scopes quiz #3

hypothesis

a statement of a relationship between two concepts

hypothesis characteristics

conditional, universal or statistical, open to testing, susceptible to testing

types of tautologies problematic for hypothesis testing

conceptual and measurement

deductive reasoning

broad to specific

inductive reasoning

spedici to broad

theory

a set of interrelated concepts from which hypotheses can be derived. acts as the logical connection between hypotheses and the empirical world, but is never directly tested.

cross tabular analysis

used to show a relationship between two variables (but only useful for variables with small numbers, make sure to convert the raw data table in column percents)

how do you move from a scatterplot to a cross tab?

by reducing the level of measurement

standard experimental design process

random assignment, treatment group with treatment outcome vs control group outcome

random assignment

subjects are assigned to the treatment and control groups randomly to help rule out alternative explanations. you want to control for characteristics that subjects bring to the experiment. cook and campbell said that random assignment is the great ceteris paribus of casual inference.

How do researchers control the independent variable?

they manipulate the treatment directly, make the test groups as similar as possible, use blind experimentation

what does blind and double-blind experimentation do?

it allows for standardization and prevents new variables from unintentionally effecting the outcome of an experiment.

treatment effect

the difference in outcomes across the treatment and control groups

internal validity

Internal validity is the extent to which a study accurately proves a cause-and-effect relationship between its variables, free from the influence of confounding or outside factors.

external validity

the observed effects of a study can be generalized to the real world

tradeoff between internal and external validity

experiments are high on internal validity but tend to be low on external validity

survey experiment

respondents evaluate hypothetical situations or vignettes

field experiments

experiments conducted in the field, offering greater realism

natural experiments

experiments created via nature

suprious assocation

the apparent relationship between two variables that are not related, but a relationship between them is produced by their relationships with a third variable.

how do you address spurious association?

statistical solutions, building Z variables into measures of x or y variables, random assignment, matching (turning Z into a constant)

simpson’s paradox

a statistical phenomenon where a trend appears in different groups of data, but disappears or reverse when those groups are combined together