Expert Judgment in Groups Chapter 13

Similar documents
Major Milestones, Team Activities, and Individual Deliverables

ADAPTIVE PLANNING. 1 Powered by POeT Solvers Limited

Enhancing Learning with a Poster Session in Engineering Economy

CHAPTER 4: REIMBURSEMENT STRATEGIES 24

STABILISATION AND PROCESS IMPROVEMENT IN NAB

(Includes a Detailed Analysis of Responses to Overall Satisfaction and Quality of Academic Advising Items) By Steve Chatman

ECE-492 SENIOR ADVANCED DESIGN PROJECT

Unit 3. Design Activity. Overview. Purpose. Profile

Librarians of Highlights of a survey of RUL faculty. June 7, Librarians of 2023 June 7, / 11

The Good Judgment Project: A large scale test of different methods of combining expert predictions

Lucy Calkins Units of Study 3-5 Heinemann Books Support Document. Designed to support the implementation of the Lucy Calkins Curriculum

Software Development Plan

Semi-Supervised GMM and DNN Acoustic Model Training with Multi-system Combination and Confidence Re-calibration

Software Maintenance

THE PENNSYLVANIA STATE UNIVERSITY SCHREYER HONORS COLLEGE DEPARTMENT OF MATHEMATICS ASSESSING THE EFFECTIVENESS OF MULTIPLE CHOICE MATH TESTS

Thank you letters to teachers >>>CLICK HERE<<<

Outreach Connect User Manual

University of Waterloo School of Accountancy. AFM 102: Introductory Management Accounting. Fall Term 2004: Section 4

10.2. Behavior models

Mike Cohn - background

Self Study Report Computer Science

Task Types. Duration, Work and Units Prepared by

First Grade Standards

An Investigation into Team-Based Planning

Visit us at:

Qualitative Site Review Protocol for DC Charter Schools

John Jay College of Criminal Justice, CUNY ASSESSMENT REPORT: SPRING Undergraduate Public Administration Major

Author: Justyna Kowalczys Stowarzyszenie Angielski w Medycynie (PL) Feb 2015

Course evaluations at Chalmers

Generating Test Cases From Use Cases

Strategic Practice: Career Practitioner Case Study

CREATING SAFE AND INCLUSIVE SCHOOLS: A FRAMEWORK FOR SELF-ASSESSMENT. Created by: Great Lakes Equity Center

Instructions and Guidelines for Promotion and Tenure Review of IUB Librarians

VIEW: An Assessment of Problem Solving Style

Using Team-based learning for the Career Research Project. Francine White. LaGuardia Community College

Integrating simulation into the engineering curriculum: a case study

National Survey of Student Engagement at UND Highlights for Students. Sue Erickson Carmen Williams Office of Institutional Research April 19, 2012

Executive Guide to Simulation for Health

Understanding Fair Trade

Value Creation Through! Integration Workshop! Value Stream Analysis and Mapping for PD! January 31, 2002!

Measures of the Location of the Data

Objective: Add decimals using place value strategies, and relate those strategies to a written method.

Student Assessment and Evaluation: The Alberta Teaching Profession s View

Rover Races Grades: 3-5 Prep Time: ~45 Minutes Lesson Time: ~105 minutes

Promotion and Tenure Policy

Probability estimates in a scenario tree

PAGE(S) WHERE TAUGHT If sub mission ins not a book, cite appropriate location(s))

Helping Graduate Students Join an Online Learning Community

PM tutor. Estimate Activity Durations Part 2. Presented by Dipo Tepede, PMP, SSBB, MBA. Empowering Excellence. Powered by POeT Solvers Limited

Guidelines for the Use of the Continuing Education Unit (CEU)

IBM Software Group. Mastering Requirements Management with Use Cases Module 6: Define the System

Arizona s English Language Arts Standards th Grade ARIZONA DEPARTMENT OF EDUCATION HIGH ACADEMIC STANDARDS FOR STUDENTS

Probabilistic Latent Semantic Analysis

1/25/2012. Common Core Georgia Performance Standards Grade 4 English Language Arts. Andria Bunner Sallie Mills ELA Program Specialists

CollaboFramework. Framework and Methodologies for Collaborative Research in Digital Humanities. DHN Workshop. Organizers:

Activities, Exercises, Assignments Copyright 2009 Cem Kaner 1


Preparing for the School Census Autumn 2017 Return preparation guide. English Primary, Nursery and Special Phase Schools Applicable to 7.

