Research Methods Correlation & Replication

Association Verbs

verbs used in each case are association verbs

Bivariate Correlation

an association that involves EXACTLY two variables

Categorical Variable

values fall in either one category or another and are not numerical.

Quantitative Variable

values fall in a numerical category (use scatterplot)

Mean

the arithmetic average

Correlational Studies

involve measuring both variables and support an association claim

Experiments

have one variable manipulated which is appropriate for testing a causal claim

Association Claim

describes the relationship between two measured variables

Effect Size

describes the strength of a relationship between two or more variables

Replication

estimating the population association is to conduct the study again and find multiple estimates

Outlier

Extreme scores, ones that stand out from the rest

Restriction Range

if there is not a full range of score on one of the variables in the association, it can make the correlation appear smaller than it really is

Correction for Restriction of Range

estimates the full set of scores based on what we know about an existing, restricted set, and then recomputes the correlation

Curvilinear Association

the relationship between two variables is not a straight line

Covariance of Cause and Effect

The results must show a correlation, or association, between the cause variable and the effect variable

Temporal Precedence

The method must ensure that the cause variable preceded the effect variable, it must come first in time

Directionality Problem

when we don’t know which variable came first

Third-Variable Problem

when we come up with an alternative explanation for the association between two variables, that alternative is some lurking third variable

“nuisance” variables, outside variables

Spurious Association

the bivariate correlation is there, but only because of some third variable

External Validity

the size of the sample does not matter as much as the way the sample was selected from the population of interest

ask whether the association generalizes to other kinds of the same topic, for example like other kinds of praise (from teachers or other adults), ask if the study was randomly sampled

Internal Validity

There must be no plausible alternative explanations for the relationship between the two variables

longitudinal designs help establish temporal precedence and multiple-regression analysis helps rule out third variables, providing evidence for internal validity

Moderator

sub-group of people (external validity)

the variable can change the relationship between the other two variables by making it more intense or less intense

grouping, answers “for whom” or “in what situation”

Homoscedastic

Heteroscedastic

Regression Equation

described the numeric relationship between the variables

Best fitting line

lets us predict scores of one variable given scores of the other in their measured units

Linear Regression

regression equation below predicts a dependent variable (DV) with one independent variable (IV)

Multivariable Regression

includes more than one predictor variable

Continuous Variables

betas represent slopes

Group Variables

beta’s represent difference scores to a reference level

Root mean squared error (RMSE)

provides a measure of the overall spread of data points away from the prediction (answers how much variability is left in Y)

R2

indicates the proportional gain in variability accounted for in predicting y from x rather than the mean of y

third-variable problem

is there a C variable that is associated with both A and B, independently?

Covariance

Do the results show that the variables are correlated?

Cross-sectional correlations

the correlation of variable when measured at the same time

Autocorrelations

the correlation of each variable with itself when measured at different times

Cross-Lagged Correlations

the correlation of an earlier measure of one variable with a later measure of another variable

interpreting intercept

value of dependent variable when all predictor variable are 0

interpreting beta

estimated difference in outcome for 1 level difference of predictor variable

Longitudinal Design

measures the same variables in the same people at several points in time

Multiple Regression

helps rule out some third variables and addresses internal validity concerns (does not establish causation)

Criterion Variable (dependent variable)

the variable researchers are most interested in understanding or predicting

Predictor variable (independent variable)

the variable that researchers are measuring for the study, change as the criterion variable is manipulated

Beta Basics

there is one beta value for each predictor variable, positive = positive r, zero = no relationship, higher = stronger relationship

P-Value

when greater than .05, the beta is considered not significant, and you can infer that its 95% CI DOES contain zero

coefficient b

represents an unstandardized coefficient

Pattern and Parsimony Approach

researchers can investigate causality by using a variety of correlational studies that all point in a single, causal direction

Parsimony

the degree to which a scientific theory provides the simplest explanation of some phenomenon

Mediator

between two of the variables in the study, why? (internal validity)

modifies the strength of the association between the independent (X) variable and the dependent (Y) variable

mechanisms, answers “why"?”

construct validity

appropriate to interrogate the construct validity of the variables in the study by asking how well each variable was measured

Statistical validity

ask about the point estimates and confidence intervals and ask whether the study has been replicated

Frequency

one measured variable

Association

multiple measured variables

causal

manipulated independent variables, measured dependent variable

data

a set of observations representing the values of some variable, collected from one or more research studies

Prediction/Hypothesis

a way of stating the specific outcome in the data that the researcher expects to observe if the theory is accurate

data

a set of observations representing the values of some variable, collected from one or more research studies

Theory

statement or set of statements that describes general principles about how variables relate to one another