A Cluster Analysis of Office Operations

64 Pages·4.223 MB·English

by  Thomas, Leon Logan

PURDUE UNIVERSITY
THIS IS TO CERTIFY THAT THE THESIS PREPARED UNDER MY SUPERVISION
by____________Leon Logan Thomas
entitued A Cluster Analysis of Office Operations
COMPLIES WITH THE UNIVERSITY REGULATIONS ON GRADUATION THESES
AND IS APPROVED BY ME AS FULFILLING THIS PART OF THE REQUIREMENTS
FOR THE DEGREE OF
Doctor of Philosophy
/K/ /CjlJ
, Professor in Charge of Thesis
^ Hear of School or Departmekt
y / j F _________
TO THE LIBRARIAN:--
THIS THESIS IS NOT TO BE REGARDED AS CONFIDENTIAL.
pbofessos nr oharoe
grad, SCHOOL FORM 9—3-49—XM

A CLUSTER ANALYSIS OF OFFICE OPERATIONS
A Thesis
Submitted to the Faculty
of
Purdue University
ty
Leon Logan Thomas
In Partial Fulfillment of the
Requirements for the Degree
of
Doctor of Philosophy
August, 19^1
ProQuest Number: 27720641
All rights reserved
INFORMATION TO ALL USERS
The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.
uest
ProQuest 27720641
Published by ProQuest LLC (2019). Copyright of the Dissertation is held by the Author.
All rights reserved.
This work is protected against unauthorized copying under Title 17, United States Code
Microform Edition © ProQuest LLC.
ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 48106 - 1346
ACKNOWLEDGEMENTS
I wish to express hqt appreciation to the members of my advisory
committee. Professors Joseph Tiffin, E* J. McCormick, F* Je Shaw,
and W, V. Owen, for their interest and cooperation in this research#
I wish, also, to thank Miss Joan Prentice and Mr. R. R. Rathfelder
for their cosqputational assistance*
Finally, I acknowledge my indebtedness to Professor C. H. Lawshe,
chairman of my committee, whose research and teaching stand as guides
for the empirical approach to the problems of psychology#
ABSTRACT
Thomas, Leon Logan, A CLUSTER ANALYSIS OF OFFICE OPERATIONS, August 1951,
52 pages, 17 tables, 6 titles in the bibliography, appendix, A research
problem in industrial psychology which represents a new methodological
attack upon the components of work#
1# The purpose of this study was a) to identify clusters of similar
elemental operations in a sample of office jobs, and b) to develop
a modification of the Tryon Cluster Analysis technique for use
with a large number of variables#
2# Using completed Job Description Check-Lists of Office Operations
from 112 office jobs in five different companies, the inter-
correlations for 79 selected check-list items were computed#
3# A modification of the Tryon Cluster Analysis technique was
developed and used to identify eight clusters of office operations
which may be thought of as the work components of a sample of
office jobs#
Ue The clusters, or components of work, identified were: a) Typing,
b) Listing and Compilation, c) Communication, d) Planning and
Supervision, e) Filing, f) Stock Handling, g) Routine Clerical
Operations, and h) Calculation#
5* A procedural outline for performing a cluster analysis with a
large number of variables was written.
ii
TABLE OF CONTENTS
Page
ABSTRACT.................. *................* ü
INTRODUCTION .............. 1
PURPOSE........ *..... 1
PROCEDURE...... 2
The Check-List. ...... 2
The Sample.. . . . . . . . . 2
The Selection Criteria..................................... 2
Limitations of the Sample.• .... 3
The Selection of I t e m s 3
The Correlation M a t r i x . ..... 3
Construction of Tryon Curves...... ....... ........... h
Curve Inspection........................................... h
First Refinement. ...... ••• 5
Second Refinement. 6
Cluster Intercorrelations. .... 6
RESULTS ............ 10
Cluster I (Typing)...... 10
Cluster II (Listing and Compilation)....................... 11
Cluster III (Communication)..• ........ 11
Cluster IV (Planning and Supervision)...................... 11
Cluster V (Filing) ..... 11
Cluster VI (Stock Handling)........... 12
Cluster VII (Routine Clerical)............................. 12
iii
TABLE OF CONTENTS (continued)
Page
Cluster VIII (Calculation)* ....... 12
Residual Items. 12
Unselected Items. 13
PROCEDURAL OUTLINE FOR PERFORMING A CLUSTER ANALYSIS
WITH A LARGE NUMBER OFV ARIABLES............ 1$
SUMMARY....... 18
REFERENCES............ 19
APPENDIX A. JOB DESCRIPTION CHECK-LIST OF OFFICE OPERATIONS.. 20
APPENDIX B. CORRELATION MATRIX OF SELECTED ITEMS.............2 5
APPENDIX C. VIEWING BOX USED FOR CURVE INSPECTION... 3h
APPENDIX D. TABLES............. 35
LIST OF TABLES
Table Page
le Distribution of Cluster Item and Residual Item
Correlations After the First Refinement...*..*.•••••••*• 7
2* Distribution of Cluster Item and Residual Item
Correlations After the Second Refinement*•••••#•••••*••• 8
3* Matrix of Cluster Intercorrelation....... ..*..*...**** 9
lu Frequency of Occurrence and Rank of Check-List Items
In a Sample of 112 J o b s * . * . ....**** 36
5* Three Items Most Highly Correlated With Each Selected Item. 37
6. Cluster I ( T a p i n g ) . ................. 38
7* Cluster II (Listing & Compilation)................. ..... 39
8. Cluster III (Communication)*• • ♦ . hi
9* Cluster IV (Planning & Supervision). h2
10. Cluster V (Filing)...... W&
11. Cluster VI (Stock Handling)....... ..... U5
12. Cluster VII (Routine Clerical). . U6
13* Cluster VIII (Calculation)................................. 1&8
111. Correlation of Residual Items With Each of the Eight
Clusters After the First*Refinement.................. U9—
l$m Correlation of Residual Items With Each of the Eight
Clusters After the Second Refinement................ $0
l6* Matrix of Residual Item Intercorrelation............... $1
17* List of Residual Items After Second Refinement............. $2
r
A CLUSTER ANALYSIS OF OFFICE OPERATIONS
The identification of components of work is an integral step for the
investigator in the area of Industrial Psychology, One phase in the con­
struction of tests for selection and placement purposes is the determina­
tion of groups of similar operations, The development of a training
program depends, in part, upon the isolation of components of work that
tend to cut across job lines. Likewise, work components contributing to
the make-up, of a job family can serve as a foundation upon which a system
for rating job incumbents can be built. Investigators in these areas are
constantly seeking methods which will give them more information about the
fundamental nature of the job families under consideration#
Research by Lawshe and others (U> 5) suggests that factorial methods
can be successfully applied to job analysis data. They have shown that
factorial methods can be used to identify components of a more fundamental
nature when applied to job evaluation systems# Tryon (6) has developed a
technique intended to accomplish results similar to those of the factorial
methods in a somewhat less mathematically rigorous manner. The Tryon
technique seemed promising and a modification was developed for use with
this particular problem#
PURPOSE
This study, therefore, represents a new methodological attack upon
the identification of components of work* Specifically, the purpose is to:
(1) identify groups, or clusters, of similar elemental operations in a
1