Mvs - Descriptives For Multivariate Statistics

Front 1

bar charts

Back 1

When you have a categorical variable as the IV, it is best to use ___ to graphically display and review the data.

Front 2

scatterplots

Back 2

When the IV is continuous, then generally a ___ is used to visualize the data.

Front 3

Q-Q plot

Back 3

A ___ can help you assess univariate normality by displaying observed values and expected values in a normal distribution. If the line is straight, then the values can be considered normally distributed.

Front 4

Shapiro-Wilk

Back 4

The ___ test provides an inferential means to test for univariate normality in small datasets, as well as multivariate normality.

Front 5

Kolmorgorov-Smirnov

Back 5

The ___ provide an inferential means to test for normality in large datasets.

Front 6

Analyze->Descriptive Statistics->Explore

Back 6

In SPSS, use ___ to provide summaries of variables. Make sure to select normality plots from the options.

Front 7

probability graph

Back 7

A ___ can be used to graphically assess multivariate normality.

Front 8

Box's M

Back 8

___ tests for equality of variance-covariance matrices, which is a way to check for equal variance across groups for multivariate outcomes.

Front 9

Tamhane's T2 or Dunnett's T3

Back 9

___ are corrections for the violation of the assumption of homogeneity of variance for multivariate outcomes, and are nearly always used.

Front 10

mean vector

Back 10

In multivariate statistics, we describe the means of all of the DVs as a ___.

Front 11

more than 5%

Back 11

If ___ of the data is missing, it is problematic.

Front 12

list-wise deletion

Back 12

___ is the most conservative method of compensating for missing data.

Front 13

mean replacement

Back 13

___ is substituting the average value for similar subjects for the missing value.

Front 14

regression-based replacement

Back 14

___ involves using a model to estimate the expected value of a missing value.

Front 15

multiple imputation

Back 15

___ uses multiple sub-datasets that estimate the missing value, similar to bootstrapping.

Front 16

eigenvector and eigenvalues

Back 16

___ provide the starting point for almost all of the multivariate statistics that we will utilize.

Front 17

eigenvector

Back 17

___ is a new variable that represents the unique variance for a given set of values.

Front 18

multicollinearity

Back 18

When two variables in a correlation matrix have a correlation magnitude greater than 0.7, we consider ___ an issue.

Front 19

correlation matrix

Back 19

Evaluation of the ___ helps identify the issue of multicollinearity.