A Method For Designing Improvements in Organizations, Products, and Services

Stuart UmplebyResearch Program in Social and

Organizational LearningThe George Washington

UniversityWashington, DC USA

E-mail: umpleby@gwu.edu

Dragan TevdovskiMathematics, Statistics and Informatics University Sts. Cyril and Methodius Skopje, MacedoniaE-mail: dragan@eccf.ukim.edu.mk

Second Conference of the Washington Academy of Sciences

Washington DC, March 2006

Introduction

A method for determining priorities for improvement in an organization

Priority means high importance and low performance

Quality Improvement Priority Matrix

The approach to design This approach to design is “piecemeal”

rather than “utopian” It is “bottom up” rather than “top down” It uses the judgments of employees or

customers Features to improve are ranked by

urgency Several projects can be worked on

simultaneously

Quality Improvement Priority Matrix

References

The method was first described by the specialists from GTE Directories Corporation in 1995

Armstrong Building Products Operation used the method in1996

Naoumova and Umpleby (2002) - evaluation of visiting scholar programs

Melnychenko and Umpleby (2001) and Karapetyan and Umpleby (2002) used QIPM in a university department

Prytula (2004) introduced the importance / performance ratio

Dubina (2005) used cluster analysis and proposed standard deviation as a measure of agreement or disagreement

Goals of the Paper

Understand more fully the priorities of the Department of Management Science at The George Washington University (GWU), USA, and the Department of Management at Kazan State University (KSU), Kazan, Russia

Use and develop new methods to compare QIPMs for two organizations

The Data

A questionnaire was given to management faculty members at both GWU and KSU in 2002

The questionnaire contained 51 features of their departments

Importance and performance scales, each ranging from 0 to 10

Evaluation

Range Mean Standard

Deviation

Importance (GWU)

4.80 7.5408 1.25207

Performance (GWU) 4.90 5.4890 1.18905

Importance (KSU) 6.00 7.3371 1.84934

Performance (KSU)

8.39 4.3529 2.49989

Dispersion in the responses

Coefficient of variation

Importance (GWU) 16.60%

Performance (GWU) 21.66%

Importance (KSU) 25.21%

Performance (KSU) 57.43%

Standardization of the importance and the performance scores

Range Min Max MeanStd.

Deviation

Importance Standardized (GWU)

3.84 3.35 7.19 6.0225 1.00

Performance Standardized (GWU)

4.12 2.73 6.85 4.6157 1.00

Importance Standardized (KSU)

3.25 2.16 5.41 3.9661 1.00

Performance Standardized (KSU)

3.36 0.20 3.56 1.7408 1.00

GWU QIPM

KSU QIPM

Ranking the Priorities

Standardized importance-performance ratio (SIP)

s

s

P

ISIP

Ranking GWU Priorities According to SIP Ratio

Rank GWU Priority Features SIP

1 Office security 1.977

2 Building/ physical environment 1.781

3Dept. organization to implement its strategic plan 1.756

4 Dept. strategic plan 1.729

5Help with writing research proposals 1.724

Ranking KSU Priorities According to SIP Ratio

Rank KSU Priority Features SIP

1 Funds to support research 24.197

2 Travel support 24.170

3 Office space for faculty 12.289

4 Projection equipment 9.387

5 Salaries 6.631

Clustering the Priorities

GWU Clusters Centers

Cluster

1 2 3 4 5

Importance Standardized (GWU) 7.15 4.92 5.83 4.3 4.22

Performance Standardized (GWU) 3.62 2.87 3.48 3.72 2.92

SIP 1.97 1.71 1.67 1.15 1.44

Clustering the Priorities

GWU Clusters Centers

Cluster

1 2 3 4 5

Importance Standardized (GWU) 7.15 4.92 5.83 4.22 4.3

Performance Standardized (GWU) 3.62 2.87 3.48 2.92 3.72

SIP 1.97 1.71 1.67 1.44 1.15

GWU Southeast Quadrant

KSU Clusters Centers Cluster

1 2 3 4 5 6 7

Importance Standardized (KSU)

4.79 5.29 4.89 4.24 4.90 4.60 4.87

Performance Standardized (KSU)

0.30 0.71 1.27 2.00 2.38 3.01 3.40

SIP 15.97 7.45 3.85 2.12 2.06 1.53 1.43

KSU Southeast Quadrant

Review of what we did (1) We used 2002 data from GWU and KSU We divided importance and

performance means by st. dev. in order to achieve a common level of agreement among GWU and KSU faculty members

Combining GWU and KSU data, we calculated the nearest whole integer mean for importance and performance

Review of what we did (2)

These means were used to create a common QIPM coordinate system

For each department the features in the SE quadrant were clustered by proximity

The clusters were ordered by average SIP, a measure of urgency

Conclusions (1) Standardizing importance and

performance scores to achieve a common level of agreement magnifies the differences between the two departments

At KSU the average importance of the features is lower than at GWU. This may mean that KSU is still struggling with basics such as salaries and office space. GWU has the luxury of concern with travel and research funds and the library collection

Conclusions (2)

Faculty members at KSU evaluate the performance of their department lower than do GWU faculty members

At KSU high priority features are mostly personal concerns such as salaries

At GWU high priority features are organizational issues such as planning