Research and Statistical Support On-line

        By Karl Ho, Statistical Consultant (kho@unt.edu)

        Research and Statistical Support, a division of AcademicComputing Services, has been going through some major changes since this past summer. Panu Sittiwong, the long-time manager of this area, returned to his native Thailand to live and work. Phanit Laosirirat, a long-time consultant in the area, completed his Ph.D. and also returned to his native Thailand. James Yarbrough, another long-time consultant, becamethe manager and two part-time employees were hired, Ajay Babar and Voyko Kavcic. Voyko and Ajay have now moved on to other endeavors and we are happy to welcome Karl Ho, a Ph.D. candidate in the Political Science Department and author of this article, as our new 3/4 time statistical consultant. - Ed.

        Research and Statistical Support (RSS) is a team of consultantsworking under the auspices of Academic Computing Services to serve the research community on campus. Our duty is to cater to the needs of all UNT researchers, faculty members and students. Services offered by RSS cover four areas: data acquisition, data analysis, application support and instructional support. Recently, we have set up our new "virtual base" on the World Wide Web. The site aims at enhancing our support servicesin these four areas. This article gives a brief explanation on our new "cyberservices."

        http://www-lan.unt.edu/RSS/

        Above is the URL for our new RSS Web page. This Web page isdesigned to act as an interface between RSS and our clientele with regards toresearch and statistical support. Primarily, it covers services in the following four areas:

        1. Data Acquisition: Providing data resources for academicresearch is one of our major tasks. As a member of the Inter-university Consortium for Political and Social Research (ICPSR), our university enjoys the privilege of accessing a wealth of data collections from the consortium located in Michigan. Largest of its kind, ICPSR archives a wide variety of data sets, from economic data such as World EconomicIndicators, to demographics like the US Bureau of Census Population data.Every year, ICPSR publishes a catalog displaying all available and updated data titles. Thanks to the Information Superhighway, the catalog is now on-line. In enhancing the search for data set titles, we provide the relevant links on the RSS home page for browsing data sets onthe ICPSR's archive records. These links include the ICPSRhomepage at The University of Michigan and a gateway at Columbia University through which we can search for data sets of interest by keywords, titles and authors. Before long, we will have our own database ready for our clients to find data set titles from the whole ICPSR archival database.

          Over the years, we have also amassed almost 600 data titles on campus ready for use. To help tap into this data source, James Yarbrough, Research and Statistical Support Manager, wrote an article, entitled "Accessing and Using ICPSR Datasets," in the September/October 1995 issue of Benchmarks (Volume 16 Number 5). The article provides detailed instructions on how to access the ICPSR data sets. Interested readers may point their web browser to: http://www.unt.edu/UNT/departments/CC/Benchmarks/benchmarks_html/sepoct95/icpsr.htm

          To order data sets, either on campus or from ICPSR, pleaseconnect to our Web page, click on the link, How to Obtain Data sets from ICPSR, and get to the instructions page. Prior to the search for data sets, we recommend that you get a computer User-ID on your preferred platform (CMS or UNIX). Since some of the data sets are huge in size, it is usually preferrable to subset the data set on CMS or UNIXbefore downloading to a microcomputer system.

          Having a computer User-ID also facilitates communication with us via electronic mail. Application for computer accounts can be made at the Computing Center Support Services in Room 119 of the Information Science Building.

          Regarding the data set hunting, you can search the right dataset either via our on-campus listings or the ICPSR searchable database. Write down the ICPSR number and data set title, fill out and submit the on-line request form, and we will process your request on the same day.

          Apart from the ICPSR data archives, data from other sourceslike Center for Research in Security Prices (CRSP) and Standard and Poor's (COMPUSTAT) are also available upon request. For more details regarding the aforementioned data sets, please refer to another Benchmarks article by our former Research Statistical SupportManager, Dr. Panu Sittiwong, in the September/October 1994 issue (Volume 15,Number 5). The URL for this article is: http://www.unt.edu/UNT/departments/CC/Benchmarks/benchmarks_html/sepoct94/stat94.htm

          In the near future, we plan to put more data links on our sitein an effort to make more data sets available and accessible to UNT users.

        2. Data Analysis: Currently, Academic Computing Services maintainsa site license for SAS and SPSS. The RSS consultants provide technical programming support for both software applications. Our support services also include installation and regular annual updates, and programming support for data management and data analyses. On our web site, we provide brief manuals and sample programs to guide users to manage and analyze their data using SAS and SPSS. A list of internet links to the statistical software providers such as SPSS, SAS, S-Plus and Statistica is also available for interested users to browse.

        3. Short Courses: Research and Statistical Support consultantsregularly offer short courses on using SPSS and SAS. Interested faculty members and students are welcome to register for these free courses. On our Web site you will find the schedules and instructions for registration. In the future, we plan to put up class notes, exercises and sample programs for users to take "cyberclasses."

        4. Sites for Statistical resources: Within this topic, we havegarnered the links of on-line academic journals regarding Statistics and methodological issues. In the near future, we will put on some more sites of statistics departments and institutes to keep you up-to-date with the latest innovations and trends in statistical analysis.

        The RSS web site is designed to enhance our support service tothe research community on campus. We invite you to point your web browser to our site, make the best use of it and give us your input towards the betterment of our service. We hope the site provides a starting point for our further communication in the near future. Only with your support will we better support your research. Should you have any suggestions or comments, please feel free to send us E-mail: James Yarbrough, Research and Statistical Support Manager (james@vm.acs.unt.edu), Karl Ho, Statistical Consultant (kho@unt.edu).

        References

        • Sittiwong, Panu. 1994. "Research and Statistical Support FromAcademic Computing Services." Benchmarks. Volume 15 Number 5.

        • Yarbrough, James. 1995. "Accessing and Using ICPSR Datasets."Benchmarks. Volume 16 Number 5.

        Previous Article <== ==> Next Article

        If you have problems or questions about this server, please contact us as soon as possible. You can send mail to the following address: www@unt.edu