HOME PRODUCTS SERVICES EVENTS FORUM CONTACTS

LOG-IN
FREE REGISTRATION
SEARCH
EMAIL

PSM FAQ

Here are some of the frequently asked questions about PSM, compiled from support staff, PSM trainers, and Guidebook authors (revised: 01/02/02). If you cannot find your answer here or elsewhere on our Web site, please call us at 703-405-2191, or email the PSM Support Center.

Table of Contents

What is Practical Software Measurement?
What is PSM Insight?
How does PSM apply to maintenance?
How can you get the developers to change what they are measuring when the issues change mid-stream on a program?
How does measurement support an IPT (Integrated Product Team)?
Why is an independent analysis capability needed?
My program's issues are different than those listed in the PSM Guidebook. How can I use PSM?
How can the evaluation of the program personnel's performance be avoided on a small project?
What are the specific policies and requirements for software measurement in my service/organization?
How do you decide when measurement results are "bad enough" (variation between plans and actuals) to merit further investigation or corrective action?
I am new to all this measurement stuff. How do I get started?


Practical Software Measurement: Objective Information for Decision Makers is the definitive guide to PSM.
Details...


PSM Insight Version 4.2.2 is full of new features to help you work smarter.
Details...

 

What is Practical Software Measurement?

Practical Software Measurement (PSM) is an objective process for analyzing project issues, risks, and financial management, focusing on software project management. PSM provides government and industry managers with objective information for making sound decisions and for meeting cost, schedule, and technical objectives. It is based on actual DoD, government, and industry experience, representing the best practices in the software and engineering communities.

Back to Top

 

What is PSM Insight?

PSM Insight is a PC-based software tool that automates the Practical Software Measurement (PSM) process. PSM Insight includes tailoring, data entry, and analysis functions to help you develop a project-specific software measurement database and analyses. While PSM Insight provides templates of commonly-used issues and measures, it is also completely flexible for you to customize analysis to project-specific needs.

PSM Insight is sponsored by the Department of Defense and the U.S. Army Software Metrics Office.

Back to Top

 

How does PSM apply to maintenance?

If the maintenance (or sustaining support) process is organized into periodic releases, PSM is implemented similar to a new systems development process. Issues are decided during the planning of the release and are based on things like schedule and staffing constraints, requirements/change request volume, etc.

If the maintenance process treats every change request as ad hoc or continuous releases, issues and measures focus more on maintaining and improving baselines over time, versus monitoring discreet events. For instance, improvement goals may focus on reducing the backlog of change requests over time, responding more quickly to customer requests, improving software reliability, and maintaining or improving software maintainability.

Finally, the issues and measures selected during a maintenance project will probably be different than those selected during the program's development.

Back to Top

 

How can you get the developers to change what they are measuring when the issues change mid-stream on a program?

The first thing to consider is whether the data being supplied really needs to change or whether the same data can be used to provide insight into new issues or areas of concern. If changes really are needed, make sure that what you need can realistically be delivered.

Think of it this way: A new issue for the government should almost always be equally important to the developer; therefore, they should already be generating some data to address the problem. Find out what that data is, and plan to use it, if possible. Also, make sure the developers understand why something represents a new issue for the government.

Finally, consider reducing the requirements to supply other measurement data (i.e., drop the requirement to provide measures for an issue which is less important or under control, or allow reporting of some data less frequently).

Back to Top

 

How does measurement support an IPT (Integrated Product Team)?

Integrated Product Teams (IPTs) integrate all areas of the program team, including the government and developer personnel, in order to improve and streamline the communications and program decision-making processes. A goal of IPTs is to de-centralize decision-making and cut down on red tape.

IPTs can be used to jointly define the measurement program requirements (tailoring). Primarily, IPTs provide a logical infrastructure for the regular review of measurement results. IPTs can analyze and discuss the meaning of measurement results; they can determine root causes, consider alternatives, and recommend, and in some cases initiate, corrective actions.

