The impacts of selected institutional characteristics on tuition pricing strategies

MISSING IMAGE

Material Information

Title:
The impacts of selected institutional characteristics on tuition pricing strategies
Physical Description:
x, 122 leaves : ; 28 cm.
Language:
English
Creator:
Hirté, John Richard, 1947-
Publication Date:

Subjects

Subjects / Keywords:
Universities and colleges -- Finance -- United States   ( lcsh )
College costs -- United States   ( lcsh )
Pricing   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1988.
Bibliography:
Includes bibliographical references.
Statement of Responsibility:
by John Richard Hirté.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001079351
notis - AFG4283
oclc - 19087255
System ID:
AA00003365:00001

Full Text














THE IMPACT OF
SELECTED INSTITUTIONAL CHARACTERISTICS
ON TUITION PRICING STRATEGIES
























By

JOHN RICHARD HIRTE


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY


UNIVERSITY OF FLORIDA


1988




















This study is dedicated to my wife, Pamela, who provided her

support, love, and encouragement throughout this lengthy process and to

Mrs. Janice Rothring, my administrative secretary, for her invaluable

professional expertise and skillful assistance.


















ACKNOWLEDGMENTS


The writer is indebted to Dr. James L. Wattenbarger, chairman

of this doctoral committee, for his guidance, encouragement, and

patience throughout all phases of the study. Similar indebtedness is

due Dr. Phillip A. Clark and Dr. John H. James for their assistance and

counsel.

Acknowledgment is also warranted for the chief business

officers of those private comprehensive colleges and universities

participating in the study and for those colleagues in higher education

who rendered invaluable support.

A final note of thanks is extended to the staff of the

university library, the university computer center, and many friends

and associates at Xavier University for their encouragement, support,

and professional expertise in the completion of this study.
















TABLE OF CONTENTS



PAGE

ACKNOWLEDGMENTS . . ... .iii

LIST OF TABLES. . . ... .vi

ABSTRACT. . . ... .... .viii

CHAPTER


BACKGROUND OF THE STUDY . .


Introduction .
Purpose and Null Hypothesis .
Delimitations and Limitations
Definition of Terms ....
Significance of the Study .
Overview of Methodology .
Overview of the Study .


of the Study.


REVIEW OF RELATED LITERATURE . .

Introduction . .
Tuition . . .
Pricing Policy Issues . .
The Pricing Decision . .
Pricing Strategies Identified .
Summary and Observations . .


THREE RESEARCH METHODOLOGY . . .


Introduction .
Selection of the Population .
Instrumentation .
Source of Data .
Variables Analyzed .. ..
Statistical Methods Employed.


TWO


















FOUR ANALYSIS OF THE DATA . .

Overview. . .
Survey Results. . .
Tests of the General Null Hypothesis..
Summary of Significant Differences. .

FIVE SUMMARY AND DISCUSSION . .

Introduction. . .
Discussion. . .
Implications for Future Research .


APPENDIX


ALPHABETICAL LIST OF PRIVATE COLLEGES AND
UNIVERSITIES USED IN STUDY .

LETTER MAILED TO THE CHIEF BUSINESS OFFICERS,
SURVEY INSTRUMENT, AND FOLLOW-UP LETTER.

RESPONSE SUMMARY FOR IMPORTANT
TUITION PRICING FACTORS. . .


4 SUMMARY OF STATISTICAL
RESULTS BY CATEGORY. .

REFERENCES . . .

BIOGRAPHICAL SKETCH . .


. 101


. 107


. 112

. 117

. 122


i j


. .
* .
. .
















LIST OF TABLES


TABLE PAGE

1. Comparison of Current Income--Private and
Public Comprehensive Colleges and Universities, 1983-84 14

2. Dominant Pricing Strategy as Identified by
the Respondents to Tuition Pricing
Assessment Survey . . ... 53

3. Dominant and Secondary Pricing Strategy
Identified by Tuition Pricing Assessment
Survey. . . ... ...... 53

4. Reported Enrollment Changes for Each
Pricing Strategy. . . 54

5. Impact of Increasing Tuition on Enrollment
for Each Pricing Strategy . ... 56

6. Ranking of Most Important Pricing Factors
as Reported in Tuition Assessment Survey. 58

7. Endowment-Related Variables for
Each Pricing Strategy . ... 61

8. Revenue-Related Variables for
Each Pricing Strategy . ... 66

9. Enrollment-Related Variables for
Each Pricing Strategy . ... 70

10. Faculty-Related Variables for
Each Pricing Strategy . .... 72

11. Expenditure-Related Variables
for Each Pricing Strategy . 74

12. External Factor Variables for
Each Pricing Strategy . ... 78
















LIST OF TABLES


TABLE


13. Private Market Share by Region . .

14. Private Market Share for
Each Pricing Strategy . .

15. Summary of Statistically Significant
Findings of Institutional Variables
for Each Pricing Strategy . .


PAGE
















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy




THE IMPACT OF SELECTED INSTITUTIONAL CHARACTERISTICS
ON TUITION PRICING STRATEGIES

By

John Richard Hirte

April 1988


Chair: James L. Wattenbarger
Major Department: Educational Leadership


The objective of this study was to identify the dominant

tuition pricing strategy utilized in private comprehensive colleges

and universities and to identify those institutional variables which

may influence the tuition pricing strategy followed.

The study methodology involved the distribution of a survey

questionnaire to 160 chief business officers of private comprehensive

colleges and universities located within the continental United

States. Completed surveys were returned by 115 (71.8%) respondents

with 4 responses deleted for reasons of inaccurate or inconsistent

data. Responses from the remaining 111 chief business officers were

used as the basis for this study. The primary purpose of the survey

was to determine the dominant pricing strategy followed, the secondary

pricing strategy used, to gather data on enrollment changes, level of


viii















tuition and to identify the most important factors used in the tuition

pricing decision.

Quantitative institutional data were secured from the National

Center for Higher Education Management Systems (NCHEMS) and merged

with the survey responses provided by the chief business officers.

These data consisting of 31 institutional variables were classified

into six major categories defined for purposes of this study as

endowment, revenue, enrollment, faculty, expenditure, and external

factors. The general null hypothesis stated that there were no

differences (p<.10) among the institutions when they were grouped by

institutional pricing strategy on any of the 31 selected institutional

variables for the four major pricing strategies identified.

Of the 111 institutions from which survey responses were

received, 47 (42.4%) followed the residual pricing policy, and 37

(33.3%) followed the peer pricing strategy. Proportional and

externally indexed pricing had 13 (11.7%) and 11 (9.9%) responses

respectively. The smallest category was differential pricing with

only 3 (2.7%) institutional business officials identifying this

strategy. During the 5-year period of 1980 to 1985, 91 (82%) of the

respondents indicated the same pricing strategy was used.

As a result of the study, it was concluded that 6 variables

out of 31 were different among institutions when they were grouped by

institutional pricing strategy. The most significant variables were
















related to endowment income as a percent of total education and

general revenues, endowment income per full-time equivalent student,

and the level of endowment per full-time equivalent student.

Also significant were the variables of the percentage change

in tuition for the period of 1972-73 to 1980-81, the 5-year change in

undergraduate enrollment, and private enrollment in the region in

which the institution was located (regional market share).

















CHAPTER ONE
BACKGROUND OF THE STUDY




Introduction


Pricing of a product in the marketplace involves the

interaction of six factors: supply, demand, price of related

products, degree of competition, quality of the commodity, and the

ability of the consumer to purchase the commodity. In the discipline

of economics these factors are treated within microeconomics as price

theory.

In a purely economic sense the consumers (students) will pay

a price (tuition) for a product (education, degree, etc.). According

to the National Center for Education Statistics (1984) tuition

accounted for 50-75% of the total institutional budget. In private

higher education tuition represents the most significant income item

in support of the institution.

The tuition price established by a college or university

policy body is complicated by a variety of factors such as the costs

of providing the services, mix of other income, perceived quality of

the institution, geographic location, market share, and competition

from other institutions. (Cohen, 1982)

Funk (1972) found that the price sensitivity of students has

become more apparent in recent years and that students now consider











price in their selection process for university attendance. Jackson

and Weathersby (1975) concluded from their review of empirical

studies that there was a negative relationship between tuition price

at a specific institution and the probability that a potential

student would attend. Nerlove (1972) found that tuition pricing was

complicated by state, federal, and institutional financial aid;

tuition price discounting through institutional merit or need-based

aid; and the growing spread of tuition price between private and

public universities.

In a recent review of pricing policies, Shaman and Zemsky

(1984) categorized pricing into six major segments: proportional cost

pricing, which sets tuition equal to some fixed proportion of the

cost of education; externally indexed pricing, which links tuition to

an index such as the Consumer Price Index; peer pricing, which sets

tuition in accordance with the competitive (peer) institutions; value

added pricing, which ties tuition to future benefits such as higher

lifetime earnings; residual pricing, which establishes tuition as the

amount required to balance the budget; and mandated pricing, which is

set by legislative act or public policy.

In a marketplace of shrinking enrollments, the Western

Interstate Commission for Higher Education (1984) estimated a

significant decline in high school graduates by 1992. They estimated

a decline of 600,000 students from a high of 2.9 million in 1982.

Given these projections, pricing will become much more critical,

especially for private comprehensive colleges and universities with

tuition revenues as the major revenue source. The problem for











university administrators will be how to set tuition so that the

institution can maintain its market share and provide continued

educational quality. (Eckstein, 1960)

In a study of public higher education price setting, Rusk and

Leslie (1978) examined 50 public institutions, one from each state.

Their results showed that the prices charged were affected by the

market area of the institution, the market share held by competitors,

the costs of public subsidies, the per capital income within the

state, and the quality of the institution.

These results, however, are inconclusive. The arena of the

private institution is different from the public because of

significant variations in the level of tuition paid by the student.

As tuition rises, the student consumer will evaluate his or her

choices more critically.

A review of the literature surrounding this pricing issue

reveals little information on the pricing strategies followed in

private comprehensive colleges and universities. The popular press

is replete each year with stories on the rising costs of private

higher education, and administrators explain increases in terms of

maintaining competitive faculty salaries, maintenance of quality, and

physical plant needs.





Purpose and Null Hypothesis



The purpose of this study was to identify the dominant

tuition pricing strategy utilized in private comprehensive colleges











and universities and to identify those institutional variables which

may influence the tuition pricing strategy followed. As defined

earlier, these pricing strategies were proportional cost pricing,

externally indexed pricing, peer pricing, value-added or differential

pricing, residual pricing, and mandated pricing. For purposes of

this study, mandated pricing was deleted because the population

investigated was private comprehensive colleges and universities not

subject to state financial regulation or allocation of funds.

The 31 institutional variables were broadly grouped into 6

categories consisting of endowment, revenue, enrollment, faculty,

expenditure, and external factors. These categories were identified

through the literature review as important factors considered by

policymakers in the establishment of tuition.

To state the problem another way: Are these institutional

variables different among reported institutional pricing strategies?

The investigation of this question was facilitated by the testing of

the following general null hypothesis: There are no differences

(p<.10) among the institutions when they are grouped by institutional

pricing strategy on any of the 31 selected institutional variables.





Delimitations and Limitations of the Study



The study was limited to private comprehensive colleges and

universities because of their dependence on tuition as the major

revenue source. These institutions were classified as comprehensive








5


colleges and universities using the National Center for Higher

Education Management Systems (NCHEMS) taxonomy developed by Mankowski

(1979). This taxonomy defines a comprehensive college or university

as one offering diverse post-baccalaureate programs, including first

professional, but not engaging in significant doctoral education.

The primary limitation of the study was the instrumentation

used to gather data. The survey instrument, designed to identify the

pricing strategy followed, was subject to the individual interpreta-

tion of the chief business officer. To overcome this inherent

problem, a precise definition was provided for each strategy and the

respondent was asked to identify the pricing strategy followed by the

institution.

The study was confined to those 160 institutions classified

as private comprehensive colleges and universities. The most recent

data reported for 1983-84 for the 31 institutional variables were

obtained from the National Center for Higher Education Management

Systems (1987). In cases of obvious data inconsistency or inaccu-

racy, the institution was deleted or the data corrected from indepen-

dent sources. Responses from 4 institutions were deleted for the

reasons stated above, leaving a study population of 111 institutions.











