Poli 3000 Midterm

Empirical

questions of fact that can be answered through information we collect

Normative

Questions of value answered through logic proofs ad philosophical debate

Applied research

identify solutions to real world problems

Theory-oriented research

based on existing theories and helps better understand political phenomena. 

Normative Theory

identify moral principles that make society better

Formal Theory

identify the implications of people acting rationally to maximize their self-interest

Explanatory Idea

when an idea describes a casual process that connects one set of facts with another set of facts

Testable Idea

researcher describes a set of conditions under which the idea should be rejected

Conceptual Questions

a question expressed using ideas

Concrete questions

a question expressed using tangible properties

conceptual definition

clearly describes the concepts measurable properties and specifies the unit of analysis

Operational definition

Describes the INSTRUMENT to be used in measuring the concept and putting a conceptual definition into practice

Conceptual dimension

Defined by a set of concrete traits of a similar type

Multidimensional concept

Has two or more distinct conceptual dimensions

Individual-level unit of analysis

When a concept describes a phenomenon at its lowest possible level

Aggregate-level unit of analysis

A collection of individual entities

Cross-level analysis

When data collected at one level of analysis is used to better understand what’s happening at another level of analysis

Ecological fallacy

When an aggregate-level phenomenon is used to make inferences at the individual level

Feeling thermometer

A visual aid that helps people QUANTIFY their feelings about people, ideas, and institutions

Systematic measurement error

Introduces consistent, chronic distortion into an empirical measurement, often called measurement bias

Hawthorne Effect

A phenomenon where one accidentally measures a subject’s response to the knowledge that they are being studied

Random measurement error

Introduces haphazard/chaotic distortion into measurement process, like fatigue or unavoidable distractions

Reliability

The extent to which it is a consistent measurement of a concept

Test- retest method

Investigator applies the measure once and then applies it again at a later time to the same units of analysis (reliability)

Alternative-form method

Investigator administers two different but equivalent versions of the instrument (reliability)

Split-half method

Internal consistency approach based on the idea that an operation measurement obtained from half of a scale’s items should be the same as the measure obtained from the other half (reliability)

Cronbach’s alpha

Compares consistency between pairs of individual items and provides an overall reading of inter-item correlation and a measure’s reliability (reliability)

Validity

the extent to which it records the true value of the intended characteristics, no systemic error/no bias

Face validity approach

Investigator uses informed judgement to determine whether an operational procedure is measuring what it is supposed to measure (validity)

Construct validity approach

Investigator examines the empirical relationships between a measurement and other concepts to which it should be related (Validity)

Codebook

Where one can find variable names, descriptions, and info about a data set

Cross-sectional dataset

A dataset that compiles information collected at one time to study properties that vary across the units of analysis

Cross-sectional study

Contain information on units of analysis measured at one point in time

Time-series dataset

A dataset that compiles information collected at different time internals to study properties that vary over time

Panel study

Contains information on the same units of analysis measured at two or more points in time

Additive index

a summation of ordinal variables, each of which is coded identically, and all of which are measures of the same concept

Automated Content Analysis

a method of transforming massive amounts of data into interval-level measures of important political concepts

Central Tendency

the variable’s typical or average value; measured by either mode, median, or mean

Centrality

the importance of nodes within the network

Cumulative Percentage

records the percentage of cases at or below any given value of the variable

Interval-level Variables

communicate the exact amount of the characteristic being measured, can be used to separate cases in groups, rank cases, and calculate differences among cases 

Likert Scale

an additive index of 5- or 7-value ordinal variables, each of which captures the strength and direction of agreement with a declarative statement

Mean Centering

involves subtracting the mean value of the variable from each observed value

Positive Skew

Distributions with a longer, or skinnier, right-hand tail

Negative Skew

Distributions with a skinnier left-hand tail

Network Analysis

used to transform social connections among people into quantitative measures and map social structures that explain political phenomena

Nodes

Units that make connections are called nodes

Edges

The connections of nodes

Nominal-Level Variable

communicates differences between units of analysis on the characteristics being measured. The least precise and used to categorize into groups

Ordinal-Level Variable

communicates relative differences between units of analysis. More precise than nominal and can be ranked 

Resistant measure of Central Tendency

what the median is called, and gives a more faithful idea of the true center than the mean

