Introduction to Statistical Analysis for Social Work | Final Exam Definitions
Chapter 7 – t Tests and Analysis of Variance
> Chi-Square Test of Association: a common statistical procedure used to
analyze the association between two nominal level variables. It is usually referred
to as chi-square.
Dependent T Test: a hypothesis-testing procedure used to decide whether the
means within two dependent samples could have occurred by sampling error;
sometimes referred to as a dependent-groups t test, correlated-groups t test,
paired-groups t test, or matched-groups t test.
Degrees of Freedom: a characteristic f the sample statistic that determines the
appropriate sampling distribution; the number of ways in which the data are free
to vary; the number of observations minus the number of restrictions places on
the data; a number related to the sample size in a way that depends on the
particular statistical technique employed; in many statistical test, degree of
freedom, or df, is needed in order to determine the p value.
Hottelling’s T Test: a statistical test used to examine the relationship among a
dichotomous independent variable and a set of two or more interval/ratio
Independent T Test: a statistical test used to decide whether two independent
samples could have occurred by sampling error.
Kolmogorov-Smirnov One-sample Test: (K-S test) a nonparametric test that
compares the overall distribution of a sample with another theoretical distribution
in relation to an ordinal or skewed interval or ratio dependent variable.
Kolmogorov-Smirnov Two-sample Test: (K-S test) a nonparametric test that
compares the overall distribution of two samples in relation to an ordinal level or
skewed interval or ratio level dependent variable. It identifies the point at which
the two samples reflect the greatest difference.
Kruskal-Wallis Test: a nonparametric statistical test that can be used to
examine the relationship between a nominal variable with more than two values
and an ordinal level or skewed interval level or ratio level variable. It is a common
nonparametric alternative to simple, one-way ANOVA.
Mann-Whitney U Test: a nonparametric test that is used to examine the
relationship between a dichotomous nominal level variable and an ordinal level or
skewed interval level or ratio level variable. It is a common alternative to the
parametric independent t test. MANOVA: (multiple analysis of variance)