Adjusted R2 Multiple coefficient of determination adjusted for sample size and number of independent variables. It is a better estimate of the parameter than is the unadjusted sample multiple coefficient of determination.
Analyses Used to describe scores on single measures and to identify relationships that may exist between scores across different measures; typically involve the use of statistics.
Assignment error Results because the random allocation of cases to levels of an independent variable typically does not exactly equalize scores on the dependent variable across groups.
Attenuation due to unreliability Formula that estimates the degree to which unreliability in either or both variables reduces a correlation coefficient relative to perfectly reliable measurement of the variables.
Bar chart A graph thatreports the values of a discrete variable on one axis and the frequency (number and/or percentage) of cases on the other.
Beta coefficient A standardized (mean = 0.0, standard deviation = 1.0 of all variables in the model) regression coefficient. In simple regression, the beta coefficient is identical to the simple correlation coefficient.
Between-cases design Design that uses two or more cases or groups of cases; each case or group experiences different levels of an independent variable. Effect of an independent variable is assessed by comparing dependent variable scores across the cases or groups of cases.
Biased relationship An observed relationship between an independent and dependent variable that under- or overstates the causal relationship. Occurs if a nuisance variable is causally related to the dependent variable and is related to the independent variable(s).
Binomial probability distribution Discrete probability distribution that results from a set of trials where the values of each trial are independent, the probability remains constant from trial to trial, and each trial can lead to one of two values.
Bouncing betas Instability in sample standardized partial regression coefficients resulting from multicollinearity.
Boundary condition Synonym for moderator variable. A relationship between an independent and a dependent variable depends on the value (boundary) of another independent variable.
Box-score method In meta-analysis, the use of statistical significance values as the dependent variable. Its use is not recommended.
Cases Entities (e.g., individuals, groups, organizations, transactions) investigated in research.
Causal conceptual relationship A relationship in which variation in the independent construct is responsible for variation in the dependent construct.