Skip to main content

Full text of "ERIC ED055586: Community College Enrollment Projections."

See other formats


ED 055 586 

JC 710 250 






Connor, Aikin 

Community College Enrollment Projections. 

American Association of Junior Colleges, Washington, 



American Association of Junior Colleges, One Dupont 
Circle, N. W. , Washington, D.C. 20036 ($1.00) 


HF-S0.65 HC-S3.29 

♦Enrollment Influences; ♦Enrollment Projections; 
♦Enrollment Bate; ♦Enrollment Trends; ♦Junior 
Colleges; Student Enrollment 


In this discussion of enrollment forecasting for 
community colleges, a new point of view is expressed. Traditional 
theory characterizes enrollment as a function of history. The 
historical approach used such methodologies as; (1) cohort, or 
percentage of survival, (2) curve-fitting, (3) ratio-method, and (4) 
correlation-analysis. However, the new perspective outlined in this 
discussion views enrollment as the product of demands for educational 
services by the student population, mediated by the operations of the 
college. Four major process areas are defined and discussed as 
appropriate to the new enrollment projection method. They are (1) 
definitions and formulas pertinent to the projection model, (2) 
outline of appropriate procedures for enrollment forecasting, (3) 
analysis of current enrollment context, and (4) collection and 
analysis of necessary institutional and. demographic data. The major 
difficulty in implementing this new method is the quantification of 
the effects of various limiting and stimulating variables on 
enrollment. In discussing the four major process areas, an 
operational definition for each variable is offered and used in a 
step-by-step procedure illustrating the new principles. Although this 
treatment of enrollment forecasting is new in relationship to the 
historical or classical methods, it does not claim to be final or 
definitive. Rather, it is advanced as an alternative procedure for 
viewing the problem of enrollment forecasting. (Author/AL) 



THE PERSON OR organization ORIG- 
_ A IONS STATED do not necessarily 







Aikln Conner 


In this discussion of enrollment forecasting for community colleges a new 
point of view is explicated. By traditional theory, enrollment has been char- 
acterized as a function of history. The perspective outlined in this discussion 
views enrollment as the product of demands for educational services by the 
population mediated by the operations of the college. 

Perhaps the major difficulty raised here is the quantification of the effects 
of limiting and stimulating variables on enrollment. An operational definition 
for each variable is offered and used in a step by step procedure illustrating 
the new principles. 

This treatment of enrollment forecasting does not claim to be final or 
definitive. The best hope is that a different light has been cast on the prob- 
lem, and that a new perspective has been gained. 

Aikin Connor 



The Historical Perspective 1 

The New Perspective 2 


Definition of the Context.-'.- . 3 

Definition of Enrollment 4 


Limiting Factors . 5 

Stimulating Factors 6 

Administrative Control of Factors - — 7 

Summary 7 


Quantifying Limiting Factors. 9 

Quantifying Stimulating Factors 9 


Definitions and Formula 10 

Outline of Procedures 10 

Analysis of Current Enrollment Context 10 

Forecasting Future Context. . . ... .. 11 

Data. ... . . 13 

Price $1.00 

ry "the. Qt£ff>er dttd The 
ffeSeC-ia-hie-rj ■ofTr, 


Copyright 1971: American Association of Junior Colleges 
One Dnpont Circle^ N;W. 

Washington, D.C. 20036 i c 



It has become commonplace to describe com muni ty 
colleges as “the fastest-growing segment of higher 
education in America." From a total of approximately 
600,000 students in I960, enrollment in two-year col- 
leges in the United States rose to approximately 2.5 
million students in 1970. 

To those in the community college movement, this 
growth has been heady and exhilarating. But to indi- 
vidual colleges, it sometimes must have seemed dis- 
concerting, if not chaotic. Some years it has seemed 
that students were virtually pounding at the college 
doors demanding admission. As the capacity for 
classes been readied without noticeably diminish- 
ing the d emands for services, new facilities have been 
required — some temporary, some per manent . 

