Chapter 14 Flashcards
What are associative analyses?
Determine where stable relationships exist between 2 variables
Define a relationship between two variables?
Relationship: consistent systematic linkage between levels or labels for 2 variables
- Levels refers to characteristics of description for interval or ratio scales
- Labels refers to characteristics of description for nominal or ordinal scales
What is a linear relationship and curvilinear relationship?
Linear: Two variables have “straight-line” relationship
y= a + bx
y is the dependent variable and x is the independent variable
Curvilinear: Some smooth pattern describes the relationship
What is a monotonic relationship?
The general direction of a relationship between two variables is known
1. Increasing relationship
2. Decreasing relationship
What is a nonmonotonic relationship?
Two variables are associated, but only in general sense. Presence or absence of one variable is associated with presence or absence of another
What are the 3 characterizing relationships between variables?
Presence: whether any systematic (statistical)
relationship exists between two variables
Pattern: the general nature of the relationship, which
may take the form of a direction
Strength of association: whether the relationship is
consistent
Step by step process for analyzing relationship between two variables?
- Choose variables to analyze
- Determine scaling assumptions of the chosen variables
- Use correct relationship analysis
- Determine if relationship is present
- If present, determine direction of the relationship
- If present, assess strength of relationship
What is the correlation coefficient and covariation?
Correlation coefficient: Index number constrained to fall between range of -1.0 and +1.0
- Communicates both strength and direction of linear relationship between two metric variables
Covariation: amount of change in one variable systematically associated with change in another variable
Correlation strength based on Coefficient size
+0.81 to +1.00, -0.81 to -1.00 Very strong
+0.61 to +0.80, -0.61 to -0.8
Strong
+0.41 to +0.60, -0.40 to -0.60
Moderate
+0.21 to +0.40, -0.21 to -0.40
Weak
+0.20 to -0.20
Very Weak
What is cross-tabulation and cross-tabulation cell?
Cross-tabulation: rows and columns defined by the
categories classifying each variable
Cross-tabulation cell: the intersection of a row and a
column
What are the 4 types of numbers in a cell of a cross-tabulation table?
- Frequencies - raw numbers in cell
- Raw percentages - cell frequencies divided by grand total
- Row percentages - row cell frequencies divided by row total
- Column percentages - column cell frequencies divided by its column total
What is a Chi-Square Analysis?
Chi-square analysis is the examination of frequencies for 2 nominal-scaled variables in a cross-tabulation table to determine if variables have significant relationship.
Null hypothesis: 2 variables not related
Chi-square statistic summarizes “how far away” from expected frequencies the observed cell frequencies are found to be.
What are the formulas for row percent and column cell percent?
Row cell percent = cell frequency / total of cell frequencies in that row
Column cell percent = cell frequency / total of cell frequencies in that column
Describe the chi-square distribution
The Chi-square distribution is skewed to the right, and the rejection region is always at the right-hand tail of the distribution.
The shape of the distribution is dependent on degrees of freedom.
