Return to Brian Schott @ GSU
January 7, 1997
The questions contained in the Introduction to Business Statistics database (IBSD) have been selected from a huge database of questions and answers developed under National Science Foundation (NSF) funding in the mid 70's. The NSF database is available in compressed form (.zip) as five ASCII files tests. A readme file is also available for the NSF database readme.txt. The readme file contains information regarding the indexing and categorizing employed by the NSF database developers and other information about the database and it's contributors. The readme file is recommended browsing for users of IBSD (described next).
I selected and organized the questions contained in IBSD during December 1996. The 811 questions were selected manually from the 3003 pages of questions and answers of the NSF database. The questions are reorganized (relative to the NSF database) to agree with the chapters covered in the course DSc310 Introduction to Business Statistics at Georgia State University. The textbook for the course is Harvey Brightman and Howard Schneider's Statistics for Business Problem Solving, 2nd Edition (South-Western Publishing Company, 1994).
The questions are best accessed from the "Contents" page qcontent.html. The Contents page displays the questions in chapter order. From the list of chapters choose one by clicking on the chapter name. A teaser phrase copied from each question is provided to inform the reader's choice of question. From any chapter's list of question teasers, click on a teaser to select a question. At the bottom of each question click on the appropriate phrase to either look at the answer or return to the question or list of questions.
The question types include multiple choice (50% -- all percentages are approximate), true/false (14%), short answer (13%), numerical answer (11%), definition (7%), and a few fill-in and essay.
Several questions have been slightly changed for inclusion in IBSD. Most of the changes reduce the calculation burden for the questions; for example calculations of the variance and sums of squares were moved from answers to questions.
I regret that the questions available in the NSF database do not reflect the emphasis we place on certain topics. For example, we place emphasis on identifying outliers using Tukey's fences and the "Empirical Rule," neither of which are covered in the NSF database. In our course we often limit our coverage to symmetric confidence intervals for the unknown population mean and the unknown population proportion, but cover both one- and two-sided tests for the population mean. Our course emphasizes regression analysis, not correlation analysis, but the NSF database reverses this emphasis. In the future I hope to revise the coverage of topics.
Please give me your comments, ideas, help, and recommendations. The IBSD database is very much under construction.