Back to Top

 

Why is an independent analysis capability needed?

Even if the government and the developer are working closely together, it is important for the government to have the ability to do their own analysis with raw measurement data independent from the developer organization. This means either having raw data delivered to the government regularly or having on-line access to the contractors data collection systems and databases.

This does not imply that the government thinks that the contractor will manipulate the data and should in no way be construed as a lack of trust. Instituting this type of capability primarily forces the government to thoroughly understand the true project status and not make any assumptions about what the developer is communicating. When both parties analyze results and then get together to discuss them, the questions they each bring to the table are the basis for high-quality communications.

Back to Top

 

My program's issues are different than those listed in the PSM Guidebook. How can I use PSM?

The PSM issues are simply mechanisms designed to help software programs identify and classify their own issues. PSM's common issues simply characterize problems which plague many software-intensive programs. Programs can either 1) use these issues as a starting point when considering what their issues really are or 2) map their specific issues to one of these common issues.

Also, remember that new issues, categories, or measures can be defined and used within the PSM Guidance. The PSM process stays the same, no matter what the issue is.

Back to Top

 

How can the evaluation of the program personnel's performance be avoided on a small project?

This is a problem on any project, really. However, it is important to consider that just because measurement results might provide some indication of a person's performance doesn't mean the results need to be used for this purpose. The decision not to use results this way must be addressed through education.

What needs to be emphasized to anyone using or with access to measurement results is that the results are simply a problem-solving tool. As Deming points out, most of the problems are in the process; poor results are usually not due to an individual's poor performance. Results should be used to identify and isolate problems, then to pinpoint what process caused the problem. For instance, poor schedule performance may appear to be due to poor project management abilities, but are, in fact, due to a poor estimating process. It's management's responsibility to help the organization develop reliable processes. When measurement data must be used for oversight purposes, make sure the people using it understand this.

Also, consider making it a rule to only show data for oversight purposes at the group level.

Back to Top

 

What are the specific policies and requirements for software measurement in my service/organization?

At the DoD level, the 5000 series requires measurement. The AIS world is specifically required to measure according to PSM's 6 common issues. In general, DoD policy is becoming less and less definitive. With this in mind, PSM is designed to help people meet whatever requirements they are faced with.

Instructors should be familiar with the latest policy statements, standards, and program requirements for measurement, both at the DoD level and also those requirements specific to the service/organization for which they will be providing training. The "Supplemental Materials" section of the Instructor Notebook contains some of this information. Other good sources for this information are the U.S. Air Force's Crosstalk,the U.S. Army OPTEC's Insight newsletter, etc.

Back to Top

 

How do you decide when measurement results are "bad enough" (variation between plans and actuals) to merit further investigation or corrective action?

There is no "right" answer to this question because variations between plans and actuals must be interpreted within the context of the program, and depend on the program's risk tolerance. In most cases, knowing when something is a "bad enough" problem is pretty obvious. People look not just at a single current variation, but also use the "preponderance of evidence" approach (e.g., integrated analysis) and consider what the trends suggest.

Many organizations set "rules of thumb" for certain issues/indicators. A standard rule of thumb is to pay special attention to any indicator with a 20% variance overall or a 10% variance in any period.

Back to Top

 

I am new to all this measurement stuff. How do I get started?

Read the book. People new to software measurement should read Part 1 of the PSM Guidebook first to get an overview of the PSM approach. People faced with the immediate task of implementing measurement on a program should use the more detailed guidance found in Parts 2,3, and 4 while they are performing the steps necessary to select and apply measurement in real-time. Students which are more comfortable learning by example should read through one of the case studies in Part 5. Students who need deeper coverage of a particular area of software measurement should refer to the bibliography in Part 6, which recommends a number of books and reports on software measurement.

Back to Top

Bottom Links Home | Products | Services | Events | Forum | Contacts