Political science research methods (quiz 1)

causal political science

1. exploratory 2.hypothesis testing *vast majority of political science research is concerned with questions of causality *vast majority is hypothesis testing

most political science is about identifying patterns

1. in human political behavior 2. in political institutions 3.in the social structures that relate to politics

replication

1. makes sure initial research was done correctly 2. get a better sense of the true value of the parameter of interest 3.improvement of the procedures

levels of measurement (the higher the level, the more information it provides)

1. nominal (different values only indicate different categories) 2.ordinal (different values allow us to rank-order the cases, to put them in a meaningful order) 3. interval (different values represent a consistent degree of variation) 4.ratio (same as interval, with zero indicating an absence of the concept)

methods of theory development

1. poaching from another discipline (deductive) 2.applying broad political science theories to specific questions 3. look at cases, selecting a manageable subset of cases with an eye towards controlling for some variation; develop a theory by searching for patterns in the data/cases (inductive)

ways to administer a survey

1. self administered 2. face to face 3.phone

four pillars of scientific research in social sciences

1. the procedures are public *facilitates replication-repeated same procedure to obtain new results 2. inference (descriptive or causal) 3.conclusions are uncertain 4.method, not the subject matter, that makes a study scientific

literature review continued

1. to familiarize the reader with the state of research on the topic, often including contrasting perspectives and viewpoints, which can serve as jumping off point 2. to describe the strengths and weaknesses of that existing research, discussion of weaknesses often jumping off point for your proposal... leads into a presentation of what you are hoping to add to existing body of knowledge on your topic

population

a collection of units about which you are interested in making claims, all of the cases or observation covered by a hypothesis

census

a gathering of data about all of the cases in the population

push poll

a poll where the intention is to transmit (typically false) information under the guise of a question

content analysis

a procedure by which communications are transformed into quantitative data

non-probability sample

a sample selected from a population using a method that is not based on a chance process....haphazard sample, purposive sample, chain referral sample

probability sample:

a sample selected from a population using some chance process.... SRS, stratified sample, cluster sample * each unit in population should have same probability of being chosen

indicator

a sign of the presence or absence (and usually the level) of the concept we're studying

hypothesis

a statement proposing a relationship between two or more variables... a testable expectation about a reality that follows from a theory. It is a statement of something that ought to be observed in the real world if the theory is correct

sample

a subset of the population of interest. you typically only gather data from a sample, want to be representative of the population, same distribution of characteristics

dependent variable

a variable that depends on, or is influenced by, another variable (our independent variable)

independent variable

a variable that helps explain some phenomenon; is a cause of the phenomenon we are observing

intervening variable

a variable that impacts the dependent variable and is itself affected by an independent variable. It helps explain how your x influences your Y

variable

an attribute of some subject that is capable of assuming any of a set of values

antecedent variable

an independent variable that precedes another independent variable in time, and affects that independent variable

random vs systematic measurement errors

bad: random measurement error in an independent variable, or systematic measurement error in your DV or IV (all create bias) not so bad: random measurement in a dependent variable

qualitative data

can cover a wider range of phenomena, highlights the complexity of social phenomena, better suited to idiographic explanations

causality

causality is partial, inconsistent across units. Partial means that there are many IVs that cause variation in each DV. Inconsistent across units means that a constant amount of variation in a specific IV does not have the same impact of the DV in each individual case. We seek to measure the average impact of some IV on some DV

problems with constructing survey questions

chain order, question order, question wording

major steps in content analysis

creating a sample, coding

modern study of politics

empirical study of politics- began with behavioral revolution of post WWII, new focus on the political behavior of individuals and groups

valid?

face validity: does a measure make sense on its face? is it reasonable? content validity: does a measure cover all of the dimensions we are interested in? does it cover the full range of meanings included within the concept? criterion / predictive validity: does the measure help us predict things we would expect it to help us predict?

traditional study of politics

focused on the formal institutions of politics and the formal relationships between those institutions

research typically involves..

formulating a question about a behavior, structure, or institution, and then seeking to answer the question. You propose to answer the question by developing a theory that helps address your question, and then developing a hypothesis that helps you test your theory.

importante

from a sample, we get an estimate about a population... *use random sampling

descriptive

goal is to clearly define and accurately measure some political phenomenon

sources of measurement error

lying, misinterpretation of questions, subjective interpretation of survey questions, etc

quantitative data

more precise, makes comparisons easier, good for hypothesis testing, less subjective, easily allows replication, better suited for nomothetic explanations

inductive reasoning

moving from a set of specific observations to the discovery of a pattern that represents some degree of order among the observations

selection bias (sampling bias)

occurs when certain types in the population are less/more likely to be chosen from the sample

empirical study

one grounded in the analyzing of evidence we observe (data), has expanded beyond behavior

Nomothetic explanation of partiality/causality

partial explanation of the general. nomothetic explanations seek to identify one causal factor that affect a general class of events, outcomes, or conditions

nomothetic explanation

partial explanation of the general. seek to identify one causal factor that affect a general class of events, outcomes, or conditions

methods of theory development

question-> theory -> testable hypothesis

gen.

research in political science is interested in factors that influence the probability* of some outcome, or influence the expected* level of some outcome

close ended

respondents have to choose from the options provided

reliable?

test-restest, split-half

validity

the correspondence between the measurements an instrument yields and the concept it is supposed to measure, used to describe the instrument or the measurements

measurement error

the difference between the real value of the attribute we want to measure and the value we come up with in the process of measuring

accuracy in measurement- reliability

the extent to which an instrument yields the same results in repeated trials

deductive reasoning

the process of reasoning from general principles to derive particular information or specific predictions

open ended survey questions

the researcher does not provide fixed response options

theory

the world is infinitely complex, to make sense of some specific aspect of the world, a theory zeroes in on some of the more important factors, and provides a causal story

idiographic

thorough explanation of the specific. Idiographic explanations seek to explain all (or almost all) of the idiosyncratic factors that caused a specific event, outcome, or condition

x--> y

variation in the independent variable causes variation in the dependent variable

literature review

what: somewhat brief written review of the existing research and theories relevant to your topic purpose: to discuss the existing body/bodies of knowledge to which your research would add, and situate your project in that body (knowledge is cumulative)

social desirability bias

when a question, even inadvertently, puts pressure on people to respond in a particular way