As enrollments and demands have increased, more 
and more attention has necessarily been directed to 
planning for community colleges. Naturally, the .first 
question voiced by planners concerned the number of 
students their colleges might expect to enroll over the 
period of the long-range plan bring developed, and 
at any one time during that period. 

The Historical Perspect i ve 

For the most part, community college planners 
have relied upon a well-developed and generally 
accepted methodology for predicting enrollments. 
This methodology consists of several procedures or 
• for base-lmes, and 

future projections. • The most c^ ose^(cxdudr; 

mg ^ of. procedures) are: (l) ^ 

hort survival, (2) -ci^ ^ ^ •. 

Jatioh^ 1960 % A 

of — techniques fa givefc^^ a brief over- 

vie n I n rmay be hdpfnl^; ■ 

■ • ciyy 

1. Cohort survival (sometimes called “Per cent 
survival") determines the extent to which a specified 
group oi individuals — a cohort — survives either by 
grade from first grade, or by year of age from birth, 
through college graduation. The survival rate is then 
applied to predicted area population predictions to 
estimate the expected college enrollment. 

2. Curve-fitting determines a relationship be- 
tween past enrollment and years, then projects to a 
future date by extending the shape of the curve thus 

3. Ratio-method describes the ratio between the 
persons currently enrolled in college and the “parent" 
population of which they are a part. By- applying this 
ratio to projections of future population, college en- 

" rollment is predicted. 

4. Correlation-analysis seeks to establish an asso- 
ciation (or a network of associations) between enroll- 
ment as a dependent variable and one or more inde- 
pendent variables. Thus, when changes are observed 
or predicted in the independent variables, enrollment 
may be predicted as a concomitant. 

fa there a thread of commonality linking these 
methods? Does some one identifiable orientation or 
point of view underlie these procedures? If so, what 
characterizes the “established perspectives"? 

Quite simply, the point of view or perspective from 
whi<h these procedures stem and which is reflected in 
most decision making in colleges fa this: The -future 
" lean best be predicted by projecting the past. 

• r Tnsmjr instances, no doubt, lessons learned from 
Tikm iy ano mpftnmg fnl And in some circumstances, 
projection techniques of ^ ^ forecasting are 

"• perfectly adequate. The discussion which follows 


steins from the basic assumption that to be of value 
to community colleges, enrollment forecasting proce- 
dures must be able to respond to -.he question “What 
if . . in such a way as to give administrators reason- 
able grounds for making planning decisions. The 
college planner-administrator must be able to predict 
the results of planning decisions in terms of enroll- 
ment before he makes those decisions. 

Thus, enrollment forecasts in community college 
planning not only must prompt planning decisions, 
they must reflect them. From this point of view the 
administrator becomes an active instrument in de- 
termining enrollment, not a passive spectator who 
merely watches events which affect student enroll- 
ment unfold, and tries to react appropriately. 

The question raised here is whether the methods 
which reflect the established perspective offer the 
community college planner the possibility of answering 
the “What if . . .?” question. Will they allow him to 
assess the effect upon enrollment of a planning or 
policy decision? 

If he relies upon cohort-survival to forecast enroll- 
ments, he cannot estimate the effect of a policy or 
planning decision upon enrollment because he cannot 
make his decision retroactive. The administrator may 
know that a certain percentage of high school juniors 
have, in the past, survived their fres hm a n year in 
college, but this information is of little value when he 
modifies the circ umstanc es in which that survival rate 
was established. 

Similarly, a curve based on a relationship between 
years and past enrollment (e.g. growth rate) cannot 
take into account the effect of even the most dra- 
matically effective planning decision upon student 
enrollment. Again, the administrator cannot m a ke 
his decision retroactive. 

Hie administrator who depends upon the ratio 
method has no way of estimating the effect of a speci- 
fic decision. Enr ollment projections which merely 
reflect changes in fixed proportion to the population 
cannot take into account changes that might result 
from an administrative action. 

