Variable
any characteristic of some event, object, or person that may vary
Dependent Variable (DV)
The variable that is measured to determine if the independent variable has any effect
Independent Variable (IV)
The variable said to be responsible for some sort of effect
Categorical Variable
a variable that categories which are different in quality, not in numerical magnitude
Quantitative Variable
a variable that has a theoretically infinite number of intermediate values, and differs in magnitude
Discrete
possible values are from a specific range (0, 1, 2...)
Continuous
possible values are from an infinite number of possibilities (8.654, 7.4563, ...)
Nominal Scale of Measurement
variable has discrete, mutually exclusive categories. No order. (ex; eye color)
Ordinal Scale of Measurement
variable has discrete categories that have specific order. Magnitude between categories is not necessarily equal. (ex; letter grades A, B, C, D)
Interval Scale of Measurement
variable values have a specific order and there is equal numerical magnitude between values.
Ratio Scale of Measurement
like interval (values have specific order and there is equal numerical magnitude between values), but also has the addition of a meaningful "zero point", where zero means the absence of some property.
Bar Graph
discrete scores, usually from nominal or ordinal scale
Pie Chart
slice of the pie represents the percentage of observations in a category
Histogram
small or large range of continuous scores from an interval or ratio scale
Polygon
large range (population or close to population) of continuous scores from an interval or ratio scale. Created by joining the top bars of a histogram
Operationalization
The process of specifying how something will be measured
True Experiment
IV (manipulated)
DV
Random assignment to groups
Can make causal inference
Quasi-Experiment
IV (might be manipulated)
DV
NO random assignment (pre-existing groups)
Some causal inference ok
Correlational Design
Variable X
Variable Y
Relation between two variables
Can't make causal inference
Population
larger than a sample, parameter
Sample
Subset of the population (smaller than population), statistic, good samples are representative of the larger population
Simple Random Sampling
A simple random sample of n subjects from a population is one in which each possible sample of that size has the same probability (chance) of being selected
Systematic Random Sampling
Ex; choose every 3rd person on a list of names
Stratified Random Sampling
Divide people into different groups or strata (gender, race, etc)
disproportional- sample is not representative of the amount in the population
proportional- sample is representative of the amount in the population
Cluster Sampling
Divide the population into clusters(city blocks)
randomly select blocks (clusters)
and sample everyone in those clusters
Selecting Appropriate Measure of Central Tendency
A. Interval / ratio ---> mean
B. Ordinal ---> median
C. Nominal ---> mode
(** can calculate measure of CT from tier below but not above)
Outlier
data points that are extremely greater than or less than the mean.
Skewed Left
longer left tail, most observations are medium/large (ex; age of death from natural causes).
Mean is most effected
mean < median < mode
Skewed Right
longer right tail, most observations are small/medium (ex; salary)
mode < median < mean
Range
Difference between that highest and lowest scores of a distribution.
Standard Deviation
The average by which a typical score deviates from the mean
Variance
Square of the Standard Deviation
Probability
The chances of an event occurring. With a random sample or or randomized experiment, the probability an observation has a particular outcome is the proportion of times that outcome would occur in a very long sequence of observations
Joint Probabilities
The probability of selecting from the sample space an element where two conditions are present at the same time P(A and B)
(venn diagram, that overlaps)
Disjoint Probabilities
Probability that events cannot occur at the same time (aka they are mutually exclusive)
Addition Rules
Disjoint Probabilities:
P(A or B)= P(A) + P(B)
Joint Probabilities:
P(A or B)= P(A) + P(B) - P (A and B)
Independent Probabilities
Probabilities are independent if the probability of one event occurring does not affect the probability the other event will occur
Multiplication Rule
Multiplication rule for independent events:
P(A and B)= P(A)*P(B)
Conditional Probability
What is the probability of A given B?
P (A I B)= P(A and B)/P(B)
** if events are independent P(A I B)= P(A) and you can use the general multiplication rule for independent events
Normal Distribution
68-95-99
Total area under curve=1 and mean=0
Symmetric
Asymptotic (extreme ends of distribution are approaching zero but never actually reach zero)
Unimodal
% of scores under curve is predictable
Why is Normal important?
Setting the total area to 1 allows us to use Z scores to calculate the relative location of an individual in a data set
Standardized Scores
to "standardize" is to transform scores that are on different metrics into a single metric so that they can be compared
Z-scores
Standardized scores in terms of standard deviation units
z= y-mean/s