SCS 2150 Lecture Notes - Lecture 17: Null Hypothesis, Multivariate Analysis, Scatter Plot
20 Aug 2016
School
Department
Course
Professor
SCS2150
March 23 2016
Variables and Quantitive Analysis
continued from last class
-bivariate analysis determines whether there is a relationship between two variables
-determination of a relationship is not proof of causation
-contingency tables/cross-tabulations allow simultaneous analysis of two variables
-identities patterns of association
-can be used for any variable type
-normally used for nominal or ordinal data
-correlation analysis: Pearson’s r(elationship)
-normally used with interval/ratio data
-values from 0 (no relationship), to +1 (perfect positive relationship), or -1 (perfect
negative relationship)
-the relationship between the variables should be approximately linear if Pearson’s r is to be
used
-can be established using a scatter plot
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
Variables and quantitive analysis continued from last class. Bivariate analysis determines whether there is a relationship between two variables. Determination of a relationship is not proof of causation. Contingency tables/cross-tabulations allow simultaneous analysis of two variables. Can be used for any variable type. Normally used for nominal or ordinal data. Values from 0 (no relationship), to +1 (perfect positive relationship), or -1 (perfect negative relationship) The relationship between the variables should be approximately linear if pearson"s r is to be used. Can be established using a scatter plot. Statistical signi cance is stated as a probability level (probability that the results are not due to chance) Statistical evident is often misused and its signi cance does not always imply real life relatability. Easier to get statistically signi cant results in larger n studies. A null hypothesis tests the signi cance of the bivariate association. Establish an acceptable level of signi cance (must be . 05 or lower)