As long as the decision of the community college 
administrator causes quantifiable changes only in 
those variables for which associations have been estab- 
lished, he may be able to assess the effect of a decision 
upon enrollment using correlation analysis. However, 
he should be prepared for surprises if his decision ex- 
tends the affected variables beyond their previously 
observed ranges. What had appeared to be a simple, 
straight-line relationship within a narrow range of 
observations may well become curvilinear when the 
range is extended. V ?- ' V.' ' 

Useful as some of these methods may. be as descrip- 
tive tools,' a new approach to the problem of estimating 
future enrollments seems to be needed.. V 

The New Per s pect i ve 

In order to avoid misunderstandings about the 
analogy used here of “perspective” a word of expla- 
nation may be called for. Perspective is meant to imply 
that the frame of reference, model, set of assumptions, 
etc., represent a consistent way of viewing enrollment 
predictions. Traditionally, this view or perspective has 
been that enrollment is a function of history: that 

future enrollments may be predicted by carefully 
analyzing past enrollments and applying a “formula”; 
in short, that the future can best be predicted by pro- 
jecting the past. 

A new perspective, then, means only that a different 
point of view is developed. The new perspective does 
not deny that the future can best be predicted by 
projecting the past; neither does it confirm it. The 
new perspective is a different way of looking '' enroll- 
ment prediction, a different conceptualization cf the 

The new perspective that will be developed here 
sees enrollment not as a function of the past but in 
terms of a continuous present. Enrollment is charac- 
terized as a measure of educational demand mediated 
by the co liege’s response to that demand. Thus, total 
enrollment is the sum of individual enrollees, each of 
whom is enrolled because his personal demand was 
responded to by the college. Forecasting college en- 
rollment is viewed as a problem of estimating the 
interaction of two factors or forces: demand for edu- 
cational services, and institutional limitations on 
meeting the demand. 

Perhaps the most compelling argument favoring 
the new perspective is that the assumptions of the 
model hold for the individual enrollee as well as fur 
the total enrollment; that the analytical procedures 
followed to forecast total enrollment may be applied 
to any individual enrollee. 

As a simplified illustration, consider an individual 
case. A person is motivated (or not) to learn, become 
educated, acquire a skill or skills, or gain status and 
credentials. He becomes counted in enrollment data 
if: (a) he is sufficiently motivated and (b) the college 
responds positively to meet his educational needs or 
demands. He is not counted in enrollment data if: 
(a) he is not sufficiently motivated, or (b) if the college 
does not respond positively to meet his educational 
needs or demands. 

Hie total enrollment of a college at any time, then, 
represents a measure of positive response to the edu- 
cational needs or demands of its community. Predict- 
ing enrollment, from this point of view, depends' upon 
accurately estimating the force of educational de- 
mands in the community served., and assessing the 
effects of institutional responses to those demands. 
The first requires foresight; the second, under- 1 


in the previous section, as well as in most discus- 
sions of the problem, these terms are used: forecast- 
ing, prediction, and projection. For the sake of clarity, 
the following distinctions will be made in this dis- 

“Forecast” means a general description of what is 
anticipated. Just as a weather forecast may be made 
over a period of time which describes atmospheric 
conditions expected as a result of various observable 
forces of low and high pressure systems, moving and 
stationary fronts, etc., an enrollment forecast should 
describe potential events and interaction of events 
which will affect enrollment. 

When the time-span is shortened to a specific point 
and conditions which will prevail are specified, a 
“prediction” is made. Tomorrow's weather is pre- 
dicted to be fair and cool because a high pressure 
system preceded by a cold front is expected to move 
into the area by tomorrow. Next year's enrollment is 
predicted to increase by so many students because a 
new industry is moving into the area which will re- 
quire skills and vocational interests not previously de- 
manded and the college is preparing to respond to the 
anticipated need. 

The term “projection” is used here to describe 
processes of extending the past into the future. A 
trend is established by observations of events, such as 
a year-by-year increase in the purchase of compact 
cars. Based on this observed trend, automobile manu- 
facturers may project a specific volume of sales of 
compact cars for the next five years. Based on s imil a r 
trends in enrollment figures, college administrators 
may project specific number? of students over the 
next five years. 

To continue the analogy of the anto manufacturer, 
he malcea his predictions of sales of compact cars so 
that he can {dan for the necessary re-tooling, re- 
designing; hiring of personnel, and advertising to 
meet the demands he has predicted. The college ad- 
ministrator does the same. 

