UIUC STAT 100 - Exam 1 (Rao)- Annie Szkapiak

Data

The representation of someone or something

Tidy Data

The way of mapping the real world to a data set

Observation Unit

The thing we are interested in

Observation

Unique observational unit that we are interested in (per ROW)

Attribute

Unique aspect of each observation that we are interested in (per COLUM)

Measures

The way in which we collect information about the obeservation

Quantitive data

values of an attribute for an observation are NUMBERS representing a quantity of something

Categorical data

values of an attribute for an observation are selected from a SET of different category LABELS

Text data

values of an attribute for an observation is text (either single words or sentences)

Rating scale data

refers to data in which the values of an attribute for an observation are selected from a predetermined rating scale.

Time series data

refers to data collected over a series of dates

Measurement

is the process by which we collect information about observational units

Reliability

refers to the extent to which the data you collect from a measure truly represents an reflects the real world characteristics of the observational unit.

Item writing

is the process of writing survey questions and answer choices (multiple choice, true/false, constructed response)

6 common principles to item writing

1. avoid colloquialisms

2. keep question short as possible

3. avoid repetitive wording

4. avoid absolute modifiers

5. avoid ambiguous answer choices

6. avoid opposing statements

Causal Inference

is an inference about which factor or factors may be responsible for causing an effect on some observed outcome

Outcome

what actually happened in the real world

Counterfactual

What would have happened in the parallel universe

(what if questions)

Causal graph

Graphic depiction of a cause and effect relationship, including other relevant factors that may change the outcome

Moderators

factors that change the effect a cause has on an outcome. When moderators are present in a relationship, we say that there is an interaction effect.

Mediators

factors that come in between an cause and the outcome

Confounders

factors that themselves have a causal effect on both the main causal factor and outcome we are trying to study

Control

is an observational unit that is as like another observational unit as possible, but who was exposed to a different treatment.

Treatments

different conditions related to the causal factor we are interested in studying

Case Control

study in which individual control observations are matched to each individual treatment observation. - Match Observations

Cohort Control

tudy in which an entire group of control observations are matched to the group of treatment observations based on their aggregate characteristics. - Match Groups

(includes averages)

Random assignment

A strategy that utilizes random components to determine which observational units receive a treatment, and which serve as a control.

Matched pairs

strategy by which two people are identified as very similar to each other in terms of the important factors, and then one person in the pair is randomly assigned to the treatment group, while the other person in the pair is randomly assigned to the control group

Block randomization

is a strategy by which an entire GROUP of similar people are blocked together, and then some are randomly assigned to the treatment group, while the others are randomly assigned to the control group.

Average Treatment Effect

computed by taking the average of the difference between ALL of the treatments and their controls

Internal validity

the degree to which a study's design supports making a causal inference. The internal validity of a study depends on the extent to which the observational units in the control group are like the observational units in the control group.