Definition of Terms



The following terms were used in this study:

Private comprehensive colleges and universities are those

institutions characterized by diverse post-baccalaureate programs,

including first professional, but not engaging in significant doctoral

education. (Mankowski, 1979)

Education and general expenditures (E&G) are expenditures made in

direct support of instruction, research, public service, student

services, institutional support, and plant operations.

Endowment funds are funds with respect to which donors or other

outside agencies have stipulated, as a condition of the gift

instrument, that the principal is to be maintained inviolate and in

perpetuity and invested for the purpose of producing present and future

income which may either be expended or added to principal (AICPA, 1975).

Full-time equivalent student (FTE) is a student equivalency factor

derived by dividing student credit hours by an average course load of

15-semester hours for undergraduates.

Institutional financial aid is financial aid funds provided

students from institutional education and general funds.

Instructional costs are those costs reported in audited financial

statements as the direct costs of instruction.

Regional market share, expressed as a percent, is computed by

dividing the number of students enrolled in private colleges and

universities by the total number of students, both public and private,

within the region. The composition of each region was defined by the

Digest of Education Statistics (1987).











Median family income is a statistical measure which represents the

midpoint of family income levels for each state as reported by the

United States Census Bureau (1980).

Tuition and fees include all tuition and fees assessed, net of

refunds, against students for education and general purposes.

Weighted average faculty salaries are salaries reported by

academic rank and weighted by the number of faculty within each rank.





Significance of the Study



As the costs borne by students become more significant, it is

important to assess the impact of various institutional variables as

determinants in setting tuition and selecting a tuition pricing

strategy. According to Harris (1970), MacKay-Smith (1985), and Ross

(1985) pricing increases are justified publicly in the context of

increasing costs, maintaining excellence, quality improvements, and

other rationales acceptable to the consumer public. The evaluation of

institutional variables implies that these factors do not change

significantly over a reasonable period of time; therefore institutional

priorities are reflected in these quantifiable measurements.

It was believed that the results of this study would advance

knowledge in the field by expanding research done in the public sector

by Rusk and Leslie (1978) and McKeown (1982) and identify those

institutional variables that may influence the setting of tuition

pricing in private comprehensive colleges and universities. In











addition, the results of this study could lead to the development of a

modified basis for determining tuition charges in private comprehensive

colleges and universities.





Overview of Methodology



A survey instrument was developed by the researcher and was

completed by the chief business officer identified at each institution.

The survey instrument was used to ascertain which pricing strategy was

followed through the forced selection of one of 5 different alterna-

tives. In addition, each participant was asked to identify other

factors considered in strategies setting the tuition price, to report

on enrollment trends, to assess the impact of rising tuition upon

enrollment, and to report the acceptance rate of new admissions. The

survey instrument is shown in Appendix 2.

The institutions identified as private comprehensive colleges

and universities were classified according to Mankowski (1979) and the

institutional data for the 31 variables analyzed were obtained from the

Higher Education General Information Survey (HEGIS) of the National

Center for Education Statistics (NCES) for the 1983-84 fiscal year.

For each college or university in the population used for this

study, the following variables were extracted from the 1983-84 HEGIS

report:

Endowment-related variables included endowment income as a

percent of total education and general revenue, endowment income per











full-time equivalent student, endowment per full-time equivalent

student, market value of endowment June 30, 1981, and endowment fund

balance as a percent of current fund expenditures.

Revenue-related variables included percentage change in tuition

for the period of 1972-73 through 1980-81, total education and general

revenues per full-time equivalent student, private gifts and grants per

full-time equivalent student, tuition and fees as a percent of total

education and general revenues, gifts and grants as a percent of total

education and general revenues, tuition and fees per full-time

equivalent student, and fall semester 1986 tuition as reported from the

survey instrument.

Enrollment-related variables included total headcount

enrollment, total full-time equivalent enrollment, and the ratio of

full-time to total headcount enrollment for the 1983-84 fiscal year.

The chief business officers reported on the survey instrument the

change in undergraduate enrollment over a 5-year period and their

assessment of the impact of tuition increases on enrollment.

Faculty-related variables included the number of tenured

9-month faculty, weighted average 9-month faculty salary, the number

of 9-month faculty, and the percent of tenured 9-month faculty for the

1983-84 fiscal year.

Expenditure-related variables included total education and

general expenses per full-time equivalent student, academic support

expenditures per full-time equivalent student, instructional costs per

full-time equivalent student, instruction as a percent of total

education and general expenditures, institutional student aid as a











percent of total education and general expenditures, student aid per

full-time equivalent student, and financial aid as a percent of

tuition.

Other external factors considered were regional market share,

1979 median family income for each state and enrolled students as a

percent of applications for each institution.

The method of analysis used in this study for interval scaled

data was the one-way analysis of variance (ANOVA) procedure followed by

the Duncan multiple range test. The null hypothesis was tested at the

p<.10 level. For nominal scaled data the chi-square test was used with

the null hypothesis evaluated at the p<.10 level. A full description

of the methodology used in this study is shown in Chapter Three.






Overview of the Study



In Chapter Two of this study the pertinent literature related

to tuition pricing in higher education is reviewed. A description of

the review of the literature regarding tuition pricing is included.

In Chapter Three is a description of the research methods

employed. Topics such as survey development, sampling procedures, and

statistical methods employed in analyzing the data are discussed.

Chapter Four contains the results of the study with reference

to the summary statistics, data analysis, statistical methods, and

testing of the general null hypothesis.

Chapter Five provides the summary and discussion from the

completed research and recommendations for future research.

















CHAPTER TWO
REVIEW OF RELATED LITERATURE




Introduction


Presented in this chapter are the results of the literature

review related to pricing of undergraduate tuition at private

comprehensive colleges and universities. The results of this review

are arranged in topical sections covering the scope of this study. The

review includes a description of the concept of tuition and related

definitions; a comparison of public and private universities with

regard to various selected characteristics; a review of the economics

of price; the pricing policy issues of product quality, pricing

objectives, and educational costs; and the descriptions of pricing

strategies identified in the literature review. Concluding this

chapter are summary and observations drawn from the review.






Tuition



Tuition is the price charged by officials of a college or

university to the student for attendance at the institution. This

price may or may not cover the actual cost of instruction, but in all

cases, this source of revenue represents a significant portion of











current income. Hopkins and Massy (1981) found that in institutions

where tuition is the dominant source of revenue, the setting of tuition

was one of the most important financial and economic decisions

undertaken during the annual budget cycle.

In a survey of 20 institutions, Saupe and Blagg (1977) reported

little consistency in the terminology used for tuition and fees. They

indicated fee structures varied considerably, and little, if any,

information existed on the design or rationale used to change or modify

fees. Dejnozka and Kapel (1982) defined tuition as

financial charges assessed students by an educational
institution (e.g. college, university, private
elementary or secondary school) to pay for instruction.
There is no universally established procedure used by
all educational institutions to determine tuition
costs. Some institutions charge tuition on the basis
of each credit hour or course for which the student is
registered. Others impose uniform charges for the
semester or year. Tuition for undergraduates is
frequently different (lower) than that charged graduate
students. Fees charged for a particular purpose (e.g.
laboratory fees) supplement rather than being a part of
tuition.

Tuition in private institutions is usually higher
than in public institutions, primarily because public
(tax) monies are used to support part (in some cases,
most) of the costs of instruction in public schools and
colleges. Tuition costs may change from year to year,
depending on such factors as nontuition support
received by educational institutions, changes in the
cost of living, and so on. In almost all instances,
tuition received from students defrays but a part of
the total instructional costs of an education. (p. 532)

Tuition represents the major source of revenue for the 160

private comprehensive colleges and universities in this study. Table 1

shows that during 1985-86, the National Center for Education Statistics

(1986) reported that in private comprehensive colleges and universities











tuition was 58.3% of current income. This compared with 19.3% for

similar public comprehensive colleges and universities.

Further analysis of Table 1 showed the distinction between

private and public comprehensive colleges and universities with regard

to other revenue sources. Private comprehensive colleges and

universities received only 2.1% of revenues from federal, state, and

local appropriations as compared to 53.1% for their public counter-

parts. As one might expect, this was the largest revenue source for

public universities in this category.

Both types of institutions received about the same percentage

of revenue from grants and contracts, and auxiliary operations. This

would indicate that this activity was very similar for both types of

institutions.

The major difference in private support for institutions was

evident in the comparison of private gifts and endowment income.

Private comprehensive colleges and universities significantly exceeded

their public counterparts in both private gifts (7.0% versus 1.7%) and

endowment income (5.1% versus 0.2%). Based upon the increased

contribution of both private gifts and endowment income to private

comprehensive colleges and universities, it was determined that these

variables should be identified as factors for consideration.

Private university leaders have found it increasingly difficult

to remain competitive in terms of tuition pricing. Brinkman (1982),

who analyzed a sample of 2,488 institutions over a 7-year period of

1972-73 to 1979-80, reached the following conclusions:

While the rate of tuition increase among private
institutions was greater than among public












Table 1


Comparison of Current Income--Private and Public Comprehensive
Colleges and Universities, 1983-84


Revenue Source:

Tuition and Fees

Appropriations: Federal, State, & Local

Grants & Contracts, Federal, State, & Local

Private Gifts, Grants & Contracts

Endowment Income

Auxiliary Operations and Sales & Service


TOTAL


Percent of


Percent of
Total Current Income

Public Private

19.3% 58.5%

53.1 2.1

6.4 5.8

1.7 7.0

0.2 5.1

19.3 21.5


100.0%


100.0%


reported.


Note: The number of institutions in each category was not

Source: National Center for Education Statistics (1986)











institutions, it was not out of step with price
movements in other areas of the economy. the rate
of tuition price increase in private institutions has
been generally consistent with the movement of the
consumer price index. (p. 9)

Other researchers have confirmed this trend since 1979-80.

Suttle (1983) explored in detail the rising costs of private education

in the context of a high-priced private institution. From his analysis

of Yale University, he concluded that the real cost of tuition grew at

an amount less than 1% over a 10-year period ending in 1982 after

adjusting for inflation in the general economy.

In an analysis of tuition trends and the tuition gap between

private and public universities, McPherson (1978) reached two important

conclusions. First, the ratio of private to public tuition over the

period of 1929 to 1974 increased only slightly from 3.6 to 4.2.

Second, the dollar difference in tuition as a fraction of disposable

personal income, adjusted for inflation, remained relatively constant.

Suttle also concluded, based upon a 25-year analysis of tuition

from 1950-51 to 1975-76, that the percent of median family income spent

on tuition and total student charges had remained remarkably constant

in a range of 31% to 35%.

In a comprehensive study of tuition and the costs of higher

education, Nerlove (1972) advocated that tuition should be a price in

the economic sense. That is, the price of tuition should allocate or

ration the supply of, and the demand for, the product called higher

education.

The tuition and fees charged by public institutions
are quite consciously set below any plausible estimate











of what it costs to educate a student; the same is true
of high quality liberal arts colleges as it is for the
major private and public universities. (p. 180)

Nerlove concluded that "subsidies provided public universi-

ties, through direct support, produced tuition below cost and have

depressed the level of tuition in private institutions" (p. 211). This

differential in tuition between public and private institutions created

"the effect of virtually strangling the private part, which is caught

between rising costs and its own inability to raise the prices it

charges for its product" (p. 212).

Based upon these studies it was decided to gather quantitative

data on the following variables: total education and general revenues,

private gifts and grants, and tuition and fees per full-time equivalent

student; tuition and fees as a percent of total education and general

revenues; private gifts and grants as a percent of total education and

general revenues and tuition reported for the 1986-87 academic year by

institution.

The conclusions drawn from these researchers depend upon the

assumptions used in setting tuition. The continuum runs from the pure

economic price allocating scarce resources to the adoption of a pricing

policy which takes into consideration the special needs of the higher

education marketplace. Gilmore and Suttle (1984) provided an excellent

summary of the many dimensions of the tuition pricing decision in

private universities. A discussion of these alternatives follows.















Pricing Policy Issues



When addressing the issue of pricing policy, the approach must

be from the aggregate to the specific. In the arena of higher educa-

tion there are two major sectors: public colleges and universities

supported by state and local governments, and private colleges and

universities supported by tuition and private philanthropy.