Because college facilities cannot be constructed over- 
night, the process of facility-planning is a continuing 
one. The administrator must know not only the condi- 
tion of the current facilities and their “fife expectancy” 
but he must look forward to specific demands upon 
the college facilities in the future. 

T - context of a rapidly changing work world, 
] FRin planning has become ah urgent need. Not 

only must the college administrator seek to meet the 
changing occupational demands of his constituency, 
he must be able to foresee how the introduction of new 
programs and the phasing out of unnecessary ones 
will affect the enrollment in those that continue. 

Enrollment forecasts are vital to the planning for 
staff, facilities, and program. Sheer numbers of stu- 
dents exert demands on faculty, administrative staff, 
counselors, librarians, custodial personnel, and other 
staff required to support the operation of the college 
and the campus. The magnitude of those numbers 
must be predicted well in advance in order to avoid 
institutional chaos and administrative trauma. 

Enrollment forecasting assumes different contexts, 
depending upon the objectives, goals, and operational 
mode of the college. In general, two-year colleges in 
the U.S. are classified as private or public, depending 
upon their governing and supporting agencies. 

Privately controlled colleges may be proprietary 
schools, such as the many business and tec hni c al 
schools operated by an individual or a profit- makin g 
company. Such schools forecast enrollment much as a 
factory must forecast production — in terms of meeting 
a demand it has fostered and encouraged. Like the 
factory, the proprietary school will do its utmost to 
satisfy the predicted demands. 

Other privately controlled colleges are operated by 
independent, non-profit organizations. These colleges 
need to mak* enrollment forecasts, too, but in many 
instances, the purpose is to allow the administrator 
adequate lead-time to recruit students (if a decrease is 
predicted) or secure funds (if an increase is predicted). 

Church-related colleges utilize enrollment fore- 
casting in the same way independent, non-proprietary 
schools use them. Long-term plans and policy deci- 
sions require an estimate of expected enrollments. In 
particular, changes in admission polities must take 
into account their effect on future enrollment. 

The great majority of students who are enrolled in 
two-year colleges are enrolled in publicly supported 
colleges. These may offer occupational curricula, 
academi c transfer curricula, community services, and 
continuing education, as ' well as virtually all com- 
binations of these programs. The administrator of a k 
public college must be as jeady to meet the educa- 
tional demands of his constituency as is the proprietary 
school administrator. He must be as sensitive to the 
effect of policy decisions on enrollment as the private : 

college administrator is. In short, the community 
college administrator must grasp the reasons under* 
lying student enrollment in order to make significant 
enrollment forecasts. 

Perhaps the greatest problem in forecasting enroll- 
ment in community colleges results from the unde- 
fined student population they serve. Unlike univer- 
sities whose students are drawn almost exclusively 
from the ranks of the so-called “college-age popula- 
tion,” community colleges serve not only the college- 
age population but a substantial segment of the 
“post college-age” population, as well. Forecasting 
procedures which rely upon statistics derived from the 
limited college-age population are likely to produce 
errors of considerable magnitude when they are used 
in community colleges. 

Definition of Enrol hngi i t 

Probably no college administrator views his institu- 
tion as a monolithic establishment. He is more likely 
to think of it as an inter-related collection of educa- 
tional services — a functional identity rather than a 
formal one. His budget requirements reflect the 
aggregate needs of all the service functions of the 
college noth internal and external. If he is committed 
to a “PPBS” approach be is made constantly aware of 
the diverse nature of his institution's identity. 

