PLSC 374
Lab Exercise (2/21/2007): Aggregate Cross-National Data
(aka Part 1 of Assignment 3)
The main purposes of this exercise are to (1) compile a small data set from published cross-national data sources, (2)
formulate and test some hypotheses of interest to political scientists, and (3) perform suitable statistical analysis.
Note: This is part one of Assignment 3.
Please read the following instructions carefully. Make sure you understand them, before you begin on the exercise.
1. This assignment is due Monday, February 26, at the beginning of the class. Ensure your computer and printer
are in working conditions.
2. Review the policies enunciated in the syllabus, and especially the instructions regarding the exercises and my
expectations on written works. For those that seek to benefit from the help of the Writing Center, start your
work earlier.
This assignment consists of several sequential and integrated steps.
(1) Refer to the table “Classification of Economies by Region and Income, FY2006” on p. 291 of World
Development Report 2006 (The World Bank, 2006), which is given to you as a handout.
(2) Choose countries from each of the following groupings according to the specified number of cases (N=18),
while ensuring proper representation of various regions:
a. Low-income (LIC): $825 or less (6 cases)
b. Lower-middle income (LMC): $826~$3,255 (3 cases)
c. Upper-middle income (UMC): $3,256~$10,065 (3cases)
d. High income: $10,066 or more (6 cases)
(3) Use the remainder of the handout, or go to WDI data query page
(www.worldbank.org/data/onlinedbs/onlinedbases.htm), or other relevant data sets from
(http://oncampus.richmond.edu/~vwang/ps374/2007Spring/PLSC374_S2007_index.htm), or Library course
website (http://library.richmond.edu/information/csguides/PLSC37406.htm), Freedom House, Transparency
International, etc. (and feel free to look elsewhere) to enter the relevant data for each case that are pertinent to
your research questions (see (4)):
a. Choose 5 variables at interval or ratio level of measurement (called scale data) (e.g., population
(millions), Gross national income (billions of dollars), Gross national income - per capita dollars, PPP
gross national income - per capita dollars, Life expectancy at birth (years), Under-5 mortality rate per
thousand, Adult literacy rate, GINI index1, External debt (total millions of dollars and percentage of
GNI), Economic growth (gross domestic product) for 1980-90 and 1990-98, Personal computers (per
1000 people), Internet hosts (per 10,000 people), Trade dependence: A proxy figure is [(merchandise
exports)+(merchandise imports)+(exports of commercial services)+(imports of commercial services)] /
GNI, etc.)
b. Create a “democracy index” (7 = most free, 1= least free) which is the inverse of the average of the
civil liberty (CL) and the political rights (PR) scores measured by Freedom House.2
c. Create a “corruption index” (10 = most clean, 1 = least clean) from Transparency International’s
International Perceptions Index 2005
d. Choose 2 variables at nominal level and 2 variables at ordinal level of measurement (e.g., “free,”
“partly free,” and “not free.”).
(4) Use this data as appropriate to construct, by hand or spreadsheet, a table that contains all the information. Make
sure you include all the sources. DUE: Monday, February 21.
(5) Enter your variables in an SPSS file. Consult An SPSS Companion to Political Analysis. DUE: Monday,
February 21.
(6) Use this data as appropriate to construct, by hand, 1 crosstab, 1 means table, and one graph.
(7) The rest of Assignment 3 will be dealt with next week.
1 This is a widely used measure for income distribution, ranging between 0 (perfect equality) and 1 (perfect
inequality), with 0.4 usually seen as a threshold for alarm.
2 For example, the average of Algeria’s PR (6) and CL (5) is 5.5. So its “democracy score” will be 2.5.