The National Commission on the Financing of Postsecondary

Education (1973) developed a series of recommendations for student

access through institutional accountability. Many of these concepts

were addressed specifically to the pricing of higher education. From a

comprehensive report (over 400 pages), Lawrence (1974) prepared a

summary highlighting the conclusions and recommendations of the

Commission. With regard to tuition price and private colleges and

universities Lawrence noted reasonable choice should be provided

"among those institutions regardless of tuition charged or family

income" (p. 2). He went on to say there must be institutional

diversity "if all reasonable needs of students and society are to be

met. A number of finance related trends. threaten diversity"

(p. 3). Strong evidence suggests a relationship between "financing and

excellence in postsecondary education" (p. 4).

Lawrence stated, "the relative availability or scarcity of

financial resources, regardless of number of sources, is probably the

most significant factor affecting institutional independence. and











funders may want to know how much institutions spend [including cost

per student] to achieve an objective and to what extent the objective

is achieved" (p. 5).

He concluded that "adequate financial resources should be

provided from state and local governments. federal government. .

students and their families. .. to the extent of their [financial]

ability. and from alumni, foundations, and corporations to ensure

high quality among both private and public institutions" (p. 6).

Breneman and Finn (1978) found that private institutions were

at a competitive disadvantage when compared on any measure related to

tuition pricing. Government, at the federal and state level, removes

part of this disadvantage through direct support to the institution or

the student. States and regional associations have adopted policies

with regard to tuition and fees. Zumeta and Green (1987) studied the

impact of state policies regarding access and choice. They concluded

that many states provided subsidies on a per student basis for

attendance at private colleges and universities.

McKeown (1982) found that policymakers in 22 states had

established criteria for charging tuition at public institutions

within their states. In all cases, there was a differential for

residency and, in many states, a differential based upon student level.

McKeown also observed that there was a trend, given traditional age

enrollment declines, to fit tuition pricing policies to the realities

of the marketplace. Lawrence (1974) concluded that private

universities have always considered the marketplace when setting

tuition.











In 1973, Mayhew summarized the Carnegie Commission report which

contained specific recommendations with regard to tuition pricing

having a significant impact on the private sector. The commission

recommended a gradual rise of public tuition levels to about one-third

of the education cost. This would have the effect of narrowing the

tuition gap and making private colleges and universities more

affordable and price competitive with their public counterparts.

The commission also recommended narrowing the tuition

differential between public and private institutions by increasing

public tuition, supporting private colleges and universities through

institutional aid, and providing state student aid programs. This

would bring parity to the current differentials existing in tuition

between the two sectors of higher education.

Moore (1978) analyzed data for the state of New York, where

many of the Carnegie recommendations had been adopted. He concluded

that by providing state institutional aid, student financial aid, and

public tuition increases, many, but not all, of the financial barriers

for attendance at a private college or university were removed. These

strategies were implemented by the state over a 4-year period from 1973

to 1977 and kept the private/public tuition ratio about constant, but

decreased the private/public enrollment ratio. This would imply that

financial access to the private sector in New York state improved

during this time period.

Halstead (1986) found that increased state aid in any form

would directly or indirectly make private tuition levels more

competitive with their public counterparts. In another study by Hearn











and Longanecker (1985), the enrollment effects of alternative

postsecondary pricing policies were explored. As they relate to

private higher education, the results confirmed earlier conclusions.

The authors identified two major schools of thought, somewhat

divergent, dealing with distributing public subsidies to the public

sector of American higher education.

The traditional approach results in the broad distribution of

public subsidies by maintaining low tuition. This approach tends to

subsidize all students, regardless of financial need. The rationale

for this approach is "the general belief that the returns to society

from a highly educated citizenry justify significant public

expenditures on education" (Hearn & Longanecker 1985, p. 459).

The targeted approach combines higher tuition and "more

generous financial aid aimed squarely at those with financial need"

(p. 459). This approach rests on two premises. Individuals and society

both benefit from education so both should pay for its cost and "public

expenditures for education should be provided so as to maximize the

difference between public returns and costs" (p. 459).

This would imply that leaders of private colleges and

universities, facing competition in the marketplace and faced with

substantial tuition differentials, would favor the targeted approach.

Concerns about raising tuition inducing enrollment declines have not

materialized. In fact, Hearn and Longanecker (1985) stated that on the

basis of existing research, there is strong evidence that responsible

movement toward targeted subsidation of postsecondary education should

not appreciably affect enrollment rates in the long run. (p. 476) They











concluded that "likewise some students attending public institutions

might choose to attend higher cost private institutions because of

decreased tuition differentials" (p. 476).

Krakower and Zammuto (1983) identified five environmental and

five institutional factors impacting enrollment in colleges and

universities. They concluded that there were three variables that had

an institutional impact on enrollment in private colleges and

universities. First, the previous year's enrollment was positively

associated with the current year. Second, federal student financial

aid was positively associated with enrollment. Third, tuition and fees

were negatively associated with enrollment.

It is clear that state and federal policies have a major impact

on tuition pricing in the private sector. According to Moore (1978)

when certain state policymakers adopted higher tuition and enhanced

financial aid, the private component became more competitive on a price

basis. Outside of these external factors the leaders of private

institutions must address their position within the private sector and

with their peer competition. In a study by Hoenack and Weiler (1975)

the relationship of cost-related tuition policies was explored in the

context of university enrollments. They concluded that there was not

enough empirical evidence to suggest that enrollments were impacted by

alternative tuition policies.

Based upon these prior studies, it was decided for purposes of

this study, to gather data on headcount enrollment, full-time

equivalent enrollment, the ratio to full-time enrollment, the change in

undergraduate enrollment over the preceding 5-year period, and an











assessment of the impact of increasing tuition on enrollment over the

same time period. These variables were categorized as enrollment

related.






The Pricing Decision



Economists deal with price on a purely theoretical basis

whereby price is the equilibrium between supply and demand. Kaysen

(1960) stated that

a well functioning price system operating under ideal
circumstances leads to the happy result that each kind
of goods and service is produced and consumed in such
quantity that. at the margin, what it is worth to
the consumers who pay for it is just equal to what it
costs the producers who supply it, and in turn, just
what it costs society, in terms of foregone alternative
uses of productive resources. (p. 55)

Because these economic ideals cannot be achieved unless a purely

competitive environment exists, other price strategies are often

followed. Mulvihill and Paranka (1967) summarized a comprehensive list

of issues dealing with the pricing of products in the marketplace.

Although not directly related to private comprehensive colleges and

universities, the concepts are important to review. Five fundamentals

were identified in addressing the pricing decision:

1. The consumer situation dealing with such issues as

utility to the buyer, the perceived quality of the product,

availability of substitute products, and relative or customary prices.

2. Cost considerations such as historical costs, costs of

production, capacity, volume, and contribution to overhead.











3. The competitive and market structures in place. Included

here are market position, share of market, leader or follower in the

market, competitive response and actions, and the market position of

substitutes.

4. The market structure and promotional policies which

include geographic area, advertising strategy, promotional programs,

and possibilities of more effective cooperation.

5. General economic considerations such as new product

introductions, expanded geographic area, and adjustments to the

cyclical nature of the economy.

Institutional leaders should consider these factors when

adopting an effective pricing strategy. Kotler (1975) discussed four

different pricing objectives which must be decided before a pricing

strategy is adopted. The adoption of one of these objectives, or a

combination, is important in developing a pricing strategy.

Business firms set their price to maximize their
profits. The theory of profit-maximizing pricing is
well worked out and widely understood requiring
knowledge of the demand and cost functions for the
product. Nonprofit organizations price according to
other standards, such as covering costs, matching
competition, subsidizing certain groups, and so on.
(p. 70)

Kotler continued

Prices are not only an indicator of the rate of
exchange but also communicators of a product's probable
quality. If a list of tuitions charged by different
colleges were shown to a group of people, they would
probably take the tuitions to be an indicator of
academic quality. (p. 71)

The pricing objectives proposed by Kotler were profit

maximization, cost recovery, market incentivization, and market

disincentivization. Each will be discussed briefly.











Profit maximization requires that leaders of an organization

estimate a demand function for a range of prices and a cost function

for a range of production to include fixed and variable costs. Total

profit would equal the difference between total revenues and total

costs. Difficulties of this approach include a tendency to maximize

short-run profits and problems associated with estimating the

appropriate demand and cost functions.

Cost recovery exists when revenues just cover costs or prices

are set to cover a portion of the cost. In many institutions,

including universities, the only objective is to cover operating costs

and, in practice, "prices charged by their organizations are largely

determined by tradition, public opinion, and actual or potential

competition rather than any scientific standard" (Kotler, 1975 p. 181).

The objective of market incentivization is to attract adoption

of a new service or program or to increase market share. Generally,

this is done through the mechanism of establishing a low price to

attract new customers. Market conditions of price sensitivity,

decreasing unit production costs, and strong competition would lend

themselves to the adoption of this objective.

Price setting with the objective of discouraging consumption

would be used under market disincentivization. This objective is

commonly used by governments to discourage consumption of tobacco

products and alcohol. Optimally, prices would be set at a level high

enough to achieve the objective desired.

In a quantitative analysis of the college choice process,

Chapman (1979) developed a statistical model in which he identified two











important factors. He stated that "of the marketing decision variables

at the control of a school, pricing policy seems to present both the

greatest problems and the greatest opportunities" (p. 40).

Chapman went on to state that "of all the marketing decision

variables, the pricing variable has a particularly significant role to

play for private colleges and universities" (p. 42). Financial aid

packaging to the student by the university allows the institution to

effectively discriminate tuition price.

The conclusions drawn from this study indicated that college or

university quality was a very important factor in the selection

process. In addition, those students from higher income families tend

to place an even stronger emphasis on quality.

The second major conclusion was that price "clearly has an

important role to play in the college choice process" (p. 54) and for

those students from lower income families, the kind and amount of

student financial aid is the single most important factor.

Tierney (1982) confirmed the importance of price to the student

as a factor in matriculation. In a study sample of 8,818 students

admitted to at least one public and one private institution, he

concluded that college quality and family income were two very

important aspects in the college choice process. The "student

assessment of an institution's value is predominately a function of

the quality of the institutional alternatives and the relative net

price of attending a particular institution" (p. 375).

In the area of consumer goods, the concept of price as a

quality indicator has received considerable attention. Cornell (1978),











in a review of literature, suggested that "in a world where it is

difficult for consumers to determine the attributes offered by

competing products, price may come to function as a signal of quality"

(p. 302). Although not directly related to private education, where

prices are generally higher, price does signal quality. In a recent

poll of college and university presidents (America's Best Colleges,

1985) many private colleges of perceived high quality were associated

with the highest annual costs of tuition. Monroe and Petroshius (1981)

found through their research that consumers associate higher prices

with higher levels of quality. Murphy (1984) observed that "in higher

education, this means that higher-status institutions can and should

charge higher tuition" (p. 84).





Pricing Strategies Identified



In developing the survey instrument to gather information on

pricing strategies, it was necessary to determine the extent of

available alternatives. This lead to the development of precise

definitions for tuition pricing strategies. From this review, the

top five pricing strategies, most appropriate for private comprehensive

colleges and universities, were used in the development of the survey

instrument.

Both private and public college and university policymakers

generally review tuition on an annual basis. In a study of setting

tuition at public universities, VanAlstyne (1977) identified five











major factors that needed to be considered in reaching the pricing

decision. These were the overall demand for education, tuition levels

at competing institutions, net revenues available to cover costs, cash

flow requirements, and the explicit student aid policy in place. He

went on to determine that "tuitions at public institutions are

frequently determined not on the basis of explicit state education

policy but as a residual calculation from the difference between budget

requirements and state appropriations" (p. 66). Two pricing strategies

were identified; residual pricing, in which tuition is set to balance the

budget, and peer pricing, which compares tuition to tuition charged at

competitive institutions.

Rusk and Leslie (1978) analyzed tuition pricing strategies

at major public universities by selecting one university from each

state. They examined five variables for their possible impact on

tuition prices at these selected institutions.

The first variable examined was the price charged by

competitors in the market area. This reflects the concept of peer

pricing. They concluded that this variable "clearly suggested. .

those responsible for establishing tuition are aware of tuition. .

levels in adjoining states. Further, tuitions. .. bear a strong

resemblance to tuitions at other in-state institutions, both public and

private" (p. 534).

The second variable was market share defined as the proportion

of enrollment of private and public universities in the state. Rusk

and Leslie concluded that in states where there were high enrollments

in private colleges and universities, relative to public colleges and











universities, tuition charges were significantly higher than the

average.