The college administrator is also aware that the 
total enrollment of his college represents an aggrega- 
tion of the enrollments for each sector of college 
activity. In other words, college enrollment is the sum 
of the enrollment of liberal arts transfer students, 
science transfer students, allied health career students, 
automotive technology career students, electronic 
technology career students, part-time non-credit stu- 
dents, etc. Those factors which affect enrollment in 
one segment of the coliege may have do effect, or 
only tangential effect, on another segment. The sig- 
nificance of this fact for enrollment forecasting is 

Using the old perspective, this diversity of function 
(based on diversity of educational need) would suggest 
different projection methods. It is safe to say, for ex- 
ample, that the bulk of students in transfer programs 
are: (1) recent high school graduates or, (2) under 25 
years old. Hus segment of the enrollment has proved 
to be the most stable, its rado-to-population most 
nearly unchanging of any group of enrollees. There- 
fore, a fairly accurate prediction might result from the 
use of the ratio method of projection. 

In another case, career enrollment is very likely to 
be associated with certain employment variables, 
suggesting correlation analysis as a projection method 
for this segment of the enrollment. In short, if one is 
to project future enrollment the method or methods 
employed should be those which most efficiently use 
the information available. 

When enrollment figures are issued for a college 
they may be: (1) headcount — current registration, 
(2) headcount — cumulative for the academic year, (3) 
full-time equivalent — current registration, or (4) full- 
time equivalent — cumulative for the academic year. 
The type of enrollment figure employed depends 
upon (in some cases) legal requirements in definitions 
or the use to be made of the data. A brief discussion of 
each type may clarify terminology. 

Headcount enrollment means simply counting all 
registered students, full-time and part-time. In some 
instances colleges count only degree-credit students; 
in other instances all registered students are counted. 
The definition or determination of who is a “student” 
and who is not raises problems in dealing with com- 
munity college enrollment data. Some colleges offer 
many short-term non-credit courses but do not count 
registrants in these courses as enrollees; other col- 
leges include them. Sometimes they are counted for 
one purpose but not another, giving the college en- 
rollment picture a thoroughly unreliable and inaccu- 
rate appearance. Any enrollment forecast, then, must 
be made in terms consistent with itself. 

Full-time equivalent enrollment is determined by a 
mathematical formula applied to headcount figures. 
The formula varies from locale to locale and almost 
always reflects state regulations. In turn, the state 
regulations reflect varying requirements. The Ameri- 
can Association of Junior Colleges, for example, uti- 
lizes a formula for FTE of full-time plus three-quarters 
part-time to establish institutional membership dues. 

Current registration or sometimes “opening fall 
enrollment” is a count (whether headcount or FTE) 
of students enrolled at a specific date. Since the 
“academic year” begins with the fail term (quarter or 
semester) usually enrollment figures are based upon 
registration at some point during the fall session. This 
date may be the final date for registration in the ses- 
sion, the last day to drop classes without penalty, or 
an arbitrary date early in the session. 

Cumulative enrollment ordinarily means the total 
number of enrollees (headcount or FTE) who have 
been in college during the academic year — fall and 
spring semesters or fall, winter, and sp£ng>guarters. 
However, practices are not standard here, either. For 
example, some schools accumulate enrollment begin- 
ning with the first session of a calendar year — the 
spring semester or winter quarter. Also, some schools 
include summer sessions and some do not. Again, 
enrollment forecasting most use definitions which are 
at least internally consistent. 

From the foregoing discussion it is obvious that 
enrollment forecasting must concern itself both with 
disaggregation of total enrollment and consistency in 
defining enrollment and its elements — students. 


From the proposed new perspective, enrollment In 
community colleges is conceived as a quantification of 
the demand for educational services mi t ig ated by the 
ability of the community college to provide those 
services. This suggests two factors that an enrollment 
forecase must deal with: (1) those factors which stimu- 
late demand, and (2) those factors which limit the 
satisfaction of demand. Although both of these factors 
will be more thoroughly discussed later, a simple 
example will be useful here. 

