Bivariate Relationships Analyzing two variables at a time, usually the Independent & Dependent Variables Like one variable at a time, this can be done.

Slides:



Advertisements
Similar presentations
Bivariate Analysis Cross-tabulation and chi-square.
Advertisements

Chapter 13: The Chi-Square Test
ANOVA: Analysis of Variance
Sociology 601 Class 13: October 13, 2009 Measures of association for tables (8.4) –Difference of proportions –Ratios of proportions –the odds ratio Measures.
Chapter 11 Contingency Table Analysis. Nonparametric Systems Another method of examining the relationship between independent (X) and dependant (Y) variables.
CORRELATION. Overview of Correlation u What is a Correlation? u Correlation Coefficients u Coefficient of Determination u Test for Significance u Correlation.
Session 7.1 Bivariate Data Analysis
PPA 415 – Research Methods in Public Administration Lecture 9 – Bivariate Association.
Matching level of measurement to statistical procedures
Correlations and T-tests
Social Research Methods
Data Analysis Statistics. Levels of Measurement Nominal – Categorical; no implied rankings among the categories. Also includes written observations and.
PPA 501 – Analytical Methods in Administration Lecture 9 – Bivariate Association.
Summary of Quantitative Analysis Neuman and Robson Ch. 11
Crosstabs. When to Use Crosstabs as a Bivariate Data Analysis Technique For examining the relationship of two CATEGORIC variables  For example, do men.
Correlation Question 1 This question asks you to use the Pearson correlation coefficient to measure the association between [educ4] and [empstat]. However,
Chapter 14 in 1e Ch. 12 in 2/3 Can. Ed. Association Between Variables Measured at the Ordinal Level Using the Statistic Gamma and Conducting a Z-test for.
Correlations 11/5/2013. BSS Career Fair Wednesday 11/6/2013- Mabee A & B 12:30-2:30P.
Week 11 Chapter 12 – Association between variables measured at the nominal level.
Leedy and Ormrod Ch. 11 Gray Ch. 14
Analyzing Data: Bivariate Relationships Chapter 7.
Mean Tests & X 2 Parametric vs Nonparametric Errors Selection of a Statistical Test SW242.
Week 12 Chapter 13 – Association between variables measured at the ordinal level & Chapter 14: Association Between Variables Measured at the Interval-Ratio.
Association between Variables Measured at the Nominal Level.
This Week: Testing relationships between two metric variables: Correlation Testing relationships between two nominal variables: Chi-Squared.
LIS 570 Summarising and presenting data - Univariate analysis continued Bivariate analysis.
1 Measuring Association The contents in this chapter are from Chapter 19 of the textbook. The crimjust.sav data will be used. cjsrate: RATE JOB DONE: CJ.
In the Lab: Working With Crosstab Tables Lab: Association and the Chi-square Test Chapters 7, 8 and 9 1.
Tests of Significance June 11, 2008 Ivan Katchanovski, Ph.D. POL 242Y-Y.
Correlation and Linear Regression. Evaluating Relations Between Interval Level Variables Up to now you have learned to evaluate differences between the.
Statistics in Applied Science and Technology Chapter 13, Correlation and Regression Part I, Correlation (Measure of Association)
Chi-square (χ 2 ) Fenster Chi-Square Chi-Square χ 2 Chi-Square χ 2 Tests of Statistical Significance for Nominal Level Data (Note: can also be used for.
Bivariate Descriptive Analysis First step in analyzing your data Three components Cross-tabulations and frequency distributions Significance testing Correlations.
1 Further Maths Chapter 4 Displaying and describing relationships between two variables.
Cross-Tabs With Nominal Variables 10/24/2013. Readings Chapter 7 Tests of Significance and Measures of Association (Pollock) (pp ) Chapter 5 Making.
URBP 204A QUANTITATIVE METHODS I Statistical Analysis Lecture IV Gregory Newmark San Jose State University (This lecture is based on Chapters 5,12,13,
Review of the Basic Logic of NHST Significance tests are used to accept or reject the null hypothesis. This is done by studying the sampling distribution.
Educational Research Chapter 13 Inferential Statistics Gay, Mills, and Airasian 10 th Edition.
Correlation & Regression Chapter 15. Correlation It is a statistical technique that is used to measure and describe a relationship between two variables.
Chapter 16 Data Analysis: Testing for Associations.
Chi-square Test of Independence
Chapter 11, 12, 13, 14 and 16 Association at Nominal and Ordinal Level The Procedure in Steps.
Practice Problem: Lambda (1)
12/23/2015Slide 1 The chi-square test of independence is one of the most frequently used hypothesis tests in the social sciences because it can be used.
PSC 47410: Data Analysis Workshop  What’s the purpose of this exercise?  The workshop’s research questions:  Who supports war in America?  How consistent.
Inferential Statistics. Explore relationships between variables Test hypotheses –Research hypothesis: a statement of the relationship between variables.
Chi Square & Correlation
Copyright © 2014 by Nelson Education Limited Chapter 11 Introduction to Bivariate Association and Measures of Association for Variables Measured.
PART 2 SPSS (the Statistical Package for the Social Sciences)
Chapter 14 – 1 Chi-Square Chi-Square as a Statistical Test Statistical Independence Hypothesis Testing with Chi-Square The Assumptions Stating the Research.
Statistics Statistics Data measurement, probability and statistical tests.
Review: Stages in Research Process Formulate Problem Determine Research Design Determine Data Collection Method Design Data Collection Forms Design Sample.
Cross Tabs and Chi-Squared Testing for a Relationship Between Nominal/Ordinal Variables.
Jump to first page Inferring Sample Findings to the Population and Testing for Differences.
Determining and Interpreting Associations between Variables Cross-Tabs Chi-Square Correlation.
Copyright © 2012 by Nelson Education Limited. Chapter 12 Association Between Variables Measured at the Ordinal Level 12-1.
Bivariate Association. Introduction This chapter is about measures of association This chapter is about measures of association These are designed to.
Other tests of significance. Independent variables: continuous Dependent variable: continuous Correlation: Relationship between variables Regression:
Association Between Variables Measured at the Ordinal Level
Data measurement, probability and Spearman’s Rho
Final Project Reminder
Final Project Reminder
Bi-variate #1 Cross-Tabulation
Chapter 14 in 1e Ch. 12 in 2/3 Can. Ed.
Learning Aims By the end of this session you are going to totally ‘get’ levels of significance and why we do statistical tests!
Social Research Methods
Summarising and presenting data - Bivariate analysis
The Chi-Square Distribution and Test for Independence
Data measurement, probability and statistical tests
Presentation transcript:

Bivariate Relationships Analyzing two variables at a time, usually the Independent & Dependent Variables Like one variable at a time, this can be done visually with charts and graphs (such as a scatterplot), and with frequency tables. To see two univariate frequency tables together at the same time, you cross-table them, that is, you create a cross-tabulation (or shorthand: Crosstab). Guidelines for creating crosstabs: (1) Put the Dependent variable in ROWS (2) Put the Independent variable in COLUMNS (3) Calculate percentages in the direction of the independent variable (Columns in this case). You are comparing the distributions of each category (value) of the independent variable with one another in terms of the categories of the dependent variable. For example, if you want to see if there is a relationship between gender and religion, you compare the values of gender (that is, male and female) across the various religions. When the number of men and the number of women are not exactly the same, you must standardize to compare by presenting the results in terms of percentages. The percentages of men who are Catholic, Jewish, etc. with the percentages of women who are Catholic, Jewish, etc. To compare, the percentages of men must add up to 100% as does the percentages of women.

Hypothesis: There is no relationship between Sex and College Status (Graduated or Left the College) Which is the Independent and Dependent Variables? What are the levels of measurement? Put into words: 74.9% of __________ have _____________. This is not the same as saying 74.9% of those who graduated are Female. If 71.8% of the entire four year period graduated, then compare the percentages of women with men relative to that 71.8%. Who tends to graduate disproportionately higher or lower than the overall rate?

This table, however, says: 61.7% of ___ are _______. To say that 80% of sociology majors are women is not the same as saying that 80% of women are sociology majors. You must always compare the categories (or values) of the independent variable by calculating percentages within each category separately. Each must add up to 100%. And if 59.5% of all respondents are Female and 40.5% are Males, then who graduates disproportionately higher or lower than their distribution in the sample?

Put into words what this table is telling us.

