Degree Type


Date of Award


Degree Name

Doctor of Philosophy



First Advisor

William Q. Meeker

Second Advisor

Steven A. Freeman


Statistical analysis is an essential tool in safety and health research. This thesis is composed of statistical analyses for three different types of safety and health projects: power-take-off entanglements, farm stress, and heart problems and diabetes, in additional the related statistical issues are discussed;In the power-take-off entanglements project, an experimental design was used to study the risk of entanglement in power-take-off driveline. The response variable was a binary variable indicating whether a dangerous "entanglement" occurred or not. The experimental factors were angle of introduction, length of specimen, and type (stiffness) of material. The main conclusion was that the entanglement risks are higher when the angle of introduction is closer to perpendicular, or when the length of specimen is increased;The farm stress study was based on a survey conducted among Iowa farmers. The purposes of the study were to determine the most stressful events/activities for farmers, demonstrate whether demographic groups affect stress levels, and to identify unnecessary survey questions. The response variables were stress levels on 62 events/activities. The main conclusion was that the respondents felt different level of stress dependent on age, gender, and so on; death of a spouse and the death of a child were found to be the most stressful events. Also, it was determined that some questions could be removed from the survey without significant loss of information;The health problems and diabetes study used survey data collected by the Centers for Disease Control and Prevention's Behavioral Risk Factor Surveillance System (BRFSS). The interesting random variables were whether the interviewee had heart disease, or/and diabetes, and personal health status. The major purposes of the study were checking whether personal habits and background were associated with exposure to diabetes or/and heart disease, and whether a person's background was associated with personal health status. Main conclusions included a strong relationship between diabetes and heart disease and that personal background and habits, such as age, gender, diet habits, etc. are associated with the proportion of heart disease or/and diabetes. In addition, personal health status is associated with age, gender and whether individuals live with partners or not.



Digital Repository @ Iowa State University,

Copyright Owner

Qi Jiang



Proquest ID


File Format


File Size

165 pages