke marijuana to why some didn t based on observations concluded that women had more to lose than men from smoking marijuana evidence women living at home were less likely to smoke marijuana than men students from Asian backgrounds also have more lose from smoking marijuana than students from non Asian backgrounds created a theory based on social constraints to explain the observed differences in the likelihood smoking marijuana IR example experience shows that theory and observation inform each other to move forward a given program of research alliances and defense spending Deductive larger states contribute more take on more burden of leadership Inductive Greece and Turkey anomalously high spending levels within NATO relative to their size compared to other states arming to consider a war against each other Links between Theory and Research deductive model research is used to test theories inductive model theories are developed from the analysis of research data Necessary vs Sufficient Causes Necessary Condition that must be present for the effect to follow X is necessary for Y IR example Free and fair elections is necessary for a stable democracy Sufficient Condition that guarantees the effect in question X is a sufficient condition for Y X is a subset of Y IR example Oppression of human rights is sufficient for coup Significance brings clarity into analyzing the correlation between variables not much exciting explanatory power from pursuit of necessary conditions tougher to determine sufficient conditions hard to satisfy both in IR never discover single causes that are both absolutely necessary and absolutely sufficient when analyzing nomothetic relationships among variables Correlation vs Causation Correlation empirical relationship between two variables such that changes in one are associated with changes in the other particular attributes of one variable are particular attributes of the other correlation is a criterion of causation but does not mean causation Causation variables must be correlated cause takes place before effect variables cannot be explained by a third variable they are non spurious necessary cause represents a condition that must be present for effect to follow sufficient cause represents a condition that if present guarantees effect in question IR example lets take statement human rights abuses by a government causes civil war civil war and human rights abuses are definitely correlated but there are many other variables in place what about ethnic tension drug trafficking type of government dictatorship Validity vs Reliability IR example for reliability Validity how dependably a measure mirrors its concept criterion for validity indirect separate standard of judgement IR example measure of nation s hostility towards each other based on content analysis of the nations newspapers if we found out that nations went to war against each other but measure did not reflect increasing hostility we would be suspicious face validity most general does measure accurately reflect the concept purely subjective amount of campaign contributions as a valid measure of political support Reliability a measure is reliable to the extent that it gives the same result again and again if the measurement in repeated test retest split half check Nomothetic vs Idiographic Nomothetic generalizing across cases identifying a few causal factors that in turn impact and explain several events rather than just one use when you want to focus on a wide range of events however lacks specificity Idiographic one individual instance in great depth all possible causes and explanations for one event explanation is limited to a singular case use when you want to fully list all the causes of one event and focus specifically on a single case IR terrorism nomothetic look at various events relating to terrorism and generalize why terrorist attacks occurred in greater depth idiographic look at all possible reasons as to why 9 11 occurred Cross Sectional vs Longitudinal Studies Cross sectional studies observations of a sample of population or phenomenon that are made at one point in time exploratory and descriptive studies conclusions are based on observations made at one time but aim at understanding causal processes that occur over time problem of generalization when conclusions are limited to one period of time subject to other tests IR example survey about favorability of Iraq war in 2003 Longitudinal studies designed to permit observations of the same phenomenon over an extended period best to study changes over time trend studies cohort studies and panel studies IR example studying the evolution of the Muslim Brotherhood in order to predict its future political moves Conceptualization vs Operationalization Conceptualization mental process where imprecise notions are made more precise produces a specific agreed on meaning for a concept for the purposes of research conceptions summarize collections of seemingly related observations and experiences Operationalization development of specific research procedures that will result in empirical observations representing those concepts in the real world process of developing operational definitions operational definition concrete and specific definition of something in terms of the operations by which observations are to be categorized beware of definitional operationalism Considerations while Operationalizing 1 must consider range of variation this should be governed by the expected distribution of attributes among the subjects of the study 2 degree of precision how find should you make the distinctions among the various possible attribution composing a given variable 3 accounting for relevant dimensions of variables 4 defining the