DEPARTMENT OF POLITICAL SCIENCE

AND

INTERNATIONAL RELATIONS



Overview of Course and Information Network

  1. CLASS 1: AGENDA
    1. Why applied statistics?
      1. Some policy and social science research issues
      2. Drugs, alcohol and violence.
    2. Course Procedures
      1. Activities
      2. The "information network"


  2. SUBSTANTIVE PROBLEMS:
    1. The effects of the welfare state:
      1. Recent changes in American welfare law reflect a widely held belief that the social assistance programs have done more harm than good. This argument in turn partially rests on "statistical analysis." As both citizens and policy analysts we ought to know how sound the evidence is.
    2. Managed health care
      1. Most proposals for health care reform involve the use of some type of health maintenance organizations (HMOs) to control costs. Some people, however, belief that the care provided by these organizations may be inferior to that available in traditional fee-for-service plans. Again, the cases depends partly on "empirical data."
    3. Is marijuana a "gateway" drug?
      1. Two states, Arizona and California, passed referenda permitting the limited use of marijuana for medical purposes. A major objection to this "soften" of laws against drug abuse flows from the belief that cannabis use leads to dependence on other more dangerous substances. Once more the argument rests on so-called hard data? But just how solid is this evidence?
    4. Reforming education
      1. No one seems satisfied with the state of American education, but there is little agreement on what will improve it? Many in congress advocate vouchers and school choice while President Clinton insists on national testing standards. Most liberals think the educational system needs more money. Who is right?
    5. Politics and public ideology
      1. Most of president Clinton's policy agenda stems from the conviction that the American people have become increasingly conservative in the last 25 years and hence no longer support many government programs and regulations. The President and his advisors invoke polls to document their beliefs. But do these supposedly hard numbers really make the case that citizens have lost faith in an activist government?
    6. Missing voters:
      1. Despite advances in education, communications, and transportation--not to mention voting rights laws--turnout in American elections has declined more or less steadily during the last half of the twentieth century. Why?


  3. DRUGS, ALCOHOL, AND HOMICIDE:
    1. The of a recent article in the Journal of the American Medical Association write: "This study supports the need to address alcohol and other drug abuse and its relationship to risk of violent death."(1) Here is part of their evidence:


Substance Abuse and Risk of Homicide By Age and Sex of Matched Cases and Controls






Substance




N


Odds*

Ratio

95%

Confidence Intervals

Alcohol
Age

< 50

244

2.0 1.3-3.0

50

126

2.6 1.5-4.7
Sex

Male

234

3.0 1.5-3.4

Female

136

1.4 .8-2.3
Drugs
Age

< 50

244

6.2 3.1-12.6

50

126

1.0 .1-7.1
Sex

Male

234

5.8 2.8-12.4

Female

136

3.0 .8-11.1

* Adjusted for drinking or other drug use

    1. Although the layout and terms in the table may be confusing, the content lies well within the grasp of anyone who has a modest background in statistics.
      1. Would you have trouble interpreting the numbers?


  1. COURSE PHILOSOPHY:
    1. This class rests on several propositions:
      1. Even average citizens can comprehend and evaluate the statistical claims and counter claims that permeate these sorts of policy debates.
      2. It is certainly imperative that policy scholars and makers know how to weigh the evidence.
      3. In order to do so one must understand a relatively small and straight forward set of statistical principles and methods.
      4. The best way of achieving this understanding is to examine "real" data and use statistical methods to analyze them.
    2. What students need
      1. An open mind and good attitude
      2. Package 1 - minimal but adequate:
        1. Class notes the text of MINITAB for Windows, and access to University personal computer.
      3. Package 2 - better yet:
        1. Class notes and access to a personal computer.
        2. Student Version of MINITAB for Windows text and disks.
      4. Package 3 - the ideal:
        1. Class notes and access to a personal computer.
        2. Student Version of MINITAB for Windows text and disks.
        3. Agresti and Finlay, Statistical Methods for Social Sciences, 3rd edition.
        4. Sage University Papers
      5. The last option is the by far the best but of course is also the most expensive. On the other hand, students at working for an advanced degree in social and policy sciences may want to start building a professional library. Nevertheless, one of the first two alternatives should be satisfactory.