Device Design And Process Window Analysis Of A Deep- Submicron Cmos Vlsi Technology (The Six Sigma Research Institute Series) By Philip E.

How to set up gradebook categories in Moodle 2.

MAINTAINING CURRICULUM CONSISTENCY OF TECHNICAL AND VOCATIONAL EDUCATIONAL PROGRAMS THROUGH TEACHER DESIGN TEAMS

Iep Data Collection Templates

Extended Common Core Social Studies Lesson Plan Template

Gr. 9 Geography. Canada: Creating a Sustainable Future DAY 1

Iowa School District Profiles. Le Mars

Clinical Quality in EMS. Noah J. Reiter, MPA, EMT-P EMS Director Lenox Hill Hospital (Rice University 00)

Maximizing Learning Through Course Alignment and Experience with Different Types of Knowledge

Faculty Meetings. From Dissemination. To Engagement. Jessica Lyons MaryBeth Scullion Rachel Wagner City of Tonawanda School District, NY

Radius STEM Readiness TM

Ohio s New Learning Standards: K-12 World Languages

If we want to measure the amount of cereal inside the box, what tool would we use: string, square tiles, or cubes?

Notetaking Directions

Grade 11 Language Arts (2 Semester Course) CURRICULUM. Course Description ENGLISH 11 (2 Semester Course) Duration: 2 Semesters Prerequisite: None

MBA6941, Managing Project Teams Course Syllabus. Course Description. Prerequisites. Course Textbook. Course Learning Objectives.

Launching GO 4 Schools as a whole school approach

SACS Reaffirmation of Accreditation: Process and Reports

Practice Examination IREB

Ministry of Education, Republic of Palau Executive Summary

Stakeholder Debate: Wind Energy

READY TO WORK PROGRAM INSTRUCTOR GUIDE PART I

Early Warning System Implementation Guide

Policy for Hiring, Evaluation, and Promotion of Full-time, Ranked, Non-Regular Faculty Department of Philosophy

Oklahoma State University Policy and Procedures

Life and career planning

TOPIC: Biennial Exempt Market Salary Survey Report and FY Structures Adjustment

FIELD PLACEMENT PROGRAM: COURSE HANDBOOK

A CASE STUDY FOR THE SYSTEMS APPROACH FOR DEVELOPING CURRICULA DON T THROW OUT THE BABY WITH THE BATH WATER. Dr. Anthony A.

STRATEGIC LEADERSHIP PROCESSES

MOTION PICTURE ANALYSIS FIRST READING (VIEWING)

Handbook for Graduate Students in TESL and Applied Linguistics Programs

Programme Specification

Economics 201 Principles of Microeconomics Fall 2010 MWF 10:00 10:50am 160 Bryan Building

August 22, Materials are due on the first workday after the deadline.

Teaching a Discussion Section

Linking the Common European Framework of Reference and the Michigan English Language Assessment Battery Technical Report

Critical Thinking in the Workplace. for City of Tallahassee Gabrielle K. Gabrielli, Ph.D.

On the Design of Group Decision Processes for Electronic Meeting Rooms

Arkansas Private Option Medicaid expansion is putting state taxpayers on the hook for millions in cost overruns

Trends in College Pricing

Best Practices in Internet Ministry Released November 7, 2008

Transcription:

Expert Judgment in Groups Chapter 13 Group judgments when estimating early in the project or for estimating LARGE unknowns. 3 Simple Rules: 1. Each team member estimates pieces of the project individually then meets with others to compare estimates 2. Don t just average our estimates discuss the differences 3. Arrive at a consensus if not, discuss differences and get buy-in from all

Figure 13-1 Reviewed Estimates Individual Estimates Individual Estimates: MRE 55% Group Reviewed Estimates: MRE 30% 92% of Group Estimates better than Individual Estimates How many experts are enough? 3 to 5 with different backgrounds

Wideband Delphi Technique 1. Each estimator gets a specification and an estimation form 2. Estimators prepare initial estimates 3. Group meets to discuss estimation issues related to the project. If there is early agreement, a devil s advocate is chosen 4. Estimates are submitted anonymously 5. Summary is represented with an iteration form 6. Estimators meet to discuss variations in their estimates 7. Estimators vote anonymously on the average estimate If there are no votes, go to Step 3 Average Estimate 0 5 10 15 20 Staff Months 8. The final estimate is the single point estimate or a range agreed to in the discussion (with the single point estimate being the expected value)

