Exam 1: Introduction to Multivariate Statistics

The use of multivariate statistical techniques has become more commonplace largely due to the increasingly complex nature of research designs and related research questions.

The sequential analysis requires the researcher to prioritize the entry of IVs into the equation or solution.

The mathematical calculations involved in multivariate statistical analyses are performed only on a correlation matrix.

Quantitative variables are also referred to as continuous or interval variables.

Having a data set with orthogonal variables is not the ideal situation.

A study appropriate for multivariate statistical analysis is typically defined as one with several dependent variables (DVs).

A dichotomous variable is one that has only two possible levels or categories.

Inferential statistics deal with collecting and analyzing information from samples in order to:

Categorical variables consist of separate, indivisible categories.

Age is a quantitative variable, but one could recode the values so that it would be transformed into a dichotomous variable.

Categorical variables may also be referred to as nominal, ordinal, discrete, or qualitative.

Using a standard analysis approach, the overlapping portion of variance is included in the overall summary statistics of the relationship of the set of IVS to the DV, but that portion is not assigned to either of the IVs as part of their individual contribution.

When conducting a multivariate analysis, the best recommendation is to obtain the solution with the largest number of variables.

In an experimental research study, if the researcher finds a statistically significant difference between two or more of the groups representing different treatment conditions, she or he can have some confidence in attributing causality to the IV.

Orthogonality is not a desirable quality for multivariate statistical analyses.

Two most commonly used measures of relationship are:

The first step in nearly any data analysis situation is to describe or summarize the data collected on a set of participants that constitute the sample of interest.

Nonexperimental research studies also enable a researcher to conclude that the IV and DV are related and infer causality.

Measures of central tendency include:

One of the major difficulties in using multivariate statistical analyses is that it is sometimes nearly impossible to get a firm statistical answer to your research questions.