Please enable JavaScript.
Coggle requires JavaScript to display documents.
Chapter 7: Measurements in the personal software process - Coggle Diagram
Chapter 7: Measurements in the personal software process
Measurement Overview
You measure to get data, and you want data to help you with the following
Gain quantitative understanding
Evaluate a product, process, or organization
Control a product or a process Make an estimate or a plan
The principal measurement categories are the following
Absolute/relative
Objective/subjective
Explicit/derived
Dynamic/static
Predictive/explanatory
Fundamental Process Measures
PRODUCT MEASURES
Product measures generally refer to the volume of product produced. They in- clude LOC, pages of documentation, numbers of screens, numbers of files
These measures may be of various product elements, such as modules, components, or manuals. You can also measure by phase, such as the amount of code produced in the implementation phase or the LOC changed
PROCESS MEASURES
Process measures quantify the behavior of your process. They are generally ob- jective, absolute, explicit, and dynamic. They also provide the elements for a host of useful derived measures.
Here, you count things that happen, such as the numbers of defects found in test, requirements changes, or milestones met.
RESOURCE MEASURES
Resource measures apply to labor hours, the principal-software development re- source. Here, the concerns are working hours, job categories, and task activities.
While common resource measures are in programmer months or weeks, mea- sures of personal time are not terribly useful if the unit is longer than minutes or fractions of an hour
Goal-Question-Metric Paradigm
Framework for guiding measurement efforts.
define the principal goals for your activity, o construct a comprehensive set of questions to help you achieve these goals.
Define and gather the data required to answer these questions.
DEFINING GOALS
You should thus have clear goals of what you are trying to do before you start gathering data.
THE GOALS HIERARCHY
The goals hierarchy can help you to put your goals in proper context.
Start by trying to connect your goals to your organization's goals. At a senior management level, these goals are probably quite stable and are often in broad- enough terms to be readily understandable.
ROLE OF QUESTIONS
Some typical metric related questions could be the following:
For each process goal, where did I start, where am I now, and where do I want to go
What is important about this goal?
What is the best that has been achieved against this goal?
Does improved performance against this goal have any absolute limit, and if so, what is it?
METRICS DEFINITION
To ensure the desired data are consistently gathered and are presented in an understandable und retrievable form, you should design forms for gathering the data.
General PSP Objectives, Goals, and Questions
One reason to study the PSP is to improve your effectiveness as a software engi- neer.
Understand how the personal software development process works. to determine the steps you could take to improve product quality.
Determine the impact of process changes on your productivity. to establish benchmarks to measure process improvement
Gathering Data
MANUAL DATA GATHERING
Examples of the need for personal judgment in PSP data gathering are the following
It would be impossible for your computer to know what you were doing when you stopped using the keyboard
It would also be impossible to count defects automatically from code changes
FORMS AND TEMPLATES
Forms and templates are important tools for gathering and using data
Designed forms are helpful when you are generating estimates and plans and pro- vide a convenient repository for the actual data at project completion.
Forms are used when the amount of data you are gathering is fixed, and tem- plates are used when the volume of data is unpredictable
THE DEFECT DATABASE
A table of the numbers of defects injected and removed by phase
Data on the numbers of defects that were in the product at phase entry but not found during that phase.
Data on the numbers and types of defects found in a specific phase.
THE PSP SPREADSHEET
Some of the analyses that can be made with these data are the following:
Regression calculations
Yield
Productivity
Charts
Establishing a Baseline for Your Personal Process
Personal process re- sults vary considerably and you may sometimes feel you are improving when you are not. You can often tell only by statistically examining a large enough volume of data to get significant results.