 UP504 • Final Review and Exam

modified: Sunday, April 13, 2008

Dates: April 9 (review); April 14 (exam)

--> DOWNLOAD April 9 in class practice questions and answer sheet. Overview

The final exam is a closed note, closed book exam. The emphasis of the exam is on understanding concepts, interpretation of statistics, and what methods address what kinds of questions. Many of the questions will require not just a general conceptual knowledge of methods, but also a more precise knowledge of terminology, calculations and meaning of statistical output (e.g., regression results or t-test/difference of means). The topics covered are all the materials from the readings and lectures we have covered in the class.

All you need to bring is a pencil or two (with eraser). You may bring a calculator, though there will be few (if any) problems requiring any calculations, and I will make the numbers simple so that you can do any calculations by hand. (A simple calculator will do; no need for a scientific or financial calculator). You will write the answers directly on the exam (no need to bring in blue books or paper). I will pass out the exam at 3:10 p.m. and collect at 4:30 p.m.

Note: Students with exam scheduling conflicts due to religious observances or related issues should contact the instructor far in advance to discuss alternative arrangements.

Topics include (but are not limited to):

 Overall Research Terms unit of analysis scientific method inductive vs. deductive  null hypothesis and research hypothesis falsification replication validity (both internal and external) reliability generalization (and the distinction between statistical and analytical generalization) U.S. Census and Census Geography the "old" metropolitan geography: MSA, PMSA, CMSA the "new" metropolitan geography: metropolitan and micropolitan areas Census tracts vs. blocks race versus ethnicity variables Census regions and divisions Data Presentation: principles of data presentation: basic do's and don'ts of graphs and tables advantages and disadvantages of different types of graphs (bar, column, scatterplot, pie, etc.) ways that graphs distort data Multiple Regression Analysis know how to interpret a multiple regression output (e.g., t score, F score, R-squared, adjusted R squared, coefficient, constant, beta, significance, standard error) the mathematical relationship between Total Sum of Squares, Regression Sum of Squares, F, R-Square, etc. how to write an equation from the coefficients and use this to make an estimate know the basic assumptions of a regression model (see Lewis-Beck, p. 26 - ) homoscedasticity vs. heteroscedasticity multicollinearity dummy variables outliers Evaluation Research program theory counterfactual (and compare approaches and pros/cons of each, e.g., before-after, random assignment, natural experiments; matched-pairs; regression) random assignment control vs. experimental group rival explanation side-effects (unintended consequences) placebo effect how regression analysis might be used in evaluation research Case Study Research statistical generalization vs. analytical generalization single- vs. multiple case studies literal replication vs. theoretical replication typical vs. exceptional cases cases as quasi-experiments Economic Analysis economic base model (multipliers) location quotients GINI coefficients (and the Lorenz curve) cost-benefit analysis shift-share analysis Basic Statistics terms (largely covered in UP503): descriptive vs. inferential statistics (can you explain the difference?) univariate, bivariate, multivariate levels of measurement:  nominal, ordinal and interval mean, median, mode correlation coefficient degrees of freedom confidence interval normal curve statistical significance distribution of sample means t score F score difference of means test difference of proportions test ANOVA - Analysis of Variance chi-square the difference between a one-tail and two-tailed test (and when to use one or the other) -see Trochim's useful guide longitudinal vs. cross-sectional data panel vs. non-panel data standard error versus standard deviation primary vs. secondary data Relationships between variables: ecological fallacy spurious relationships intervening variables correlation vs. causation scatterplots (x-y plots) linear vs. nonlinear relationships measurement of the strength of a relationship (r) versus the statistical significance of a relationship (t-score and corresponding prob-value) symmetrical vs. non symmetrical relationships dependent vs. independent variables Survey Research: sampling terminology:  population, sampling frame, sampling element, sample, sample size, response rate, nonresponses probabilistic and nonprobabilistic sampling types of sampling:  simple random, systematic random, stratified, clustered, etc.  -- how they are done;  assumptions of each;  advantages and disadvantages of each.  use of weights to adjust for non-proportional sampling (e.g., with stratified sampling) sampling fraction measures and concepts questionnaire design:  do's and don'ts of wording, ordering, filtering, open- vs. closed-ended questions, etc. Demography fertility, mortality, migration (components of change) cohort survival analysis age-specific data (e.g., age-specific fertility) fertility rates "replacement level" (for fertility rates) life expectancy (e0) components of life tables (qx , lx , dx , Lx , Tx , ex ) and basic equations, e.g., qx = dx/ lx; ex = Tx/ lx population forecasting age pyramids net vs. gross migration types of growth functions (e.g., linear, compounded, exponential)