Scope and Methods GWU Milne Fall 2019

Heuristics: experiential reality, agreement reality, causal reasoning, probabilistic reasoning. - correct answer Human inquiry Inaccurate observation, overgeneralization, selective organization, illogical reasoning. - correct answer Problems with human inquiry Objective method for gaining scientific knowledge based on universally accepted rules and principles for collecting, analyzing, and evaluating data. - correct answer Empirical research Application of the well-defined principles of empirical research to the study of political phenomena. - correct answer Political science Use verified results to construct causal laws that explain why phenomena behave the way they do. - correct answer Goal of political science Empirical verification, non-normative, transmissible, general, explanatory (causal relation: predictive and probabilistic, theory), provisional. - correct answer Characteristics of scientific knowledge Evaluative, value-laden, prescriptive, should/ ought. - correct answer Normative knowledge Factual, objective, what is/ why/ what might be in the future. - correct answer Non-normative Process of reasoning from specific observation to general principle or theory. - correct answer Induction Process of proceeding from general principle or theory to specific observations. - correct answer Deduction 1: Formulating research process 2: Literature review 3: Formulating hypothesis 4: Defining concepts 5: Operationalizing concepts 6: Measuring data 7: Statistical analysis 8: Drawing conclusions 9: Writing research report - correct answer Political research process Statement that indicates there is no relationship between the variables in a hypothesis. We test the null hypothesis rather than the hypothesis when we use statistical analysis. Null is assumed to be true unless results from the research and statistical analysis results indicate otherwise. - correct answer Null hypothesis A specific, testable statement of a relationship, derived from a theory. States the nature of the relationship between two or more variables (at a minimum, have to have at least one independent and one dependent variable). - correct answer Working/ alternative hypothesis Empirical statement: Educated guess about relationships that really exist; not what ought to be true or what researcher wants to be true. General: Should explain a general phenomenon rather than one occurrence. Plausible: Should be logical to assume the relationship will be confirmed. Specific: Should be able to state a directional hypothesis if appropriate; should be carefully defined; unit of analysis should be clear. Consistent with Data: Stated in the way that researcher intends to test it. Testable: Variables must be measurable and data must be obtainable and be able to indicate whether hypothesis is correct or incorrect. - correct answer Scientific hypothesis characteristics A variable that reflects a quantity or amount. Consists of numbers. Differences between numbers or categories can be expressed numerically. - correct answer Quantitative variable Variable that reflects a quality or category. Variables that have discrete categories, usually designated by words or labels. Differences between categories cannot be expressed numerically. - correct answer Qualitative variable Variable of primary interest in study. Hypothesis predicts changes to it. Variable that is influenced or changed by the independent variable. - correct answer Dependent variable Manipulation or variation of this variable(s) is hypothesized to cause changes in the dependent variable. Variable that is used to explain a dependent variable. - correct answer Independent variable A variable that is related to both the independent and dependent variables that may enhance or hide the relationship. Must be controlled or internal validity between the independent and dependent variable is reduced - correct answer Confounding variable A variable that occurs between the independent variable and the dependent variable and affects the relationship between them. - correct answer Intervening variable A variable that occurs prior to all variables and that may affect other independent variables. - correct answer Antecedent variable Values of the control variable are "held constant," while the relationship between the other two variables is analyzed. Often, we can control confounding variables by having a relatively large sample size as compared to the size of the population being studied. - correct answer Control variable A pattern, not just simple random chance, of association, dependence, or covariance of the values of one variable with the values of another. - correct answer Relationship Broadly refers to any relationship in which the variables perform in a synchronized manner; however, does not tell us whether one causes the other. - correct answer Correlational relationship A relationship between two variables in which changes in the value of one variable affect changes in the value of the other variable. - correct answer Causal relationship Correlation: You can have correlation without causality, but you can't have causality without correlation. Time Order: Cause must proceed effect. Non-spurious: Real correlation; the relationship is not coincidental correlation between two variables shown to be caused by some third, non-related variable. - correct answer Criteria for causality Complete Causality Exceptional Cases Majority of Cases - correct answer False criteria for causality Apparent causal relationship between two variables that is actually due to one or more other variables. Two variables appear to be correlated (i.e., that one variable is having an effect on the other) but the relationship exists only because another variable is affecting both. - correct answer Spurious relationship Variables vary together in the same direction. As one increases, the other increases. As one decreases, the other decreases. - correct answer Positive/ direct relationship Variables vary in the opposite direction. As one variable increases, the other decreases. - correct answer Negative/ inverse relationship Precise statement indicating the nature and direction of the relationship between the variables. - correct answer Directional hypothesis Cannot have a directional hypothesis if a nominal variable (qualitative variable without rank order such as race, political party, region, religion, etc.) is included in the hypothesis. Hypothesis can only state that a relationship will occur. - correct answer Non-directional hypothesis Includes direction of relationship, if applicable. State the independent variable first, then the dependent variable (cause and effect). Include the unit of analysis. For dichotomous variables such as gender, use one of the values of the variable to specify the predicted outcome. Refer to the variables by name, not by "independent" and "dependent variable. - correct answer Components of a strong working hypothesis Keep it simple! Don't state the opposite of the hypothesis - simply state there is no relationship between [fill in independent variable(s) name(s)] and [fill in dependent variable name]. Do not include direction even if it was included in the hypothesis. State the independent variable first, then the dependent variable (cause and effect). Include the unit of analysis. For dichotomous variables such as gender, do not use one of the values to specify the relationship - keep it high level. - correct answer Components of a strong null hypothesis Statement of the meaning of a particular concept. Best construct is a widely accepted conceptual definition from your lit review. Subjective. - correct answer Conceptualization