But how do we know if the differences between categories is big enough? What if we find that 75% of men own Toyotas and 79% of women own them? Is 4% a large enough difference or is that just sampling error? To decide if a difference is significant enough to hold a press conference, we must use some statistical tests which will tell us what the odds are – the probability – that these findings occurred by chance alone, that is, by accident and not a real finding. If the odds are small, we have a significant finding, because the probability of the finding happening by accident is so small that it must be due to a real impact of the independent variable on the dependent variable – not an accidental impact. For tell this you have to look for two things: (1)The Value of the Statistic (2) The Probability of that statistic occurring by chance If the probability of a statistic occurring by chance is less than 5% (p <.05), then you reject the null (or accept the positive) and declare that there is a relationship between the independent and dependent variables.

Chi-Square: a measure of association between the independent and dependent variables (usually nominal or ordinal measures). If the probability of obtaining a particular Chi-Square value by chance alone is less than.05, then we declare we have supported our hypothesis (or rejected our null). We hold a press conference and declare that indeed there is a relationship between the independent and dependent variables. Then we state in words what the relationship is (such as, women are more likely than men to vote Independent). For the following data, (a) state the null hypothesis being tested (b) What are the independent & dependent variables? (c) What levels of measurement are they?

The value of Chi-square does not tell you much in and of itself. You must depend on the probability level to tell you if it is significant and then all it tells you is that there is an association between the variables. However, there are statistics that can tell you how strong a relationship is between your variables, not just whether there is one or not. These are called correlations. They tell you how much of variability of the dependent variable is explained by knowing the variability of the independent variable. Nominal variables: Lambda Ordinal variables: Gamma, Spearman’s rho Interval/Ratio variables: Pearson r All correlations have two components: (1)The value which ranges from 0 to 1.0, where 1.0 is a perfect strong correlation and 0 is no correlation at all. (2) For those variables that have a direction (an order: ordinal, ratio measures), a plus or minus sign to indicate a positive or inverse relationship

A Lambda correlation of.75 between race and religion tells us that this is a strong relation (it’s close to 1.0) and therefore the variation in religion among our sample can be explained by the variation in race. You would then look to see which religions depend on which races and report that information (such as Whites tend to be Protestant, Latinos are Catholic, and so on). A guideline: Correlations between 0 and.30 tend to be weak Correlations between.30 and.70 tend to be moderate Correlations between.70 and 1.0 tend to be strong A Pearson r correlation of -.60 is just as strong as one that is.60, and stronger than a correlation of.50, for example. The minus sign just tells us that it is inverse: those who score low on one variable, score high on the other. It does not mean it is weak or less than any positive correlation.

Visual Version of Correlation: Scatterplots Pearson r =.84

Certain correlations also tell us the proportion reduction in error or PRE. This means that the proportion (or percentage) of errors that are made in predicting the values of a dependent variables is reduced by knowing the values of an independent variable. For example: A Lambda of.45 between race and religion would indicate that 45% of the errors in explaining the variability of religion among the respondents in our sample are reduced by knowing the variability of races in the sample. For Lambda and Gamma, PRE is simply the correlation coefficient. (Multiply by 100 to get a percent instead of a proportion). For Pearson r and Spearman’s rho, you must square the correlation value to determine the proportion of error reduction (r 2 or rho 2 ). So a Pearson r correlation of -.50 between high school GPA and SAT scores would suggest that.25 or 25% of the errors in predicting SAT scores would be reduced once we know the respondents’ high school GPAs.

Put these findings into words

Review (1) Determine the independent and dependent variables in the hypothesis. (2) Label the levels of measurement for each variable. (3) Decide the appropriate statistics to use. (4) Evaluate the value of the statistic and the probability (or significance) level. (5) If the p-value is less than.05, then reject the null and accept the positive hypothesis. (6) If the statistic is a correlation (lambda, gamma, Pearson r, Spearman rho), then determine the PRE. (7) Put the findings into words for (a) fellow statistics experts and (b) for the general public on your Facebook page or Twitter feed!

Example There is no relationship between High School GPA and SAT scores. There is a relationship between High School GPA and College GPA. (1) What are the independent & dependent variables? (2) Levels of measurement? (3) Which statistic do you use? (4) What do the values of the Pearson r mean? (5) What are the Significance levels? (6) What are the PRE interpretations? (7) Put into words for a statistical audience and for the general public.