A Comparison of RADC and NASA/SEL Software Development Data. DACS Technical Monograph Series. Revised
Abstract
In September of 1978, Richard Nelson of Rome Air Development Center (RADC) completed a report entitled Software Data Collection and Analysis in which he examined several statistical relationships within the RADC Software Productivity Database. The relationships studied attempted to relate the size of a software project with various other metrics describing the development process: The seven primary relationships studied by Nelson are given below: Project Size vs. Productivity (source lines per manmonth) Project Size vs. Development Effort (manmonths) Project Size vs. Development Duration (months) Project Size vs. Average Manloading (manmonths per month) Project Size vs. Number of Errors Project Size vs. Spatial Error Rate (number of errors per 1000 source lines) Project Size vs. Effort Based Error Rate (number of errors per 10 manmonths of development effort). This report summarizes the results of a similar examination of all but one of these relationships when data from the NASA/SEL databases is merged with the RADC data. The relationship between Project Size and Average Manloading will not be examined because of the different methods used on computing this metric for the two databases. However, another possible relationship, given below, is examined. Project Size vs. Temporal Error Rate (number of errors per month of development time).
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 1982
- Accession Number
- ADA131756