The third variable examined was college and university costs,

net of any public subsidies. The major cost identified was faculty

salaries. One might expect those institutions with high faculty

salaries would have the highest tuition but this was not the case due

to the many different techniques of funding state-supported higher

education. In those states with the highest support for public higher

education there were generally lower tuition levels.

The fourth variable was the relationship between per capital

state income and tuition charged. Two interesting trends emerged here.

First, in states with higher than average per capital income students

were charged higher than average tuition and, second, in states with

high student aid programs there were higher tuition levels.

The last variable examined was institutional quality. Rusk and

Leslie (1978) utilized a numerical rating system based upon the Gourman

Report, assigning a quality index to each institution. Higher tuition

was expected to be positively correlated to quality but the results

were inconclusive.

Rusk and Leslie (1978) identified two more pricing strategies

used by state universities. These were proportional pricing, which

sets tuition to cover a certain percentage of total current funds

revenues, and setting tuition to cover instructional costs. Although

their study dealt entirely with public universities, the results and

conclusions were important to this study of private universities.

These findings were consistent with those identified by Lamoureux











(1976) in an earlier study of pricing strategies used in marketing

continuing education in which the concept of marginal pricing was

discussed.

In a later study by Johnson (1979), an analysis was made of

differential tuition levels and instructional costs. He examined

institutions in six states that established tuition as a percentage of

instructional costs. The conclusions were interesting in that there

appeared to be no basis for setting tuition as a function of costs.

He concluded that in these states policymakers based tuition on

comparable levels in nearby states (peer pricing), high enough to

satisfy revenue requirements (residual pricing), and in line with

historical costs borne by the students (proportional pricing).

In the development of a planning model for Stanford University,

Hopkins and Massy (1981) identified tuition as the major revenue

source. In institutionally determining parameters for future

projections, they were very concerned with Stanford's tuition vis a vis

its competitors (peer pricing). In addition, projections for future

tuition increases were made at 4% above the consumer price index

(externally indexed pricing). They also recommended that certain

programs charge higher tuitions than others thereby introducing the

concept of differential pricing. This implies that institutional

leaders may consider more than one pricing strategy in determining new

tuition levels.

Ihlanfeldt (1981) concluded after a study of pricing policies

in private higher education that "in the independent sector, pricing

appears to be more of a residual policy matter than one related to











market forces. that is, historically, increases in tuition have been

determined after revenues from other sources have been forecasted in

relation to expected operating cost" (p. 100).

Other considerations by management in setting tuition price

include the potential enrollment decline associated with higher

tuition, the spread between public and private tuition, the "desired

amount of revenue necessary to balance the budget" (p. 115), and

stability of revenues over the long run. Ihlanfeldt also suggested

that differential pricing be considered in those programs where costs

are significantly higher or economic returns to the graduates are

clearly enhanced.

In another study of state higher education executive officers

Viehland, Kaufman, and Krauth (1982) examined the emerging trend in

public higher education to index tuition levels to the cost of

instruction. Both the Carnegie Commission on Higher Education (1973)

and the Committee for Economic Development (1973) advocated increasing

tuition to one-third and one-half of educational costs respectively.

Viehland et al. found "that 30 states do not have an

established policy for determining tuition. tuition is determined

in an ad hoc manner. best described as incremented pricing" (p.

335). In three other states policymakers had an established policy

for setting tuition and had "written and formally approved statements

of the factors to be considered in determining tuition levels, but no

specific formula was used" (p. 335).

In 17 states tuition was indexed to a specific measure. In one

state a peer institution review was used, and in another the Higher











Education Price Index was used. In all remaining states tuition was

indexed to the cost of instruction or what was defined in the first

chapter as proportional pricing strategy.

Viehland et al. indicated that the indexing of tuition had

wide appeal because it introduced some degree of rationality into the

process. Policymakers could routinizee the process, which appeals to

decision makers at all levels" (p. 336). Also, by linking tuition to

the costs of education, increases could be partially covered by student

sources and tuition viewed as another income source to the state.

Trends such as these in the public sector tend to influence pricing

decisions in the private sector. The survey instrument developed for

this study was used to determine the pricing strategy followed by

private comprehensive colleges and universities.

The last major pricing strategy identified through the

literature review was differential pricing in which tuition differed by

program to reflect differences in instructional costs or differences in

the perceived value of the program. Shaman and Zemsky (1984)

identified this as value added pricing.

A more sophisticated approach to price setting in higher

education was proposed by Troutt (1983) in which he used linear

programming as a tool to assist university administrators in solving

the tuition increase dilemma. The complexity of this method, and the

lack of practical application for colleges and universities, precluded

consideration of linear programming as a viable tuition pricing

strategy for this study. This approach, however, certainly could be

helpful in developing alternative scenarios for tuition pricing with

the objective of maximizing revenues.











Yanikoski and Wilson (1984) cited three major reasons for

traditional differences in pricing undergraduate education. First,

"certain programs simply cost more for a university to operate and a

portion of this higher cost should be passed on to students" (p. 738).

Second, in certain disciplines future income potential justifies a

higher investment cost on the part of the student. Third, higher

tuition in certain high demand or over-enrolled programs may serve

"as a signal of quality and selectivity" (p. 738).

Leaders of certain institutions may not want to consider

differential pricing because they "will not have sufficient program

diversity. .. [and] production costs of specific programs at these

institutions typically are difficult to assess because faculty and

facilities cross departmental boundaries so frequently" (p. 745).





Summary and Observations



The literature review suggested that the approach to the

tuition pricing decision and the determination of the factors to be

considered were complex activities. The purpose of this review was to

identify pricing strategies and their application to private

comprehensive colleges and universities. These results were used to

design the survey instrument identifying pricing strategies in use and

to ascertain if selected institutional variables would be different

among these strategies.











Tuition pricing remains one of the most pressing concerns of

the policymakers in higher education. With declining student

demographics and generally lower inflation, innovative approaches must

be adopted to achieve both institutional objectives and consumer

satisfaction. In addition, institutional leaders will have to

continually respond to changes in the structure of, and support for,

student financial aid. Institutional quality will continue to be

associated with price and those institutions with this reputation will

be in a more favorable position to deal with tuition pricing decisions.

















CHAPTER THREE
RESEARCH METHODOLOGY




Introduction


Presented in this chapter is the research methodology employed

by the researcher to gather data for the analysis of tuition pricing

strategies followed by private comprehensive colleges and

universities. Additionally, 31 institutional variables were

identified which may have an impact on the tuition pricing strategy

followed. There were three components to this process: First, a

survey instrument was designed to gather institutional responses.

Second, the quantitative institutional variables for the 1983-84

fiscal year were selected from the data base available from the

National Center for Higher Education Management Systems (1987).

Third, statistical techniques and data base management tools were

employed to merge the two sources of data and analyze the information.





Selection of the Population



After an extensive review of the literature, it was determined

that those institutions which met the criteria of high dependency on

tuition as the major revenue source were classified by Mankowski











(1979) as private comprehensive colleges and universities. This

taxonomy was based on the classification of institutions developed by

the Carnegie Foundation for the Advancement of Teaching (1976).

During 1983-84, 160 institutions were classified in this grouping and

it was determined that data would be gathered for the entire

population. Over time this population is relatively constant with

adjustments made for changes in the classification of institutions,

new institutions, or institutional closures. Four institutions were

deleted from the list due to their location outside of the continental

United States. A complete listing of the 156 institutions that

constituted the study population with the headcount enrollment for

fall 1983 is displayed in Appendix 1.





Instrumentation



The survey instrument was designed after a comprehensive

review of the literature indicated that pricing strategies employed in

private colleges and universities were not easily ascertained.

Research suggested little consistent application of any one pricing

technique.

The recommendations of Tull and Hawkins (1976) regarding an

approach to survey design were followed in this study. They

recommended that "questionnaire construction techniques focus on seven

areas: (1) primary considerations, (2) question content, (3) question











wording, (4) response format, (5) question sequence, (6) physical

characteristics of the questionnaire, and (7) pre-test" (pp. 241-242).

The initial survey instrument was designed with these seven

considerations in mind and was reviewed by 15 business officers from

institutions similar to those under investigation. These professionals

were asked to review the instrument and make appropriate recommenda-

tions for changes to eliminate any possible ambiguous definitions.

Based upon their suggestions, a revised instrument was developed with

eight questions that would gather the required institutional informa-

tion for determining pricing strategy and testing the general null

hypothesis.

The most current list of private comprehensive colleges and

universities was secured through the National Center for Higher

Education Management Systems from the 1983 HEGIS report. The name and

address of the chief business officer at each institution was obtained

from the Directory of Higher Education (1985). For this study, the

chief business officer was defined as the person having the

responsibility for the financial and budgetary affairs of the

institution.

A personalized cover letter was addressed to each chief

business officer with an explanation of the research and the survey

instrument. The survey instrument was individually addressed and

printed to improve both the quality and appearance of the document.

The surveys were mailed by first class mail with a stamped

self-addressed envelope to facilitate response. The follow-up survey

was printed on high quality blue bond paper to attract the attention











of the respondent. The survey instrument used in this study was

specifically designed to determine pricing strategy currently

followed and to gather additional information on enrollment, tuition

charged, admissions efforts, and other price setting considerations.

Participants in the survey response were assured institutional

anonymity and were offered a summary of the final survey results.

The survey instrument (Appendix 2) contained eight specific

questions which would provide the following information.

1. The dominant pricing strategy currently utilized by the

institution. The respondent was asked to select only one pricing

strategy from the definitions listed on the survey instrument.

2. The dominant strategy used 5 years prior. The respondent

was asked to select only one pricing strategy from the definitions

listed on the survey instrument.

3. The combination of pricing strategies utilized and other

decision factors considered in making the pricing decision.

4. Increase or decrease in enrollment over the 5-year period

1982 through 1986.

5. An assessment of the impact of increasing tuition on


enrollment by the chief business officer of the institution.

6. Identification of the five most important factors

considered by the chief business officer in setting tuition. A

standardized coding scheme was developed for this response and is

shown in Appendix 3.

7. Tuition data for the fall term 1986 which would update

information obtained from NCHEMS.











8. Percentage of enrolled students to the total number of

applications. This was gathered to develop a measure of institutional

selectivity which is associated with quality.

One hundred and sixty surveys were mailed to respondents and,

after written follow-up, 115 surveys were returned. Telephone

follow-up was carried out on 7 of the returned surveys to clarify

responses. The relatively high response rate of 71.8% was attributed

to interest in the topic and the researcher's offer to share the

results. The responses to each question were entered into the data

base for each university. Four institutional responses were deleted

for reasons of inaccurate or inconsistent data.

The survey results were tabulated using the crosstabs

subroutine available in the Statistical Package for the Social

Sciences (Nie et al. 1981). The objective of this analysis was to

determine the following information:

1. For the 111 institutions in the study, what was the

primary pricing strategy in place at the time of the survey in 1986?

For purposes of this study the differential and proportional pricing

strategy were combined due to the small number of differential

responses (3) and the inherent similarities of the two methods.

2. For the 111 institutions in the study, was the pricing

strategy identified in question one different from the pricing

strategy used 5 years prior?

3. Do the policymakers at the institutions in the study use

more than one pricing strategy and, if so, what was the relative

weight of each strategy in the determination of tuition?











4. Over the prior 5-year period did undergraduate enrollment

at the institutions in the study, increase or decrease?

5. What was the impact of increasing tuition on enrollment in

the opinion of the respondent? Data in this category was classified

as negative impact or positive impact.

6. The chief business officers' responses to the most

important factors considered in setting tuition were coded according

to the classification in Appendix 3 and ranked from highest to lowest

factor.

7. The 1986 tuition for each institution in the study was

collected to check the reasonableness of the reported tuition data on

the data tape obtained from the National Center for Higher Education

Management Systems (1987).

8. The percentage of new enrolled students to total

applications was used to determine a measure of institutional

selectivity which has been widely attributed as a measure of

institutional quality.





Source of Data



Each institution was identified by the Federal Interagency

Committee on Education Code. This is a unique 5-digit code assigned

by the United States Department of Education to specifically identify

an institution for federal reporting in the Higher Education General

Information Survey Report (1983).











