POLS 626
Exam 3 Review Sheet
Last updated: 4/27/05
This review is intended to
give you a general idea of what may appear on the exam. The exam will cover material from the
assigned readings and material that was presented in the lecture.
Exam period:
May 2
(Mon) + 2
May 3
(Tue) + 2
May 4 (Wed) to May 5 (Thu) + 0
May 6 (Fri) -10
You will not be able to take the exam after May 6.
You will not be able to ask me questions once the exam
period starts.
Please sign up to take the exam by visiting the
following web page: www.bsu.edu/webapps/tlrs
(Sign up for 1hr). Please bring your ID when you take the exam.
REVIEW
II. Causality (
III.
Levels of measurement
Nominal, Ordinal, Interval/Ratio
IV.
Descriptive Statistics
Measures of central
tendency (Mean, median,
and mode)
Measures of
variability (Range,
variance, and standard deviation)
V.
ANOVA
Interpretation (No calculation)
VI.
Measures of Association
A. Crosstabs
Know how to produce a cross-tabulation
Know how to read a cross-tabulation (both bivariate and multivariate)
Know how to control for a third variable (Nominal and
ordinal variables only)
B. Tests of statistical significance and measures of
association
Know how to write a hypothesis
Know how to test a hypothesis
Statistical significance
Strength of association
Direction of association
Substantive interpretation
Bivariate association
Cramer’s V, Tau b, Tau c, Pearson’s r
Controlling for a third variable (Nominal and ordinal
variables only)
Cramer’s V, Tau b, Tau c
VII.
Two-variable regression analysis
Intercept and slope coefficient (estimation and
interpretation)
Population/sample regression function (Diagram for
PRF and SRF)
The difference between ui
and ui hat and yi
and yi hat
Goodness of Fit
Assumptions
Calculating ui
hat and yi hat
VIII.
Multiple regression analysis
Intercept and slope coefficients (interpretation)
Assumptions
Testing Hypotheses
Against One-sided and
Two-sided Alternatives, P-values and t Tests
Example 4.3
Economic, or Practical, versus Statistical
Significance
Confidence Intervals
Testing Multiple Linear Restrictions: The F Test
The R-Squared Form of the F
Statistic
The F Statistic for Overall Significance of a
Regression
IX Multiple Regression Analysis: Further Issues (Ch6)
Standardized
coefficients
Logarithmic
functional forms
Adjusted
R-squared
X. Dummy
Variables (Ch7)
Table
7.1
A
single dummy independent variable
Equation 7.1
Figure 7.1
Example 7.1
Using
dummy variables for multiple categories
Allowing
for different slopes
Equation 7.16
Equation 7.17
Figure 7.2
XI. Heteroskedasiticity (Ch8)
Testing
for heteroskedasticity
Equation 8.14
The White Test
Equation
8.19
Equation
8.20
Weighted
Least Squares Estimation
Pp270-274
FGLS pp276-280
XII. Specification and Data Problems (Ch 9)
RESET
as a General Test for Functional Form misspecification
Pp292-293
Using
Lagged Dependent Variables as Proxy Variables
Pp300-302
Example 9.4
Missing
Data
Pp309-310
Outliers
and influential observations
Standardized residual
XIII. Autocorrelation
Ch
10 Serial correlation pp333-334
Ch
11 The Durbin-Watson Test pp397-399
Ch
12 Differencing and Serial Correlation pp409-410
XIV. Limited
dependent variable models (Ch17)
Pp553-558
Figure
17.1
Interpreting
the logit model pp559-565
Pseudo
R squared
Calculating
predicted probabilities
XV. The following formulae will be provided. The tables for t, F, and DW will also be provided
SIMPLE REGRESSION ANALYSIS
Estimation of SRF
Hypothesis
Testing
GOODNESS OF FIT
LOGARITHMIC FUNCTION
MULTIPLE REGRESSION
Hypothesis Testing
Confidence
Intervals
Testing
Multiple Linear Restrictions
The F Statistics for Overall Significance of a
Regression
A
single dummy independent variable
(7.1)
Interactions
involving dummy variables
(7.16)
Test
for Hetroskedasticity
(8.20)
Testing
Multiple Linear Restrictions
RESET
Restricted: 
Unrestricted: 
Logit
Logarithmic
function
XVI. The following instructions
will be provided
Durbin-Watson
Test
The null hypothesis: No autocorrelcation
d
< dL Reject
the null hypothesis
d > dU Fail
to reject the null hypothesis
dL <d < dU Inconclusive
|
Measurement Types |
Statistics |
|
Two nominal variables |
Lambda/ Cramer’s V |
|
One nominal one ordinal
variables |
Cramer’s V |
|
Two ordinal variables |
|
|
Dependent=I/R, Independent= nominal or ordinal |
ANOVA: Eta
squared |
|
Two I/R variables |
Pearson’s corr. Coeff or Regression |
Note: You will not be allowed
to bring any notes to the lab. Please ask a lab assistant for the formulae
sheet, the statistical tables, and scrap paper. Do not start the exam without
the formula sheet, the statistical tables, and scrap paper. You will be
allowed to use a calculator and pens/pencils. Any other electronic device
including cell phones are not allowed.