More Powerful but Still Easy to Use Data Analysis with Jmp

The new version of JMP (Version 5) has the ideal toolset for providing graphical as well as analytical solutions necessary for the designing, analysis and presentation of the results for many research studies. Its dynamic exploratory graphical interface along with its new tools and the improved scripting language JSL provide the user with functionality that is ahead of many programs currently in market. Yet it is still very simple and easy to use for the analyst without requirement of knowledge of any programming language unless the user needs to add to JMP's existing set of tools. Some of the multivariate data analysis capabilities of JMP5 will be demonstrated using real life data on bone mineral density of women. Introduction: The developer of JMP Dr. John Sall wrote“JMP is a smaller sibling to SAS® aimed at scientists, engineers, and other researchers who need to analyze data. JMP is to SAS like a spreadsheet is to a database, smaller and geared to interactive desktop uses, but able to merge into the larger enterprise easily. One of the most prominent uses of JMP is to design and analyze experiments. JMP has always been strongest in its graphical approach to analyzing data. There is a graph for almost every statistic, and most of the graphs are interactive.” [Sall and Maahs-Fladung (2002)]. JMP Version 5.0, launched in June 2002 includes five new platforms but due to space limitations we will explore only two of them in this paper, namely, PARTITION, a tree based recursive-modeling tool, and NEURAL NET, a simple neural network modeling tool with one hidden layer (JMP(2002)). The data collected on bone mineral density in women will be used to demonstrate these two techniques. The condition of decreased bone density in women is known as osteoporosis. A study was conducted at the University of Arkansas to assess risk factors associated with osteoporosis in women. Bone density measurements were collected on the left femur using Dual X-ray Absorbtometry (DEXA). The third National Health and Nutrition Examination Survey (NHANES III) was the instrument used in this study. It is an adult household questionnaire, which contains questions about demographics, physical activity, health history, diet recall, and food frequency. In the diet recall and food frequency questions, the subjects were asked to recall their food intake in the past 24 hours. Clinical data was collected through physical examination at mobile examination centers (MEC), results of blood and urine specimens collected, bone density measurements, etc. The bone mineral density variables recorded using DEXA are as follows: bone mineral density of femur neck (in gm/cm-sq, BDPFNBMD), bone mineral density of trochanter region (in gm/cm-sq, BDPTRBMD), bone mineral densty of intertrochanter region (in gm/cm-sq, BDPINBMD), bone mineral density of total region (in gm/cm-sq, BDPTOBMD). See Fig. 1. Categorical as well as continuous measurements were recorded for each subject using NHANES III. The categorical variables recorded were race-ethnicity (DMARETHN), body mass index (BMPBMI), smoker status (Smoker, non-smoker, or former smoker), alcohol consumption per day (ALCOHOL: < 1 drink or 1+ drinks), estrogen level (ESTROGEN: replete or not), calories expended (CalExp). The continuous measurements were recorded for following variables: Figure 1: X-ray of a left-femure Joint Statistical Meetings Statistical Computing Section