A data tape was secured from the National Center for Higher

Education Management Systems information services for all 160 private

comprehensive colleges and universities in the group under study for

the 1983-84 fiscal year. Information was extracted on data for

endowment, revenue, enrollment, faculty, and expenditures. These data

were reported for each institution through the Higher Education

General Information Survey (1983). Data inconsistencies and obvious

errors were corrected and four institutions were deleted from the

study due to significant inconsistencies or inaccuracies. This

resulted in a study population of 111 private comprehensive colleges

and universities.

Data elements utilized for the purpose of this study were

from institutionally reported information provided through the annual

Higher Education General Information survey (1983). More specifically,

the information in the beginning of the next section was extracted

from the file for each institution. Based upon the review of the

literature, these selected variables were anticipated to be important

factors in the selection of a tuition pricing strategy.

Market share information for each institution in the study was

obtained from the Digest of Education Statistics (1987) for the fall

1985 reporting period. Regional market share, expressed as a percent,

was computed by dividing the number of students enrolled in private

colleges and universities by the total number of students, both public

and private, within the region.











The region of the country was identified according to the

taxonomy developed by the United States Office of Education for use in

the annual Higher Education General Information Survey. The regions

of the country were grouped into two segments. Those regions having

market share below 22.7% were defined as below average market share.

Those regions having market share above 22.7% were defined as above

average market share.

Data for median family income for 1979 was collected from the

United States Census Bureau (1980). This data provided the researcher

with median family income data for each state in which a private

college or university was located.



Variables Analyzed



Information was gathered and classified under the

following 31 variables for each of the 111 institutions comprising the

study population:


Endowment-Related Variables
Endowment income as percent of total E&G revenue
Endowment income per FTE student
Endowment per FTE student
Market value of endowment, June 30, 1981
Endowment fund balance as percent of current
fund expenditures

Revenue-Related Variables
Percent change in tuition 1972-1973 to 1980-1981
Total E&G revenues per FTE student
Private gifts & grants per FTE student
Tuition & fees as percent of total E&G revenue
Gifts & grants as percent of total E&G revenue
Tuition & fees per FTE student
Table of annual tuition costs











Enrollment-Related Variables
Total headcount enrollment
Total FTE enrollment
FTE as percent of total student enrollment
Change in undergraduate enrollment
Impact of tuition increase on enrollment
reported from the survey instrument

Faculty-Related Variables
Tenured 9-month faculty
Weighted average 9-month faculty salary
Number of 9-month faculty
Percent of 9-month faculty tenured

Expenditure-Related Variables
Total E&G expenses per FTE student
Academic support per FTE student
Instructional costs per FTE student
Instruction as a percent of total E&G expense
Instructional student aid as percent of total E&G expense
Student aid per FTE student
Financial aid as percent of tuition

External Factors
1979 median family income for each state
Enrolled students as percent of applicants
Regional market share


This data file was merged with the data from the 115

institutions responding to the survey to create a unique data record

for each institution in the study population. As noted, data from four

institutions could not be utilized due to inaccuracies or the inability

to match the Federal Interagency Committee on Education codes which

identified each college or university. Calculations were performed on

the remaining 111 institutions to determine pricing strategies and test

the general null hypothesis.











Statistical Methods Employed



Data analysis was facilitated using the University Computing

Center at Xavier University in Cincinnati, Ohio. The Statistical

Package for the Social Sciences (SPSS) release 9.0 version H (Nie et

al. 1981) installed on an IBM 4341 Model 2 was utilized for the

statistical analysis. This statistical package contains a system of

subprograms which allows the researcher a variety of descriptive and

analytical operations for data research. The program allows the user

to manipulate the data to develop frequency tables, recalculation of

data, and utilize a wide variety of statistical subroutines.





One-Way Analysis of Variance



The quantitative data for each institution were subjected to

the statistical technique of one-way analysis of variance (ANOVA) to

ascertain if these institutional variables were significantly different

within the study population. This test was used on all of the

variables listed in the prior section except the change in

undergraduate enrollment, the impact of tuition increase on enrollment,

and private enrollment in region (regional market share).

In this statistical procedure the relationship between a

dependent variable (institutional variable identified) and one or more

independent variables (pricing strategy) was explored. More

specifically, for each dependent variable identified in the six major











categories, the sample means were assumed to be equal when they were

grouped by the four possible pricing strategies. This was the general

null hypothesis used for the purpose of this study.

For each independent variable there was a probability

distribution. The means of each one of these distributions fall on the

regression curve which described the relationship between the

dependent and independent variables.

In this study, no assumptions were made with regard to the

statistical relationship between dependent and independent variables.

Neter and Wasserman (1974) stated that the "reason why analysis of

variance exists as a distinct statistical methodology is that the

structure of the independent indicator variables permits computational

implications which are explicitly recognized in the statistical

procedures for the analysis of variance" (p. 422).

This procedure allows the researcher to analyze a dependent

variable in relationship to each independent variable. For purposes of

this study, a single factor was identified and labeled as the pricing

strategy. The averages, or means, of each variable were compared for

each factor.

In analyzing the data, the following procedures were followed

as outlined by Mason (1982). For each pricing strategy, the following

general null hypothesis was stated: There are no differences (p<.10)

among the institutions when they are grouped by institutional pricing

strategy on any of the 31 selected institutional variables. For

example:


HO = U1 = U2 = u3 = U4











where ul, u2,... uj = the mean of the dependent variable under
analysis and HO is the null hypothesis.

If the null hypothesis is rejected, in other words, if there is

a statistical difference, the H1, alternative hypothesis, will be

accepted.

The level of significance selected for the one-way ANOVA

(F test) was <.10 which represented the probability of rejecting the

null hypothesis when, in fact, it was true. In this study, this would

mean that there was less than a 10% chance that the conclusions drawn

from the one-way analysis of variance was in error. The <.10 level of

significance was selected due to the nature of the population in the

study and the large sample response rate.

The F test is based upon the development of the F ratio which

is calculated for each variable as follows: (Mason, pp. 398-404)


estimated population variance based on
F = variation among the sample means
estimated population variable based on
variation within samples


where the numerator has K-1 degrees of freedom and the denominator has

N-K degrees of freedom where K equals number of pricing strategies and

N equals the number of observations for a particular variable.

The SPSS program calculates the F ratio for each variable under

consideration and compares this to the table value for the .10 level of

significance. It also computes the F probability, which corresponds to

the F ratio, which is used to accept or reject the null hypothesis at

the <.10 level of significance.











Therefore, in the evaluation of each variable, if the

calculated F probability was less than .10, the null hypothesis was

rejected in that instance, leading to the conclusion that there were

differences in the means of the variable under analysis that occurred

other than by chance. If the F probability was greater than .10, it was

assumed that, statistically, there were no differences in the means of

the variable under analysis, and the null hypothesis was accepted in

that instance.





Duncan Multiple Range Test



This test is a "systematic procedure for comparing all possible

pairs of group means. The groups are divided into homogeneous subsets,

where the difference in the means of any two groups in a subset is not

significant at some prescribed level" (Nie et al. 1981 p. 427). For

purposes of this study a significant level of .05 was used to evaluate

the difference in the means of any two subsets.

This procedure described by Winer (1971) provides a basis for

determining if there are differences in the means of the institutional

variables when grouped by institutional pricing strategy, and if there

are differences, what groups are different. The 28 interval scaled

variables were subjected to the Duncan multiple range test to ascertain

if any difference existed. This was done in conjunction with the

one-way ANOVA procedure.











Chi-Square Test

Data that were nominal scaled were analyzed using the chi-

square test. The variables subjected to the chi-square test were the

responses to survey question four, extent undergraduate enrollment

changed; survey question five, impact of increasing tuition on under-

graduate enrollment; and regional market share data used to classify

institutions into below average market share or above average market

share regions.

The purpose of the chi-square test according to Mason (1982)

"is to determine how well an observed set of data fits an expected set"

(p. 418). The null hypothesis states that "there will be no difference

between the set of observed frequencies and the set of expected

frequencies; that is, any difference can be attributed to sampling"

(p. 418).

The general formula for the chi-square test is


2 2
Ssum (fo f)

X fe

where fo is the observed value and fe is the expected value.

The decision rule for accepting or rejecting the null

hypothesis is formulated by referring to a chi-square table for the

p=.10 significance level with row minus one times column minus one

degrees of freedom.











The general limitation of the chi-square test states "that for

more than two cells, chi-square should not be applied if more than 20%

of the fe cells have frequencies less than five" (p. 423). This rule

was adopted for the analysis of the data in this study.

If the calculated chi-square value is greater than the table

value, the null hypothesis is rejected and the alternative hypothesis

accepted. The chi-square statistic was derived for the three variables

previously mentioned and the results are displayed in Chapter 4.

The survey results and statistical analysis of the data

including the one-way ANOVA, Duncan test, and chi-square test are

discussed in Chapter Four.

















CHAPTER FOUR
ANALYSIS OF THE DATA




Overview


The population of institutions used in this study were 160

private comprehensive universities as identified using the taxonomy

developed by Mankowski (1979). Of the 115 college and university

chief business officers responding to this survey, 4 of their

institutions could not be utilized for data coding reasons. The

remaining 111 institutions, therefore, were used as the basis for this

research.

As described in Chapter Three, the survey data were merged

with market share data, 1979 median family income data, and the

National Center for Higher Education Management Systems (1987) data

by institution to create a unique file record for each institution.

These data were subjected to the following statistical analyses.

First, the institutional data were tabulated to identify the

basic characteristics of the data such as frequency count by category,

means, and distributions. This was helpful in determining if there

were missing records, or incomplete data, for each institution.

Second, the interval scaled data which included endowment,

revenue, enrollment, faculty, and expenditure-related variables were

statistically analyzed using the one-way ANOVA technique, with a











follow-up procedure using the Duncan multiple range test. Included in

this category were data for median family income and admission ratios

for each institution.

Third, nominal scaled data collected from both the survey and

other sources were analyzed by using the chi-square procedure. The

data subjected to this analysis were the extent undergraduate

enrollment changed over the last 5 years (survey question 4), the

impact of tuition increases on undergraduate enrollment (survey

question 5), and private enrollment in the region in which the

institution was located (regional market share).

The variables under consideration were grouped in six major

categories based upon the logical relationship of the underlying data.

A summary of the variables and the assigned category was displayed in

Chapter Three and appears in Appendix 4.





Survey Results



The researcher, by the use of the survey instrument described

in Appendix 2, submitted eight questions to the chief business officer

of each institution. The purpose and design of the survey instrument

was discussed in Chapter Three.

The first three questions of the survey instrument were

designed to furnish the researcher with information on the pricing

strategy followed by the chief business officers of private

comprehensive colleges and universities. Table 2 provides a summary











of the responses. Of the 111 institutions from which responses were

received, 47 (42.4%) followed the residual pricing policy, and 37

(33.3%) followed the peer pricing strategy. Proportional and

externally indexed pricing had 13 (11.7%) and 11 (9.9%) responses

respectively. The smallest category was differential pricing with

only 3 (2.7%) institutional business officials identifying this

strategy.

These results confirmed findings from the literature review

which indicated that peer and residual pricing strategies were the two

most dominant strategies, that proportional and external pricing were

used to a lesser extent, and that differential pricing had only

limited acceptance.

During the 5-year period of 1980 to 1985, 91 (82%) of the

respondents indicated the same pricing strategy was used. Twenty

respondents indicated that proportional (6), external (2), peer (6)

and residual (6) had been used in the past. No institution identified

differential pricing as a strategy for this 5-year period.

If secondary pricing strategies were used, each respondent was

asked to distribute these strategies over the five categories totaling

100%. The results of this distribution are shown on Table 3. In

response to question 3 of the survey instrument the chief business

officers identified peer pricing as the most important secondary

component in setting tuition, although their primary strategy

identified was proportional, externally indexed, or residual.

In those institutions where proportional, externally indexed,

or residual pricing was reported as the dominant strategy, peer pricing











was weighted 12.3% to 19.4% as an important component of setting

tuition by the chief business officer. In institutions where peer

pricing was identified as the dominant strategy, the policymakers

weighted externally indexed pricing (13.8%) as a component in the

tuition pricing decision. Three respondents indicated differential

pricing as the dominant strategy and in those cases residual pricing

(16.3%) was also used.

Although not displayed in the table, in most institutions the

policymakers tended to use only two pricing strategies to develop the

pricing decision. Use of a third strategy in no case exceeded 7.0% of

the pricing decision for any dominant pricing strategy identified.

