important terms

MEASUREMENT

The assignment of numbers to objects

according to rules.

NUMBERS

Logical symbols used to represent

objects, having the properties of

Identity, Order, and Additivity.

OBJECTS

Items in the natural world: All have

the property of Identity, some also can

be Ordered, while others have

Identity, Order, and Additivity.

IDENTITY

Unique properties of the elements can

be distinguished from one another.

ORDER

Elements can be ordered from

smallest to largest on some dimension

all share.

ADDITIVITY

Elements can be added to one another

producing a unique, and meaningful,

third element.

NOMINAL SCALE OF MEASUREMENT

Numbers have been assigned to

objects, but the numbers and objects

only share the property of Identity.

ORDINAL SCALE OF MEASUREMENT

Numbers have been assigned to

objects, but the numbers and objects

only share the properties of Identity

and Order.

INTERVAL SCALE OF MEASUREMENT

Numbers have been assigned to

objects and the numbers and objects

share the properties of Identity, Order

and Additivity. Equal Intervals

between numbers correspond to equal

differences in magnitudes of the

objects but “0” corresponds to an

arbitrary level of the variable.

RATIO SCALE OF MEASUREMENT

Numbers have been assigned to

objects and the numbers and objects

share the properties of Identity, Orderand Additivity. Equal Intervals

between numbers correspond to equal

differences in magnitudes of the

objects and “0” corresponds to natural

absence of the variable.

DAVID HUME

18th Century philosopher who defined

the foundation of causation in the

behavioral or social sciences as

“Regularity of Sequence.”

REGULARITY OF SEQUENCE

David Hume’s criterion for causation:

One must be able to manipulate the

hypothesized cause, and show the

effect occurs when the cause is

presented, and the effect is absent

when the cause is absent.

EXPERIMENTAL DESIGN

Research in which the investigator

manipulates the presence and absence

of the cause (IV) and observes the

“regular changes” in the effect (DV).

NON EXPERIMENTAL DESIGN

Research in which the investigator

measures, but does not manipulate, an

IV, while also measuring an outcome

or DV. Offers no evidence for

regularity of sequence/causation.

NON EXPERIMENTAL DESIGN

 Research in which the investigator

measures, but does not manipulate, an

IV, while also measuring an outcome

or DV. Offers no evidence for

regularity of sequence/causation.

EXPERIMENTAL VARIABLE

A variable that can be manipulated in

a research design. It is not an inherent

property of the research participants

(e.g. exercise, knowledge, sleep, diet,

etc.).

QUASI EXPERIMENTAL VARIABLE

A variable that CANNOT be

manipulated in a research design. It is

an inherent property of the research

participants (e.g. sex, age, race, etc.).

FACTORIAL DESIGN

An experiment in which two or more

independent variables are used, with

each having two or more levels.

INTERACTION

The effects of one independent

variable are different at different

levels of the other independent variable.

MAIN EFFECT MEANS

The mean of all subjects on one level

of an independent variable, ignoring

the classification across a second

independent variable.

ARTIFACTUAL MAIN EFFECT MEANS

A significant difference between row

or column means when a statistically

significant interaction is also present.

DEGREES OF FREEDOM

The number of pieces of information

free to vary in a computation.

TYPE I ERROR

The probability of rejecting the null

hypothesis when it is true.

TYPE II ERROR

The probability that you will fail to

reject the null hypothesis when it is

false.

POWER

The probability of rejecting the null

hypothesis when it is false.

STANDARD ERROR OF ESTIMATE

The average error of predicting Y from

X using a regression equation.

STRENGTH OF EFFECT

The amount of variance in the

dependent variable that can be

predicted from knowledge of the

independent variable, in a group

comparison statistical test

CHI SQUARE

Used to determine if a relationship

exists between two variables

measured at the nominal scale of

measurement.

PHI COEFFICIENT

Used to measure the relationship

between two variables measured at

the nominal scale of measurement.

PEARSON PRODUCT MOMENT CORRELATION COEFFICIENT

Used to describe the relationship

between two variables measured at

the interval or ratio scale of

measurement.

SPEARMAN RANK ORDER CORRELATION COEFFICIENT

Used to describe the relationship

between two variables measured at

the ordinal scale of measurement.

STANDARD DEVIATION OF Y

The average error of predicting Y using the mean of Y as the predictor

for everyone.

REGRESSION

The use of a linear function to predict

values of Y from X.

POINT BISERIAL CORRELATION

Used to describe the relationship

between a nominal scale variable and

an interval or ratio scale of.

CHI SQUARE EXPECTED FREQUENCIES

The frequencies of a contingency table

that are specified by the null

hypothesis for a Chi Square test.

CHI SQUARE OBSERVED FREQUENCIES

The frequencies found in sample data

for the cells of a contingency table.

CHI SQUARE MARGINAL FREQUENCIES

The row totals and column totals of

the observed frequencies in a

contingency table.

GOODNESS OF FIT CHI SQUARE

Chi square test used to examine

specific predictions by a experimenter

on a nominal variable categorized on

a single dimension.

SCALE OF MEASUREMENT APPROPRIATE TO CHI SQUARE

Numbers have been assigned to objects

but the objects share only the property

of identity with the numbering system.

BIVARIATE DISTRIBUTION

Pairs of measures on two variables

collected on the same subjects.

POSITIVE CORRELATION

As values of X increase, values of Y

increase.

NEGATIVE CORRELATIONAL RELATIONSHIP

As values of X increase, values of Y

decrease.

A CORRELATION EQUAL TO ZERO

None of the variance in Y can be

predicted from X.

A CORRELATION EQUAL TO +1.0 OR -1.0

100% of the variance in Y can be

predicted from X.

REGRESSION EQUATION

Linear function that predicts Y from X

using slope and intercept values.

SLOPE

The rate of change in Y associated with a 1-unit increase in X.

INTERCEPT

The value of Y when X is equal to zero.

RESIDUAL

The difference between observed

values of Y and those predicted from

the regression equation.

STANDARD ERROR OF ESTIMATE

The average error of predicting Y from

X using the regression equation.

COLLINEARITY

High correlation of a predictor

variable with one or more other

predictor variables in multiple

regression.

BACKWARD STEPWISE ENTRY

The entry of all predictors into a

multiple regression analysis in a single

step, and then the use of

mathematically controlled processes

to remove the weakest, non-

statistically significant predictors.

HIERARCHICAL ENTRY

Investigator controlled entry of

predictors into a multiple regression

analysis based on past research

findings and theoretical predictions.

FORWARD STEPWISE ENTRY

Mathematically controlled entry of

predictor variables into a multiple

regression analysis based on the

strength of their relationship with the

Y variable.

FORCED ENTRY

The entry of all predictors into a

multiple regression analysis in a single

step.

STANDARDIZED BETAS

Slopes converted to Z-score units, which

provide an unbiased basis for comparing

the contributions of multiple predictors to

changes in Y.

BETAS

Slopes in raw score units. Their size reflectsthe mean and standard deviation of the raw

score scale used to measure the X variable,

rather than its effect on changes in Y.

R SQUARE CHANGE

The increase in prediction associated with

each X variable when it is entered into a

multiple regression equation.