One factor that stimulates demand for community 
college services is the opportunity for employment 
of college graduates. A factor which limits the satis- 
faction of that demand by the community college is 
staff. For example, if a specific occupational program 
c xries virtually certain employment as a reward for 
its completion but the college employs only one staff 
member for that program, enrollment will be limited 
to that number which Jne single staff member can 
accommodate'. On the other hand, no matter how many 
instructors the college may have available for, say, 
widget technology, if there is no market £<*■ widget 
technicians, it seems unlikely that enrollment will be 

Limitiiiff Factors 

While idly observing the flow of automobile traffic 
on the street below an office window, if one were to see 
the cars on the street suddenly start up and move when 
a green light appeared, one might well infer a causal 
relationship. The sudden appearance of the green 
light might appear to cause or stimulate the movement 

liberalization of admission policies, there is a causal 
relationship to be inferred. However, just as the green 
traffic light merely allows the movement of auto- 
mobiles, liberalizing admission policies simply allows 
enrollment to increase. It is important to realize that 
the traffic will move on a green light only if there is 
traffic; and enrollment increases will follow liberal- 
ized admissions only if there is a commensurate de- 
mand for admission. To get a dearer picture of the 
many ways in which colleges limit the satis f a c tion of 
educational demands, let us consider a few specific 

Admission policies. Many institutions boast that they 
maintain an “open door” admissions policy, allowing 
any high school graduate to enroll. However, the 
“door is dosed to everyone who has not gradua te d 
from high school (or passed an equivalency exam). The 
would-be student who dropped out of high school be- 
fore graduation most return to school or pass an 
equivalency exam before the college will respond to 
lus educational demands. Other community colleges, 
such as those in California, allow any high school 
graduate of any age, or any adult over 18 years of age 
who can benefit from college, to enroll. The high 
school junior, aged 16 or 17, who could easily profit 
from college work may be excluded. The college will 
not respond to his education demands. Those colleges 
which require a specific high school grade-point 
average or a mimmnm score on a standardized test 
deliberately choose not to respond to the educational 
needs of those potential students who do not meet 
these criteria. 

Facilities. Enrollments are limited by the size, kind, 
and location of facilities. How. many students will 
existing facilities accommodate? What kind of pro- 

AKhoughwe_ ^ 

the flow oCtzaffiq, thtynuady aBowtraffic to move, 
we donot : in file 

operation qf colleges. We are sometimes led to con- 
0 "- fha*-, because enrollment increases follow .a 

WiM ' ■ V | : ;|||: . ; ■ . @ ■ ;1@| 

grams or cnrncola are aictafed by available facilities? 
How proximate are the college’s facilities to its clien- 

tele? If 

existing facilities can accommodate 


students, either facilities must be expanded or enroll- 
ment must be limited to 1,000. If the facilities are 
loca ted t hr ee miles from town, enrollment is limited 
to those who can arrange for time and transportation to 

Program. Enr ollments are possible only in existing 
pr ogr a ms or curricula. A college that restricts its 
curriculum to transfer programs will limit its enroll- 
ment to those whose educational needs can be thus 
ye** ft mmode < ’^ Conversely, a college that offers 
only vocational and occupational programs will enroll 
none whose career or educational needs require 
completion of a four-year curriculum. 

Staff. No matter how liberal a school’s admissions 
policies are, bow large and well-located its college 
faciliti es and no matter how extensive its curricular 
offerings, enro llmen t will be limited by the size, 
quality, and style of the staff. The most obvious limita- 
tion is size. While faculty-student ratios vary from 
program to program, in all programs there are limits 
beyond which the faculty-student ratio cannot be 

Less obvious but frequently more important is the 
limitation brought on by low-quality faculty. The com- 
munity college, like most public institutions, develops 
a kin d of image by which it is known and evaluated. 
When this image includes a poorly qualified faculty, 
student demand win be low. 

The style of the staff is the most elusive of all 
limitation^- No matter how large nor how well quali- 
fied the staff may be, students respond im m edia tel y 
to its style. If that style is too reminiscent of high school 
teachers, successful high school students are often re- 
pelled. Even in colleges where unsuccessful h igh school 
students are admitted, faculty style may exert a severely 
limiting effect. 

Stimulating Factors 

Distinguishing between stimulating and limit i n g 
variables is sometimes difficult; recognizing stimulat- 
ing factors is no easier. Just as a limiting variable is 
not fu ncti on?! unless it actually does limit the satis- 
faction of a demand (if there is no traffic, the traffic 
fight does not have anv effect!, a stimulating factor is 
not functional unless it actually stimulates edu c a tio na l 
demand. •; ;••• 

3here are at least three reasons why recogn itio n of 

' rfimnlatinp variable can be a prohlexn. One is the 
O act that students (or would-be students) respond 

diff erently to the same stimulus. 'Hie draft law exempt- 
ing students stimulated many “new” students to 
enroll who would probably not otherwise have con- 
sidered college. But it only stimulated demands from 
males of draft age. 