Question 4 on the survey instrument was intended to identify

the increase or decrease in undergraduate enrollment during the prior

5-year period. Table 4 shows the observed data reported compared to

the expected data. Respondents reported that in 57.8% of the insti-

tutions there was an increase in enrollment and in 42.2% of the

institutions there was no increase in enrollment. Data were

incomplete for two institutions. Only one expected value was less

than five (4.4) which indicated the chi-square test could be used to

test the null hypothesis.











Table 2

Dominant Pricing Strategy as Identified by the Respondents
to Tuition Pricing Assessment Survey



Pricing Strategy Number of Percentage
Identified Respondents of Respondents

Proportional 13 11.7

Externally Indexed 11 9.9

Peer 37 33.3

Differential 3 2.7

Residual 47 42.4



TOTAL 111 100.0


Table 3

Dominant and Secondary Pricing Strategy
Identified by Tuition Pricing Assessment Survey




Dominant Number of Secondary Strategy Weighted
Pricing Strategy Respondents Identified Percent

Proportional 13 peer 12.3

Externally indexed 11 peer 18.0

Peer 37 externally indexed 13.8

Differential 3 residual 16.3

Residual 47 peer 19.4



TOTAL 100


Note: Weighted percent is the average reported use of the secondary
pricing strategy.











Table 4

Reported Enrollment Changes for
Each Pricing Strategy


Change in Enrollment
1980 to 1985

Pricing Strategy Increased Decreased Total

Observed Expected Observed Expected


Proportional 11 8.5 4 6.5 15

Externally indexed 2 6.6 9 4.4 11

Peer 23 20.7 13 15.3 36

Residual 27 27.2 20 19.8 47


TOTAL 63 63.0 46 46.0 109

PERCENT 57.8 42.2 100.0











,2
These results, significant at the <.10 level, (3,N = 109) =

9.11, p<.10 lead the researcher to reject the null hypothesis that

changes in enrollment had no impact on pricing strategy reported.

Table 5 displays the chief business officers' responses to the

question of the impact of increasing tuition on enrollment, 44 (41.9%)

institutional respondents reported that rising tuition had a negative

impact on enrollment. The remaining 61 (58.1%) respondents reported

that there was no impact on enrollment associated with the increases

in tuition.

One expected cell value was less than five so the chi-square

test was applied to the data. The null hypothesis stated that there

were no differences (p<.10) in reported impact of increasing tuition

upon enrollment by institution when grouped by pricing strategy.


The results were not significant at the <.10 level,X2 (3, N

= 105) = 6.06, p>.10, indicating that there was not enough evidence to

suggest that the tuition impact on enrollment had any effect on the

selection of a pricing strategy.

Each institutional respondent was asked to identify the five

most important factors used in setting tuition for the next fiscal

year. A total of 475 responses was tabulated for this question.

Each survey was coded to reflect the scheme of classification as

displayed in Appendix 3. As expected, the five major pricing

strategies were identified during this process.

A ranking of all of the responses is shown in Table 6 with a

description of each classification. The peer pricing strategy was













Table 5

Impact of Increasing Tuition on Enrollment
for Each Pricing Strategy


Impact

Pricing Strategy Negative No negative Total

Observed Expected Observed Expected


Proportional 7 6.3 8 8.7 15

Externally indexed 7 4.3 3 5.7 10

Peer 10 14.6 25 20.4 35

Residual 20 18.8 25 26.2 45


TOTAL 44 44.0 61 61.0 105

PERCENT 41.9 58.1 100.0











identified by 94 (19.8%) institutional leaders as one of the five most

important factors in setting tuition. This confirmed earlier

reported findings of the significance of this strategy. The residual

strategy was identified by 63 (13.3%) of the respondents as one of the

five most important factors.

The use of an external index was identified by 11.5% of the

respondents as the most important factor in setting tuition. The

market reaction to change in tuition, the levels and availability of

financial aid, and the need for faculty and staff salaries were

reported from 7.5% through 9.5% of the total responses. Proportional

pricing, identified in the literature review as a pricing strategy,

was reported by only 24 (5.1%) of the respondents. Those chief business

officers using differential pricing identified this factor in only 13

(2.7%) responses.





Tests of the General Null Hypothesis



Six major categories of variables were identified as having a

potential impact on the pricing strategy followed by each institution

in the study population. These were endowment-related variables,

revenue-related variables, enrollment-related variables, faculty-

related variables, expenditure-related variables, and external

factors. A complete summary of each category with the associated

variables was provided in Chapter Three.











Table 6

Ranking of Most Important Pricing
Factors as Reported in Tuition Assessment Survey


Most Important Factor Number of Percent of
In Setting Tuition Responses Responses

Peer pricing strategy 94 19.7

Residual pricing 63 13.3

Use of an external index 55 11.5

Market reaction to change in tuition 45 9.5

Levels & availability of financial aid 44 9.3

Faculty and/or staff salary requirement 35 7.4

Proportional cost pricing 24 5.1

New programs/new equipment 21 4.4

Enrollment/economies of scale 19 4.0

Physical facility needs 17 3.6

External environment factors 13 2.7

Differential pricing 13 2.7

New/alternative tuition funding available 11 2.3

Trend in prior tuition increases 9 1.9

Quality/cost considerations 6 1.3

Cost containment 5 1.1

Collective bargaining arrangements 1 .2

TOTAL 475 100.0



Note: A complete description of factors is displayed in Appendix 3.











The reported pricing strategies were used to group the

institutions to determine if there were significant differences

(p<.10) among them in regard to the 31 institutional characteristics

(variables). For interval scaled data, the one-way analysis of

variance (ANOVA) procedure was used in conjunction with the Duncan

multiple range test. For nominal scaled data, the chi-square test was

utilized. A complete summary of the categories, the test statistic

used, statistical significance, and other summary data are provided in

Appendix 4. As has been noted, for purposes of this analysis, the

three institutions where the respondents identified differential

pricing as their dominant strategy were folded into the proportional

category. This was done because of the limited number of observations

and the logical similarity between these two pricing strategies.

For each category of variables, a summary table was prepared

to describe the results of the analysis displayed in rank order by

level of significance. In each case, the following general null

hypothesis was tested: There are no differences (p<.10) among the

institutions when they are grouped by institutional pricing strategy

on any of the selected 31 institutional variables.





Endowment-Related Variables



This category consisted of sets of data which measured the

endowment in terms of five different variables. Table 7 shows a

complete summary of the endowment-related variables, the F test values











and the reported means for each variable. Endowment income as a

percent of education and general revenue measured the relative

contribution of endowment revenue to total operating income. When the

institutions were grouped by the four possible pricing strategies, the

difference among the four groups with regard to endowment income as a

percent of education and general revenue was significant, F (110,111)

= 4.47, p<.10. The general null hypothesis, therefore, for this

variable was rejected.

Based upon the Duncan test there was not enough evidence in

the data to indicate that there was a significant difference (p<.05)

among external, residual, and peer strategies, or that there was a

difference among peer and proportional strategies. The mean endowment

income as a percent of education and general revenue was lowest

(2.4%) for those institutions reporting the external pricing

strategy and highest (7.6%) for those reporting the proportional

strategy.

External Residual Peer Proportional

2.4% 2.8% 6.7% 7.6%




The external and residual strategies differed significantly from the

proportional strategy.

The endowment income per full-time equivalent student was also

significant. When the institutions were grouped by the four possible

pricing strategies, the differences among the four groups with regard

to endowment income per full-time equivalent student was significant,

F (110,111) = 3.49, p<.10. The general null hypothesis, therefore,

for this variable was rejected.











Table 7

Endowment-Related Variables for
Each Pricing Strategy


Pricing Strategy


Variable F Propor-
Description Value tional External Peer Residual

Endowment income as %
of Total E&G revenue 4.47 7.6% 2.4% 6.7% 2.8%

Endowment income per
FTE student 3.49 $ 1,973 $ 234 $ 818 $ 303

Endowment per FTE student 2.30 $26,046 $ 3,381 $10,694 $ 7,368

Market value of endowment,
June 30, 1981 (thousands) 1.23 $33,040 $11,239 $33,247 $17,179

Endowment fund balance as %
of current fund .90 103.7% 35.6% 82.5% 66.9%
expenditures



Note: All statistics were calculated using the ANOVA procedure.
(110 degrees of freedom, sample size of 111, and significance
level of .10.)











Based upon the Duncan test there was not enough evidence in

the data to indicate that there was a significant difference (p<.05)

among external, residual, and peer pricing strategies. However, these

strategies differed significantly from the proportional strategy. The

mean endowment income contribution per student was $1,973 for

proportional and averaged from $234 through $818 for the other three

categories.

External Residual Peer Proportional

$234 $303 $818 $1,973




Endowment per full-time equivalent student measured the

relative wealth, or capitalization, of the institution. Generally,

one may conclude, based upon this sample, that those institutions

following the proportional pricing strategy were generally better

endowed than all others. When the institutions were grouped by the

four possible pricing strategies, the differences among the four

groups with regard to endowment per full-time equivalent student was

significant, F (110,111) = 2.30, p<.10. The general null

hypothesis, therefore, for this variable was rejected.

Based upon the Duncan test there was not enough evidence in

the data to indicate that there was a significant difference (p<.05)

among external, residual, and peer strategies, or that there was a

difference between peer and proportional. The mean endowment per FTE

student was $3,381 for those institutions reporting the external

pricing strategy and averaged from $7,368 through $26,046 for the

other three categories.











External Residual Peer Proportional

$3,381 $7,368 $10,694 $26,046




The external and residual strategies differed significantly from

the proportional strategy.

The average market value of endowment at June 30, 1981, was

not statistically different among the four pricing strategies. When

the institutions were grouped by the four possible pricing strategies,

the differences among the four groups with regard to market value of

endowment was not significant, F (110,111) = 1.23, p>.10. The

general null hypothesis, therefore, for this variable was accepted.

The last variable in this category was endowment fund balance

as a percent of total current funds expenditures. This was another

relative wealth measure and the researcher expected a statistical

significance. When the institutions were grouped by the four possible

pricing strategies, the difference among the four groups in regard to

the endowment fund balance as a percent of total current fund

expenditures was not significant, F (110,111) = .90, p>.10.

From the analysis of these five variables, endowment plays a

significant role in the pricing strategy followed. Throughout the

analysis, and based upon this population, those institutions following

the peer and proportional strategy were more dependent on endowment as

measured by the income variable and benefited from the higher

endowment per student and relative endowment size.











Revenue-Related Variables



There were seven revenue-related variables within this

category. These variables were the percentage change in tuition from

1972-73 to 1980-81, total education and general revenues per FTE

student, private gifts and grants per FTE student, tuition and fees as

a percent of total education and general revenues, gifts and grants as

a percent of total education and general revenue, tuition and fees per

FTE student, and the 1986 institutional annual tuition costs. Table 8

shows a complete summary of the revenue-related variables, the F test

values, and the reported means for each variable.

The percentage change in tuition for the 9-year period

1972-73 through 1980-81 indicated a significant difference among

institutions. When the institutions were grouped by the four possible

pricing strategies, the differences among the four groups with regard

to the percentage increase in tuition for the 9-year period 1972-73

through 1980-81 was significant, F (110,111) = 2.38, p<.10. The null

hypothesis, therefore, for this variable was rejected.

Based upon the Duncan test there was not enough evidence in

the data to indicate that there was a significant difference (2<.05)

in residual, peer, and external pricing strategies, or that there was

a difference in external and proportional strategies.

Residual Peer External Proportional

140.0% 147.9% 152.0% 190.0%











The residual and peer strategies differed significantly from the

proportional strategy.

Differences in total education and general revenues per

full-time equivalent student were slightly above the threshold of .10

level of significance. When the institutions were grouped by the four

possible pricing strategies, the difference among the four groups with

regard to education and general revenues per FTE student was not

significant, F (110,111) = 2.1, p>.10. The null hypothesis,

therefore, for this variable was accepted.

Since the results of the F test were very close to the

significance level of .10, the Duncan test indicated that there was

not enough evidence in the data to indicate a significant difference

(p<.05) among residual, external, and peer pricing strategies, or that

there was a significant difference among external, peer, and propor-

tional strategies.

Residual External Peer Proportional

$8,911 $9,640 $10,516 $18,184




The residual strategy differed significantly from the proportional

strategy. Proportional had average revenues of $18,184 per student

while residual had a mean of $8,911 per student. Given this range,

one may conclude that there was a difference between the relative

level of revenues for these two categories.

