Each year over 6,000 workers are killed while earning a living in the United States. From these deaths, how can researchers determine if employees in the manufacturing industry are at greater risk of a traumatic occupational fatality than those employed in agriculture or any other industry? Similarly, does the risk of traumatic occupational fatality differ among states? To answer such questions, two measurements are normally used: frequency of occurrence and incidence rate. These measures are used to identify worker groups at greatest risk of fatal injury, target research and prevention activities, and evaluate the impact of these activities. Developing the best methods to accurately identify worker groups at greatest risk of losing their life while at work is always important, but even more so in times of limited government resources. The accuracy of a traumatic occupational fatality incidence rate depends on how closely the denominator and numerator represent the same population. Selecting data from different employment programs for the denominator when calculating incidence rates using the National Traumatic Occupational Fatalities surveillance system for the numerator has shown dramatically different results. This analysis points out that researchers must carefully: choose the data they employ; evaluate the meaning of calculated incidence rates; and document the data sources to ensure proper interpretation by others. Additional research and evaluation are necessary to improve data sources, analytical methods and tools to ensure effective resource allocations for the occupational safety and health field.