MUED 6440 Ch. 1 Flashcards
A square symmetrical matrix; each row and each column represents a definite variable; located at each intersection of a row and column is a bivariate correlation between the two variables
Correlation Matrix
Organization of raw scores, or data, where rows represents cases (participants) and columns represent variables
Data Matrix
The collection of sample means for all possible random samples of a given size (n) that are obtained from a population. The values in this distribution are not scores, but statistics (sample means); a distribution of statistics is called a sampling distribution.
Distribution of sample means
The size of the treatment effect the researcher wishes to detect with respect to a given level of power; denoted as ES or partial η²
Effect size
Statistical tests of predictions made about a sample
Hypothesis testing
In hypothesis testing, the pre-established probability of being incorrect; also known as the alpha level
Level of significance
In hypothesis testing, the pre-established probability of being incorrect; also known as the level of significance
Alpha level
An extension of scalar algebra where mathematical operations are performed on an ordered array of numerical values
Matrix algebra
The arithmetic average of a set of scores
Mean
The most frequently occurring score in a distribution
Mode
The score in the distribution that divides the upper 50% of scores from the lower 50%
Median
Statistical analysis that involves more than one dependent variable
Multivariate statistical analysis
Predicts that the only differences existing between groups are chance differences that represent only random sampleing error
Null hypothesis
Perfect nonassociation between variables
Orthogonality
The appropriate measure of correlation when variables are expressed as scores
Pearson r
A value that indicates the percentage of scores that fall at or below a given score
Percentile rank
The probability of rejecting H₀ when H₀ is in fact false; equal to 1 - β
Power
In significance testing, the probability of being incorrect in drawing a conclusion about the populations
Probability level
One half of the difference between the 3rd quartile (i.e., the 75th percentile) and the 1st quartile (i.e., the 25th percentile)
Quartile deviation range
Prediction that one method or group is expected to be better than the other; in other words,. that the two group means are not equal and therefore represent a true difference in the population; also known as the alternative hypothesis
Research hypothesis
Prediction that one method or group is expected to be better than the other; in other words,. that the two group means are not equal and therefore represent a true difference in the population; also known as the research hypothesis
Alternative hypothesis
The expected, chance variation among sample means
Sampling error
Researcher assigns priority for the entry of variables into the equation (solution); first variable to be entered is assigned both its unique variance and any overlapping variance it has with other IVs; upon entry, lower-priority variables are then assigned their unique variance and any remaining overlapping variance
Sequential analysis
Procedures used to determine if the difference between sample means is substantial enough to justify concluding that a true difference exists in the population as well
Significance tests
The appropriate measure of correlation when variables are expressed as ranks
Spearman rho
All variables are entered into the solution simultaneously; overlapping variance (i.e., variance in the DV explained by more than one IV) is ignored when assessing the contributions of individual IVs to the variability of the DV
Standard analysis
An average distance of scores away from the mean; the square root of the variance
Standard deviation
The amount by which one can expect sample means to differ if other samples from the sassier population are used; indicates how well a sample represents the population from which it was selected
Standard error of the mean
A transformed score derived from the manipulation of a rash score that expresses how far away from the mean a given score is located, usually reported in standard deviation units
Standard score
Precursor to the variance-covariance matrix where the deviations has not yet been averaged; also called cross-product matrix
Sum-of-squares
Precursor to the variance-covariance matrix where the deviations has not yet been averaged; also called sum-of-squares
Cross-products matrix
A z-score expressed on a different scale; multiply the z-score by 10 and add 50
T-score
Error made when it is concluded that a null hypothesis is false, even though it is actually true; probability of this error is defined as α
Type I error
Error made when it is concluded that a null hypothesis is true, even though it is actually false; probability of this error is defined as β
Type II error
Having or characterized by one dependent variable
Univariate
A measure of how far a set of numbers is spread out from their average value; the average of the sum-of-squares
Variance
Used when scores are measured along a continuous scale; a square, symmetrical matrix where the elements of the main diagonal are the variances for each variable and the elements on the off-diagonals are the covariances between pairs of variables
Variance-covariance matrix
A standard score that indicates the distance away from the mean a score is in terms of standard deviation units; calculated by subtracting the mean from the raw score and then dividing the value by the standard deviation
z-score