Private gifts and grants per full-time equivalent student

showed a large range of values, most notably between proportional and

residual. When the institutions were grouped by the four pricing











Table 8

Revenue-Related Variables
for Each Pricing Strategy


Pricing Strategy


Variable F Propor-
Description Value tional External Peer Residual

% Change in tuition 72-73
through 80-81 2.38 190.0% 152.0% 147.9% 140.0%

Total E&G revenues per
FTE student 2.10 $18,184 $9,640 $10,516 $8,911

Private gifts & grants
per FTE student 2.03 $ 6,656 $ 709 $ 1,059 $ 538

Tuition & fees as % of
total E&G revenue 1.90 53.5% 59.2% 56.2% 61.2%

Gifts & grants as % of
total E&G revenue 1.72 10.0% 7.1% 9.0% 5.9%

Tuition & fees per FTE
student .31 $ 5,381 $5,381 $ 5,647 $5,275

Annual tuition costs
fall 1986 .12 $ 6,492 $6,745 $ 6,740 $6,791


Note: All statistics were calculated using the ANOVA procedure.


(110 degrees of freedom, sample size
level of .10).


of 111, and significance











strategies, the differences among the four groups with regard to

private gifts and grants per FTE student was not significant, F

(110,111) = 2.03, P>.10.

Since the results of the F test were very close to the

significance level of .10 the Duncan test indicated that there was not

enough evidence in the data to indicate that there was a difference in

residual, external, and peer pricing strategies, or that there was a

difference in external, peer, and proportional strategies.

Residual External Peer Proportional

$538 $709 $1,059 $6,656




There was a difference between the residual and proportional pricing

strategy. This confirmed the results of the previous variables because

this variable would be included in the total education and general

revenues. It appeared that those institutions following proportional

pricing had significant outside income sources beyond tuition. This

was supported by the variable of tuition and fees per student and the

variable annual tuition costs. These variables indicated no

significant differences in the amount of tuition charged to the

students in both relative and absolute measures.

The variables of tuition and fees, F (110,111) = 1.90, p>.10,

and private gifts and grants, F (110,111) = 1.72, p>.10, as a percent

of total education and general revenues, showed no significant differ-

ences among the mean values. This indicated that in all pricing

categories tuition and fees provided about the same relative share of

total income.











There were no statistical differences in the tuition and fees

per full-time equivalent student, F (110,111) = .31, p>.10, and the

annual tuition costs effective fall 1986, F (110,111) = .12, p>.10.

This was an interesting result in that one might have anticipated

differences due to the pricing strategy in place. This did indicate,

however, that regardless of pricing strategy followed, the average

tuition charged was, within this group of institutions, approximately

the same. These small differences in average tuition and fees and

the annual tuition costs for fall 1986 support the premise that

tuition pricing is critical for private comprehensive universities

from a competitive perspective. It was interesting to note here that

these institutions under analysis had tuition revenues as a percent of

total revenues approximating the same as nationally reported data.

For the comparative period the gifts and grants as a percent

of total revenues were in excess of nationally reported data by the

National Center for Education Statistics (1986). Since the composi-

tion of institutions making up the reported population was unknown and

access to the underlying data was limited, no statistical conclusions

could be drawn. Table 1 displayed the aggregate financial revenue

summary for private and public comprehensive universities for the

1983-84 reporting period.





Enrollment-Related Variables



This category contained three interval variables related to

enrollment. The one-way ANOVA procedure was used to determine if

there were any statistical differences among pricing strategy for the








69


variables of total full-time equivalent enrollment, total headcount

enrollment, and full-time equivalent enrollment as a percentage of

total headcount enrollment. Table 9 provides a summary of the

enrollment-related variables with the F values and means of the

reported data.

The general null hypothesis stated that there are no

differences (p<.10) among the institutions when they are grouped by

institutional pricing strategy on any of the 31 selected institutional

variables.

For all of the variables within this category the results were

not significant. When the institutions were grouped by the four

possible pricing strategies, the difference among the four groups with

regard to total head enrollment was not significant, F (110,111)

1.99, p>.10. The same results were found for total FTE enrollment, F

(110,111) = 1.80, p>.10 and for the ratio of FTE to total headcount

enrollment, F (110,111) = .16, p>.10.

Based upon the evaluation of all the variables within this

category, it appears that enrollment-related factors had little or

nothing to do with the pricing strategy followed. The research

expectations were that differences would be shown in this category.

In an earlier study by Chisholm and Cohen (1982), it was found that

levels of enrollment were subject to variation based upon the complex

interaction of tuition price, institutional quality, and the mix of

part-time and full-time students.











Table 9

Enrollment-Related Variables
for Each Pricing Strategy


Pricing Strategy


Variable F Propor-
Description Value tional External Peer Residual

Total headcount enrollment 1.99 3,706 4,590 3,276 4,563

Total FTE enrollment 1.80 2,921 3,830 2,703 3,441

FTE as % of total student
enrollment .16 83.0% 84.4% 84.3% 81.8%


Note: All statistics were calculated using the ANOVA procedure.
(110 degrees of freedom, sample size of 111, and significance
level of .10)











Faculty-Related Variables



Faculty-related variables included measures of the number of

tenured 9-month faculty, weighted average 9-month faculty salary,

number of 9-month faculty, and the tenure ratio for each institution.

Based upon the literature review and the justifications used for

increasing tuition, it was anticipated that differences might be found

in these variables by pricing strategy. Table 10 provides a summary

of the faculty-related variables, F values, and the means for each

variable. The null hypothesis stated that there were no differences

in the means of these reported data at the .10 level of significance.

When the institutions were grouped by the four possible pricing

strategies, the difference among the four groups with regard to the

number of tenured 9-month faculty was not significant, F (110,111)

= .46, p>.10.

Based upon the same analysis of weighted average 9-month

salary the results were also found not significant, F (110,111)

= .40, p>.10. Neither the number of 9-month faculty, F (110,111)

= .37, p>.10, nor the percentage of tenured 9-month faculty, F

(110,111) = .20, D>.10, were statistically significant.

The tenure percentages and overall salary averages were within

a tight range for all pricing strategy groups of private comprehensive

universities. The range of highest to lowest salaries was about

$2,000 reflecting the general parity of salaries within this institu-

tional group. This researcher did not utilize total compensation,

which includes benefit costs, due to the difficulty in securing this

information by institution.











Table 10

Faculty-Related Variables
for Each Pricing Strategy


Pricing Strategy


Variable F Propor-
Description Value tional External Peer Residual

Number of tenured
9-month faculty .46 85.6 87.2 77.4 94.0

Weighted average
9-month faculty salary .40 $21,552 $21,564 $23,570 $22,900

Number of 9-month
faculty .37 135.6 143.5 132.5 154.2

% of 9-month faculty
tenured .20 53.8% 48.1% 53.7% 54.0%


Note: All statistics were calculated using the ANOVA procedure.
(110 degrees of freedom, sample size of 111, and significance
level of .10)











Expenditure-Related Variables



There were seven expenditure variables analyzed within this

category. These variables measured the total education and general

expenditures per FTE student, academic support per FTE student,

instructional costs per FTE student, instructional costs as a percent

of total education and general expenditures, institutional student aid

as a percent of total education and general expenditures, student

financial aid per FTE student, and financial aid as a percent of

tuition. Table 11 shows a summary of the expenditure-related

variables, F values, and the mean data reported for each variable.

When the institutions were grouped by the four possible

pricing strategies, the differences among the four groups with regard

to education and general expenditures per FTE student were not

significant, F (110,111) = 2.11, p>.10. The null hypothesis,

therefore, for this variable was accepted.

Based upon the Duncan test there was not enough evidence in

the data to indicate that there was a significant difference in

residual, external and peer strategies or that there was a significant

difference (p<.05) in external, peer, and proportional strategies.

Residual External Peer Proportional

$7,179 $7,591 $8,607 $15,706




There was a significant difference between residual and proportional

pricing strategies.











Table 11

Expenditure-Related Variables for
Each Pricing Strategy


Pricing Strategy


Variable F Propor-
Description Value tional External Peer Residual

Total E&G expenses per
FTE student 2.11 $15,706 $7,591 $8,607 $7,179

Academic support per
FTE student 1.93 $ 1,928 $ 539 $ 732 $ 607

Instructional cost per
FTE student 1.91 $ 4,155 $3,230 $3,123 $2,799

Instruction as % of
total E&G expenditures 1.91 37.4% 41.8% 36.4% 38.1%

Institutional student aid
as % of total E&G
expenditures .81 12.0% 14.1% 13.1% 13.9%

Student aid per FTE student .26 $ 661 $ 690 $ 727 $ 651

Financial aid as % of
tuition .19 13.1% 13.2% 14.2% 12.6%


Note: All statistics were calculated using the ANOVA procedure.


(110 degrees of freedom, sample size
level of .10)


of 111, and significance











When the institutions were grouped by the four possible

pricing strategies, the differences among the four groups with regard

to academic support per FTE student were not significant F (110,111)

= 1.93. D>.10. Instruction costs per FTE student were not

significant F (110,111) = 1.91, p>.10, but the Duncan test

indicated differences in two subsets. There was not enough evidence

in the data to indicate that there was a significant difference (p<.05)

in residual, peer, and external pricing strategies, or that there was

a difference in peer, external, and proportional strategies.

Residual External Peer Proportional

$2,799 $3,123 $3,230 $4,155




There was a significant difference between the residual and

proportional pricing strategies.

Instruction costs as a percent of total education and general

expenditures did not show any statistical difference, F (110,111) =

1.91, p>.10. The range of means was from 37.4% for proportional

through 41.8% for the external category. The overall percentage

commitment to instruction was 37.8% which approximates other reported

national data for these institutions. One may conclude that, across

pricing strategies, these institutions made approximately the same

commitment to both academic support and instructional costs.

Based upon the Duncan test there was not enough evidence in

the data to indicate that there was a significant difference (p<.05)

in peer, proportional, and residual pricing strategies, or that there

was a difference in proportional, residual, and external pricing


strategies.











Peer Proportional Residual External

36.4% 37.4% 38.1% 41.8%




There was a significant difference between peer and external pricing

strategies.

The three variables related to institutionally-based financial

aid were student aid as a percent of education and general expendi-

tures, student aid per full-time equivalent student, and financial aid

as a percent of tuition. These variables measured the relative

institutional commitment to financial aid and the relative discounting

of tuition revenues.

When the institutions were grouped by the four possible

pricing strategies, the difference among the four groups with regard

to these three financial aid measure were not significant. No

statistical difference was observed in institutional financial aid as

a percent of total education and general expenditures, F (110,111)

= .81, p>.10; student financial aid per FTE student, F (110,111)

= .26,p>.10; and financial aid as a percent of tuition, F (110,111)

= .19, p>.10.

No statistical differences were observed in these three

variables among each pricing category, indicating that the

institutions were making very similar financial aid commitments and

relative level of expenditures. This is confirmed by the small range

of means for these variables as depicted in Table 11.

The results of the evaluation of the expenditure-related

variables indicate that these variables do not appear to have any

significant differences among pricing strategy. Although, 16








77


respondents (14.4%) indicated either a proportional cost or

differential cost pricing strategy, the results did not indicate a

significant difference in the underlying major cost components. This,

however, should not lead to the conclusion that costs are not

considered during the process of setting tuition.





External Factors



The last category considered variables external to the

institution. Three variables were evaluated as part of this

component. These were regional market share, 1979 median family

income for each state, and enrolled students as a percent of total

applications. Shown in Table 12 are the data related to median family

income and enrolled students as a percent of total applicants.

Median family income for 1979 was not statistically different

when the institutions were grouped by the four possible pricing

strategies, F (110,111) = 1.87, p>.10. It was interesting to note,

however, that the externally indexed pricing strategy had the highest

level of median family income. Intuitively, this is correct because

cost of living and other indices are used as a basis for adjusting

tuition, and median family income is adjusted accordingly.

There was no statistical difference in the percentage of

enrolled students to total applications, F (110,111) = .28, p>.10.

This would imply that, on average, these institutions were

approximately the same in the selectivity of students. This does not

appear to have had any influence on the pricing strategy selected.