A second cause of difficulty is that sti mul ati o n is 
often peculiar to a situation. Occupa tion a l require- 
ments of a specific community may stimulate enroll- 
ment in the local community college in programs 
Awrigneri to train for the needed occupations. Those 
same programs, however, would probably not stimu- 
late enrollment if occupational needs were different. 

Third, and most important, no matter how stimu- 
lating an event or events may be potentially, if a 
limiting condition stands squarely in the path of 
satisfaction of the demand stimulated, no enrollment 
incr ease (and therefore no stimulation) will be noted. 
In a college with stringently limiting admission 
polities and operating at capacity, even so strong a 
stimulan t as the draft law exempting students would 
be unlikely to have a noticeable eifect. 

When considering factors that could sti m u l ate 
educational demands it might be well to divide the 
general category into two sp e c ific sub-categories. 
One of these sub-categories would contain those 
factors which derive from local or regional events, or 
forces; the other would include those which arise out of 
national or international events or forces. A ga in , a 
few specific examples may clarify the concept. 

Local Stimulating Factors. One type of sti m u l at in g 
factor deriving from a local event might be the devel- 
opment of a new local industry which promises em- 
ployment to college graduates. While it is true that 
many of those who would pursue a training course 
required by the new industry might already be 
enrolled in some less promising program, the likeli- 
hood of employment would stimulate demands from 
those not enrolled. 

£ second situation stimulating educational demand 
is one which increases the target population or poten- 
tial clientele of the college. Examples of this kind -of 
event are commonplace, the opening of a military 
base being a striking one. Not only does such an event 
create a new clientele, it altos the charac t er istics of 
tire community and the educational demands to which 
tiie college must respond. 

National Stimulating Factors. A perfect example 
nf » d-imnlating faftnr deriving from a national event 
- has mentioned previously; the draft law which 

jwmipted students. It is safe to say that no college was 

an touched by this event (although seme may not 
have responded to the increased d em a nds it stimu- 
lated). It remains to be seen whether the stimulation 
created by the exemption of students from the draft 
nuA> a lasting and permanent change in terms of 
educational expectations. 

Administrative Control 

offered which could become careers. One look at a 
map of the community convinced the president of 
Wilson College that the location of the college facilities 
severely limited enrollment of students from that high 
school. By leasing the necessary facilities from the 
high school and arranging efficient transportation 
from that area of the community to the main college 
campus he effectively removed a very real limitation to 
enrollment from an important segment of the college’s 

From the foregoing examples of factors stimulating 
and limiting college enrollment it is evident that many 
times the college administrator has little control. Be- 
cause of varying degrees of autonomy or local 
authority among colleges it is difficult to make general 
statements regarding administrative control. How- 
ever, examples of controls which, at least in some 
colleges, may be exerted will give direction to the 
college administ rator who is familiar with the con- 
straints of his own situation. 

The administration of Wilson College is co mm i t ted 
to the philosophical view that the community college 
should serve the educational needs of individuals 
within its community. The president recognizes that 
some high school students are very well piepared for 
college studies before thair graduation from high 
school. In fact, some students with high academic 
potential may get “turned off” to school if their need 
for more challenging coursework is not met. 

With this in mind, the college administrator seeks 
to remove limitatio ns which keep potential stu d ents 
from college enrollment. First, he finds that state law 
restricts enrollment to those who have completed 
hi gh school (or its equivalent) or are 18 years or older. 
However, he finds nothing in the state law which 


Realizing that many potential students were not 
sufficiently motivated to enroll in Wilson College, the 
administration hypothesized that the college was not 
^lyi tnmnnifaring '* with this segment of its community. 
To stimulate enrollment of these s t udents, a ca m paig n 
of counseling visitations to local high schools and 
shopping centers was mounted. The increased 
visibility of the college in this sector of the community 
stimulated a demand for college services. 

