PUP3002 module D

ill-informed skepticism

disbelief or distrust in a scientific claim that is based less upon rigorous evaluations of competing arguments and evidence and more upon political orientation or vested interest.

ill-informed adherence

the adoption of, or belief in, a claim is based less upon rigorous evaluations of competing arguments and evidence and more upon political orientation or vested interest.

causal interference

the process of drawing a conclusion about causal relationships between various inputs and policy outputs and outcomes.

repetition effect

any change in a subject’s performance due to repeating the experimental condition.

fundamental problem of causal interference

the difficulty that we face in experimental situations where we cannot expose the same individual to both treatment and control at the exact same moment in time.

selection effect 

when non-random selection into treatment would likely interfere with our ability to draw a valid causal inference

confounding variables

when we have a set of variables that are correlated with both our treatment and our potential outcome

spurious relationship

when a relationship between two variables appears to exist, due to the confounding presence of a third variable,

minimum conditions

ensuring all other variables that are correlated with both our treatment and our outcome are randomly distributed across the groups in order to draw a valid causal correlation.

observational study

the assignment of subjects to the treatment and control groups is not under the analyst’s control. done when a classic experiment is not feasible, possible, or ethical.

internal validity

the likelihood or level of confidence that a causal inference drawn from an analysis reflects the “true” underlying causal relationship.

external validity

whether the findings from a study based upon a sample or an experiment can be extended or generalized to a larger population.

central tendency

information about the location of the middle and shape of the distribution of a data set

descriptive statistics 

help describe both the average response and dispersion of opinion around the average for a given population of stakeholders or clientele. 

correlation

whether and how strongly two variables tend to co-vary

scientific uncertainty

the process of estimating a precise expression of uncertainty in some claim that derives primarily from three sources of error: measurement error, random error, and sampling error.

random error

the stochastic or random probability that some events are simply not perfectly predictable.

population

the entire set of individuals or cases with which an analyst is interested in examining or learning abouts

sample

a subset of the population

parameters

characteristics of a population

statistics

the corresponding characteristics from a sample

biased statistics

statistics that do not reflect the true value of the population parameter due to error

hypothesis testing

a process in which scientists attempt to determine whether sufficient evidence for the research hypothesis has been presented to reject the null hypothesis.

type I error

when one rejects a true null hypothesis. false positive

type II error

when one fails to reject a false null hypothesis. false negative.

policy-oriented learning

“the relatively enduring alterations of thought … that are concerned with the attainment or revision of the precepts of one’s belief system”