Table 12


External Factor Variables
for Each Pricing Strategy


Pricing Strategy


Variable F Propor-
Description Value tional External Peer Residual


1979 median family income
by state 1.87 $20,189 $21,528 $20,512 $20,247

Enrolled students as %
of applicants .28 35.7% 36.9% 37.6% 34.7%



Note: All variables were evaluated using the ANOVA procedure.
(110 degrees of freedom, sample size of 111, and significance
level of .10)











For purposes of evaluating the importance of market share,

which was defined for this study as the percentage of private

enrollment within the region, the chi-square procedure was used. The

null hypothesis stated that there were no differences in market share

when institutions are grouped by pricing strategy (p<.10).

The review of the literature identified market share as an

important factor in setting tuition. The United States Department of

Education divides the United States into eight regions. Based upon

data secured from the Digest of Education Statistics (1987), the

regional market share was computed by dividing the number of students

enrolled in private colleges and universities by the total number of

students, both public and private, within the region.

Table 13 shows the computed market share percentage for each

of the eight regions. The results indicated that the northeast and

mid east dominate the relative levels of private education with market

shares of 49.3% and 38.3% respectively. The great lakes, plains,

southeast, and rocky mountain areas all have approximately 18% and 22%

private enrollment. The far west and southwest have the smallest

market share with 12.3% and 10.8% respectively. For purposes of this

study, it was decided to collapse these eight regions into two major

categories based upon the logical distribution and the natural breaks

in the data. They were defined as higher than average market share,

and lower than average market share.

Table 14 shows the results of market share data by pricing

strategy reported. There were no expected cell values less than five.

The null hypothesis stated that there are no differences in market

share by institution when grouped by tuition pricing strategy (p<.10)











Table 13

Private Market Share by Region


Region Private Market Share Classification

New England 49.3%
Mid East 38.3% Higher than Average

Plains 22.3%
Great Lakes 20.7%
Rocky Mountains 18.1% Lower than Average
Southeast 17.9%
Far West 12.3%
Southwest 10.8%

Overall 22.7%



Table 14

Private Market Share
for Each Pricing Strategy



Market Share

Higher than Lower than
Pricing
Strategy Average Average Total
Observed Expected Observed Expected


Proportional 9 7.5 7 8.5 16

Externally indexed 4 5.2 7 5.8 11

Peer 12 17.3 25 19.7 37

Residual 27 22.0 20 25.0 47


TOTAL 52 52.0 59 59.0 111

PERCENT 46.8 53.2 100.0











The results of the chi-square test were significant at the p<.10

level, 2(3, N = 105) = 6.26, p<.10. These results indicated the null

hypothesis should be rejected and the alternative hypothesis accepted

that there is a difference in market share when the institutions are

grouped by tuition pricing strategy.

A clear distinction emerged between peer and residual pricing

strategies. Over half of those institutions adopting the residual

strategy were in areas of higher than average market share while peer

institutions had the most reported in the lower than average market

share regions. One might assume that in those institutions where the

policymakers used the externally indexed strategy there was not as much

concern with the relative price of private education within the

region. Those institutions reporting a proportional strategy tended

to be in areas of higher than average market share.

Specifically, 27 out of 52 respondents (52%) indicated that

their institutions fall in higher than average market share regions

and follow the residual pricing strategy. This implies less emphasis

or concern for the pricing among competitors. Also, the costs of the

total institutional operations were considered in the pricing

decision.

Twenty-five out of 59 respondents (43%) following the peer

pricing strategy fell in the lower than average market share regions.

This implied an increased sensitivity to price of tuition when the

regional market share was low or when a greater number of public

institutional alternatives existed. This confirmed other reported











information that region of the country tended to have an impact on

both the level of tuition and type of strategy followed.

These findings confirm earlier results by Rusk and Leslie

(1978) regarding the impact of geographic location on the

establishment of tuition price. Although this study pertained to

pricing strategies as contrasted to the actual amount of tuition, it

was an important finding that market share was a significant factor

considered in the tuition pricing decision.





Summary of Significant Differences



A summary of the major findings of the study determined to be

statistically significant is displayed on Table 15. A discussion of

these findings, as well as a review of other items not found to be

significant, is included in Chapter Five. A complete summary of all

variables analyzed is shown in Appendix 4.











Table 15

Summary of Statistically Significant Findings of Institutional
Variables for Each Pricing Strategy


Variable F
Description Value Significance

Endowment income as % of
total E&G revenue 4.47 .01

Endowment income per FTE
student 3.49 .02

% change in tuition 1972-73
through 1980-81 2.38 .07

Endowment per FTE student 2.30 .08


Chi-square
Value Significance

5-year change in undergraduate
enrollment 9.11 .02

Regional market share 6.26 .09

















CHAPTER FIVE
SUMMARY AND DISCUSSION




Introduction


The purpose of this study was to identify the dominant

pricing strategy utilized by policymakers in private comprehensive

college and universities and to identify those institutional

variables which may influence the tuition pricing strategy followed.

The 31 institutional variables were grouped into six categories

defined as endowment, revenue, enrollment, faculty, expenditure, and

external factors. The general null hypothesis used to evaluate these

categories stated that there were no differences (p<.10) among the

institutions when they were grouped by institutional pricing strategy

on any of the 31 institutional variables.

From the review of the literature, covering predominantly the

period of 1965 to the present, five dominant pricing strategies in

use were identified. These were proportional cost pricing, externally

indexed pricing, peer pricing, differential pricing and residual

pricing. For purposes of this study, the differential pricing group

was combined with the proportional group due to the small sample size

and the underlying similarity of the groups which were cost based.

An 8-item survey instrument was designed and distributed to

chief business officers at 160 private comprehensive colleges and











universities. The statistical data analyses were based on 111 usable

responses. Data from each institution for 31 identified quantitative

variables were analyzed to determine the differences existing among

the institutions when grouped by tuition pricing strategy. Responses

from the survey and the associated variables were compiled and

subjected to statistical analyses designed to test the general null

hypothesis and identify the dominant pricing strategy.

Mean values for each interval variable were calculated and

analyzed using the one-way analysis of variance (ANOVA) procedure

with a follow-up Duncan test (p<.05). For those variables having a

nominal scale, the chi-square statistic was used. Statistically

significant results were reported at the p<.10 level. If a signifi-

cant value was determined for a particular variable (less than .10),

it was assumed that the probability of the sample variable occurring

by chance, if the population means were equal, was less than 10%. In

these cases, the general null hypothesis was rejected and the alter-

native hypothesis accepted. The alternative hypothesis stated that

there were differences among the institutions when they are grouped

by institutional pricing strategy on any of the 31 institutional

variables.

The 31 variables were grouped into six major categories

consisting of endowment, revenue, enrollment, faculty, expenditure,

and external related variables. A summary of these categories is

provided in Appendix 4. Statistically significant results were

determined for six different variables based upon the one-way ANOVA

and chi-square procedure. A summary of these significant variables











was shown on Table 16. There were other variables discussed in

Chapter Four that were not statistically significant, but appeared to

provide a fairly strong distinction between pricing strategies.

Further research efforts could clarify these results.





Discussion



The survey results from 111 respondents indicated two

dominant pricing strategies. Residual pricing was identified by

42.4% of the respondents while peer pricing was used by 33.3% of the

chief business officers. Together these two strategies accounted for

over 75% of the respondents and, coupled with the secondary strategies

identified, were clearly the most important pricing strategies in use by

private comprehensive colleges and universities. This confirmed

earlier research by VanAlstyne (1977), Johnson (1979), and Hopkins and

Massy (1981). To a much lesser degree, proportional (11.7%),

externally indexed (9.9%), and differential (2.7%) were identified.

The use of residual pricing as the most identified strategy confirmed

the findings of Ihlanfeldt (1981). In these institutions tuition was

set to balance the annual operating budget within the context of

remaining price competitive with other private comprehensive colleges

and institutions.

The use of peer pricing as a major pricing strategy clearly

indicated the market's sensitivity to the price of tuition. This was

confirmed by the survey results which indicated that the market











reaction to changes in tuition was a significant consideration by

those responsible for setting tuition. It was anticipated that the

differential strategy would have been reported by a higher number of

respondents. The research results were consistent with private

comprehensive colleges and universities which generally did not have

specialized differential priced programs such as law and medicine.

Although not statistically significant, those institutions following

the external pricing strategy were associated with the highest levels

of median family income. It is logical to assume that this relation-

ship does exist due to the use of this measure in various published

indices.

It was interesting to note that many of the items stressed

publicly as reasons for increasing tuition were not highly ranked by

the survey respondents in the listing of much important factors.

These included new programs and equipment, quality improvement, and

quality consideration. Future researchers might specifically

identify a set of factors and produce a survey to elicit these

responses. It had been anticipated that these factors would have

been identified by many more of the institutional respondents.

(Association of Governing Boards of Colleges and Universities, 1986)

The most important statistical results were in endowment-

related categories. Throughout the literature review, the topic of

endowment and its impact on an institution was rarely mentioned in

the context of pricing strategy. Endowment as a broad measure

reflected the capitalization of the institution and allowed the

institution to provide improved educational and support services.











Included in this category were faculty development, specialized

research, and student financial aid.

Endowment income as a percent of total education and general

revenue was the most significant factor associated with a pricing

strategy. Those institutions following the proportional strategy had

the greatest relative use of endowment income as a supplement to

tuition income and were significantly different from institutions

using the externally indexed or residual methods. This was confirmed

by the measure of endowment income per student and the endowment per

student.

The conclusions drawn from these results are that policy-

makers using the proportional pricing method relied on endowment

income to provide a much more significant contribution to income than

those identifying the external or residual strategy. The proportional

category was also significantly different from the external, residual,

and peer categories in terms of endowment income per FTE student.

This confirms the emphasis placed by policymakers in institutions

following the proportional strategy on the use of endowment funds to

support current levels of expenditure.

The level of endowment per FTE student was also significantly

different between those institutional respondents identifying the

proportional strategy and those following the externally indexed or

residual strategies. This measure of relative institutional wealth

enabled those institutions the following proportional strategy to

remain price competitive while providing a higher level of

expenditures per student.











Those institutions with these higher levels of endowment also

reported the largest commitment of total expenditures per FTE student.

This confirmed the ability of those institutions, on average, to

provide services to students over and above basic instructional

services traditionally linked to tuition.

The distinction continues when evaluating total education and

general revenues. The proportional strategy was statistically

different than the residual category, but the mean dollar differences

of education and general revenues per FTE student were more apparent.

The proportional group had total revenues on average more than twice

that of the residual group.

The 5-year change in undergraduate enrollment was signifi-

cantly different among pricing strategies when evaluated by the

chi-square test. Those institutions having increases in enrollment

above the expected were proportional and peer. The group identified

as externally indexed had experienced a decline in enrollment over

the 5-year period. The residual group had reported responses

approximately equal to the expected responses. It appeared from

these data that during this time period the institutions following

proportional, peer, and residual strategies were more successful in

stabilizing or increasing enrollment than the institutions reporting

externally indexed strategy.

The proportional pricing group of institutions segmented

revenues quite effectively, keeping tuition charges about the same as

other pricing categories but providing total expenditures at

substantially higher levels. Also, private gifts and grants were

significantly higher per student for this group of institutions.











Another significant conclusion to be drawn from the research

was the effect of market share on the selection of the pricing

strategy. Mulvihill and Paranka (1967) indicated the importance of

market share in dealing with the issue of pricing. For purposes of

this study regional market share was defined as the percentage of

enrolled students in private colleges and universities to total

students, both public and private, within the eight national regions

(Table 13). Rusk and Leslie (1978) concluded that in states with

high market share, average tuitions were higher. This research

clearly identified two major groups as discussed in Chapter Four.

Peer pricing institutions tended to be found in areas of lower than

average market share where sensitivity to price is more important.

Residual institutions tended to fall in higher than average market

share areas, implying a greater ability to pass on to students the

costs of education.

The percentage change in tuition and fees over the 9-year

period 1972-1973 through 1980-1981 was a significant variable in

differentiating those institutions following the proportional

strategy from those following either peer or residual strategies.

The proportional group, basing tuition on some component, primarily

instructional, had the higher rate of increase (190%); however, the

current level of tuition and fees for all groups of institutions was

about the same with no statistical difference. This implied that the

1972-1973 level of tuition was below the average for the proportional

group of institutions. Given the substantial increases in the

underlying cost categories, those institutions basing tuition on this