Effectiveness of Wideband Delphi Author s experience For the first 25 groups I worked with Wideband Delphi reduces estimation error by an average of roughly 40% Two-thirds of his groups produced more accurate estimates using Wideband Delphi 10 Groups that produced the worst estimates, 8 of 10 improved accuracy by 60% Author is a believer especially for avoiding erroneous results. Law of Large Numbers? The correct answer lies somewhere between the lowest and highest estimate But 20% of groups initial estimation ranges did not include the actual number

The Truth is Out There Correct Answer Range of Initial Individual estimates In none of the observed cases Did the final Wideband Delphi Estimate move in this direction (out of the initial range, away From the correct answer) In one third of cases, the final Wideband Delphi estimate moved Into this range (out of the original range, toward the correct answer)

When to Use Wideband Delphi Where in the wide part of the Cone of Uncertainty Author s experience Reduced estimation error from 290% to 170% (a 41% reduction) 170% error is high! A reduction from 50% to 30% is a 40% reduction NOTE The studies give this method mixed reviews. The approach is costly burning hours of staff time But again early in the Cone there may be no better approach. (project is complex, new, etc.)

Wideband Delphi Process Flow Planning Kickoff Individual Preparation Estimation Assemble Tasks Review Results Planning Defining and Scoping Large problems decomposed Estimated separately, possibly by different team Provide all the information available Participants (experienced and trained) Moderator Project Manager 2 to 4 other estimators

Wideband Delphi Process Flow Planning Kickoff Individual Preparation Estimation Assemble Tasks Review Results Kickoff Up to an hour Estimators given problem specs and whatever assumptions or project constraints apply Review estimation objectives Discuss problem and whatever estimation issues exist Moderator decides if the team is ready to roll At this point: Appropriate team members selected Kickoff meeting complete Team agrees to estimation goal and units PM can participate Estimators have the information needed to participate effectively

Wideband Delphi Process Flow Planning Kickoff Individual Preparation Estimation Assemble Tasks Review Results Individual Preparations Participants develop initial list of tasks Each estimates the effort needed for each task Break each activity into tasks that are more easily estimated Rule: no task should take more then 20 hours Record any tasks for related or supporting activities Record your assumptions Guidelines: Assume one you will perform the tasks Assume all tasks performed sequentially Assume you can devote uninterrupted time to each task In calendar time, list any known waiting times you expect

Estimation Form Task Estimate #1 Change #1 Change #2 Change #3 Change

Wideband Delphi Process Flow Planning Kickoff Individual Preparation Estimation Assemble Tasks Review Results Estimation Moderator collects estimates & creates chart Round 4 Round 3 Round 2 Round 1 0 500 1000 1500 2000 2500 Labor Hours (tasks only) Estimates are anonymous Estimators each read their task list (include assumptions and whatever questions or issues they want addressed)

Estimation Combining task lists leads to a more complete list than any one estimator produces limit maybe several dozen Discuss assumptions, estimation issues and questions about the problem Converging on a shared set Then each estimator modifies their estimates concurrently Revising their task list revising estimates based on new info Moderator collects estimates and plots the next Round Additional rounds with same type of discussions will narrow the distribution

Estimation Continue, until: Completed at least 4 rounds Estimates have converged to an acceptable range (defined in advance) Allotted estimation meeting time is over (typically 2 hours), or All participants are unwilling to alter their latest estimates Moderator roll: Controls meeting Time-boxed discussions (15 to 20 minutes) Facilitate particpation Preserves anonymity of estimators

Wideband Delphi Process Flow Planning Kickoff Individual Preparation Estimation Assemble Tasks Review Results Assemble Tasks Defining and Scoping Large problems decomposed Estimated separately, possibly by different team Provide all the information available Participants (experienced and trained) Moderator Project Manager 2 to 4 other estimators

Wideband Delphi Process Flow Planning Kickoff Individual Preparation Estimation Assemble Tasks Review Results Assemble Tasks Create single Master Task List Merge individual assumptions, quality- and process-related activities, overhead tasks and wait times. Note. All estimators should define their tasks clearly, so that merging is consistent (no same task name, different specs)

Wideband Delphi Process Flow Planning Kickoff Individual Preparation Estimation Assemble Tasks Review Results Review Results Team reviews summarized results and reaches agreement (buy-in) on final result 30- to 60-minute review meeting Retrospective on the process identifying possible improvements

Completing the Estimation Exit criteria: The overall task list has been assembled Summarized list of estimating assumptions is complete Estimators have reached consensus on how their estimates were used and synthesized into a single set with an acceptable range.