  2. APPLIED STATISTICS - AN OVERVIEW OF ACTIVITIES:
    1. Data description and reduction: summarize with numbers and graphs the main features in a data set. Features include variation, commonalities, differences, trends, patterns of association, and the like.
    2. Modeling: develop simple equations that show interrelationships among variables.
    3. Inference: making "educated" guesses about "populations" based on relatively small samples. What, for instance, can one say about American public opinion on the bases of 500 interviews?


  3. THE INFORMATION NETWORK:
    1. As ever one knows the world depends increasingly on the electronic transmission of information. This course is no exception. Quite apart from any statistical calculations you will be performing with a computer, you need to be familiar with electronic mail (e-mail) and the Internet
    2. The information network or "virtual reality" as the public calls it, consists of sets of "data"--documents, batches of numbers, images, computer programs and so forth--located in various places throughout the world. For our purposes it consists mostly of a class internet site or web page plus electronic mail addresses.
    3. Figure 1 presents a simple over view.


      1. One simply uses a "terminal" (that is, a simple video and keyboard) or personal computer to send e-mail and visit the class web site (the place where notes, data, announcements, and so forth are kept).
    1. Of course you have many choices or ways to "connect" to this information network.
      1. A University of Delaware "X-terminal."
        1. These devices, located in several places throughout the campus, consist of only a monitor and keyboard. They are not computers and cannot by themselves store information or run programs. If you use an X-terminal, any programs (e.g., e-mail) that you start and any data or information that you collect will be stored on one of the University's "mainframe" or large computers (e.g., Copland, brahms, Strauss). The terminal itself is not connected to any storage device. As students become familiar with personal computers, they rely less and less on these simple terminals.
      2. A University of Delaware desktop computer.
        1. Nearly all of the university's computers that are located in public sites are linked to the information network by high speed lines.
        2. With a PC one has several choices for computing and accessing the information network. The PC can be used as if it were a terminal--that is a monitor and keyboard--that is linked to the mainframes and from there to the network. Or, it can run its own software to connect to the network. The latter option allows one to store data and text directly on the PC's disk drive.
        3. Most UD personal computers use Windows 95.
      3. Your own computer that is linked via a modem to the University's network or to an Internet Service Provider's (ISP) network which in turn connects to the University.
        1. Your computer can act as "terminal" (see above) or run its own software that allow you to browse the internet and send and receive e-mail.
    2. Terminals versus personal computers.
      1. Nearly everyone will use a desktop computer, even if belongs to the University or someone else. There are, however, so-called terminals (often labeled X-terminals) scatter around campus.
      2. Moreover, for some purposes a personal computer may be used as a terminal.
        1. For example, many people use the University's mainframe electronic mailing system, Pine, to receive and send messages. Pine "runs" on a central computer, Copland, and can be accessed from a PC that is connected via phone lines or cables to the University's network.
        2. When doing so, however, the software--usually it will be telnet--turns the computer into a simple (vanilla) terminal
        3. Consequently, I use the term telnet or telnet session to mean the use of a PC as a simple terminal.
      3. Hence, it's important to know what each device does.
        1. Look at the attached figures.
      4. When used as a terminal a PC loses many of the advantages of Windows. But one can run a windows program such as MINITAB or WORD at the same time one has a telnet session open. And it's possible to copy information from the telnet window and paste it in the windows application.


  1. NEXT TIME:
    1. Windows and the basics of MINITAB and SPSS


1. Source: Frederick P. Rivara et al., "Alcohol and Illicit Drug Abuse and the Risk of Violent Death in the Home," Journal of the American Medical Association, August 20, 1997, page 572. Go back

Go to Statistics main page

Go to H. T. Reynolds page.

Copyright © 1997 H. T. Reynolds
This Home Page was created by WebEdit,Wednesday September 3, 1997