Standardizing

tells you how many standard deviations above or below the mean you are

Dependent Variable

the variable that represents the effect in a casual explanation

Independent Variable

variable that represents a casual factor in an explanation

Model

a simplified, abstract representation of some larger and more complicated subjects

Rational Actor

makes deliberate decisions to advance his or her own interests

Social-Psychological Actor

makes decisions based on gut feelings rather than thoughtful analysis

Probabilistic theories

means there are always expectations to the rules

Deterministic theories

explanations that leave no room for error 

Casual Mechanism 

an internal link that acts as a go-between or mediator between an independent variable and a dependent variable 

Research Hypothesis

a testable statement about the empirical relationship between an independent variable and a dependent variable

Cross-Tabulation

A table that shows the distribution of cases across the values of a dependent variable for cases that have different values on an independent variable 

Mean Comparison Table

a table that shows the mean of a dependent variable for cases that have different values on an independent variable

Direct relationship

a relationship that runs in a positive direction: An increase in the independent variable is associate with an increase in the dependent variable 

Inverse relationship

a relationship that runs in a negative direction: An increase in values of the independent variable is associated with a decrease in the dependent variable

Linear Relationship

an increase in the independent variable is associated with a consistent increase or decrease in the dependent variable

Negative relationship

the typical value of a dependent variable decreases by the same amount for each unit change in the independent variable

Curvilinear Relationship

the relationship between independent variable and dependent variable depends on which interval or range of the independent variable is being examined 

positive relationship

as an increase in the independent variable occasions an increase in the dependent variable

Research design

an overall set of procedures for evaluating the effect of an independent variable on a dependent variable

Experimental design

ensures that the test group and control group are the same in every way, except one —— the independent variable 

Selection Bias

When nonrandom processes determine the composition of the test and control groups. 

Post-treatment measurement

the dependent variable is measured again for both groups

Internal Validity

the effect of the independent variable on the dependent variable is isolated from other olausible explanations

External Validity 

the results of a study can be generalized —- its findings can be applied to situations in the natural world

Sampling Frame

is the population the researcher wants to analyze and the source from which the samples are drawn.

Response Bias

when some cases in the sample are more likely than others to be measured

Cluster samples

used when population of interest is hard to define but occupies a definite geography

Snowball sample

researcher asks the people they select for analysis to help identify others who could participate in the research

Purposive Sample

researcher selects cases that offer the best test of the research hypothesis

Post hoc theorizing

changing the hypothesis and underlying theory after collecting data in order to predict results in line with the data. Scientific analysis/ fruad

P-hacking

involves purposely manipulating statistical analysis to achieve statistically significant results

Compositional difference

is any characteristic that varies across categories of an independent variable

controlled comparison design 

the researcher to observe the effect of the independent variable on the dependent variable while holding constant other plausible causes of the dependent variable 

controlled comparison

examining the relationship between an independent and dependent variable, while holding constant other variables suggested by rival explanations

Zero-order relationship

an overall association between two variables that does not take into account other possible differences between the cases being studied

Confounder

a pretreatment variable that is related to both the treatment and the outcome

Controlled comparison table 

presents a cross-tabulation between an independent and dependent variable for each of the control variable 

Partial relationship

summarizes a relationship between two variables after accounting or a rival variable

Spurious Relationship

The empirical association between X and Y turns out to be completely coincidental —not causal at all

Additive Relationship

the control variable is a cause of the dependent variable but defines a small compositional differences across values of the independent variable

Interaction Relationship

the relationship between the independent variable and the dependent variable depends on the value of the control variable 

Matching methods

attempts to replicate random assignment in an observational setting

Difference-in-differences design

when they suspect that variables other than the independent variable have changed in the before and after time periods, but these “other factors” are numerous or not clearly known

Inferential statistics

a set of procedures for deciding how closely a relationship we observe in a sample corresponds to the unobserved relationship in the population from which the sample is drawn 

Sample statistic

an estimate of a population parameter, based on a sample drawn from the population

Random sampling error 

defined as the extent to which a sample statistic differs from a population parameter

Standard Error

statistic tells us how much we can expect a sample statistic to vary from the population parameter

Sampling distribution

It shows the expected distribution of sample statistics, like sample proportions or sample means

Central limit theorem

established statistical rule that tells us that if we were to take an infinite number of samples of size n from a population N members, the sample means will follow a normal distribution