These examples could be multiplied many times 
over, but perhaps these few will suggest ways in which 
administrative action can effect or control factors 
relating to college enrollment. 


The conceptual framework developed above results 
from what was termed earlier as the new perspective. 
College enrollment reflects the action and interaction 
of stimulating and limiting forces. Some of these 
forces stem from local situati ons and may be subject 
to control by the college administrator. Other forces 
are derived from national (or state) sitestions or events 
and are probably not subject to manipulation by the 

restricts attendance to those who are properly 

onmllad — that, kind of regulation 13 local and subject 
to his modification. Although he cannot offer regular 
college credit to those potential students who may 
choose to attend college classes, he can arrange for 
^ v^ ^t-hy- jyrarniTiati nn for them when they become 
eligible for regular enrollment. By using the authority 
he had, this president effectively removed or modi- 
fied a limitation on his institution’s response to an 
educational need in the community. 

The administration of Wilson College also recog- 
nized another educational demand from its com- 
munity. Among the high schools in the area, Wilson 
College serves me school which w« traditionally 
noted for the high level of attainment of its “voca tion a l 
education” gw** 1 **"*- Yet •few of those g r ad u ates 
eu~”^ : ^ coBe®v even thougi many programs were 


college administrator- 

It is obvious, of course, that the proposed concep- 
tualization does not immediately solve all the prob- 
lems of enrollment forecasting for the community 
college. For example, nothing is said about specific 
procedures for collecting and analyzing data. But the 
problems of enrollment forecasting are specific to the 
situation, and ultimately must be solved within a given 
context- Data collection and analysis are dictated by 
the requirements of a specific situation. The purpose 
of this conc ep tualization and, indeed, of "the new 
perspective is not to solve all the problems of enroll- 
ment projection in community colleges but to provide a 
framework for developing appropriate data collection 


by defining the context in 
which enrollment forecasting must operate and the 
nature of the critical factors involved. 

• ■ i 


6. There will be no major changes in requirements 
for admissions to post-secondary institutions. 

7. The post-secondary education attendance rate 
will increase as progress continues in making appro- 
priate programs of post-secondary education more 
accessible and in providing necessary assistance and 
encouragement for high school graduates to pursue 
additional education (Minnesota, p. 52) 

Perhaps the appeal of the old perspective for enroll- 
ment forecasters is that it presents few problems of 
quantification. When it uses data those data are 
“hard” and numeric. A certain percentage of students 
“survive” from high school to college, a proportion of 
a population attends collie, a tendency of enrollment 
to increase is noted in numerical terms, etc. 

The difficulty presented by the new perspective to 
the enrollment forecaster is that quantification is not assumptions. 

ready-made. The survival rate of students, the pro- (l) It was assumed that the number of freshmen entering 

portion of a population attending collie, and the • colleges or universities in the fall of 1967 would 

tread of ;inroflinmi^ equal 68S per Cent of the 1967 high school graduates 

per cent would increase 2 per cent per 
are constant or changing at a % ^private rallies are assumed to get 10 

~ mappropnate far community college planners may foe 

be illustrated by the following, excerpt." from an /v . - ... . . ' 

enrollment forecast. .. . assumed to be 50 per cent. 

In projecting future ypulafions^and enrollments. It r . , ^Another assumption* that there will be no major 

Xn^^pe^anOn^WUl ^, wstr ’ rw*P<wnnn c ' rjonroccinnc '*T»ii otai* m nnkltA . 

If the forecaster is prepared to live with such as- 
sumptions, his data are readily available and quanti- 
fied. If, however, he proceeds from the new perspec- 
tive he cannot accept such assumptions and must 
find m eans of measuring or quantifying data relevant 
to the interactive forces of stimulation and limit ition. 

Quantifying Limiting Factors. To arrive at a measure 
of the effect of a limiting factor, one assumption will 
be made: The entire population would be enrolled if 
there were no limiting factors. The corollary to this 
is, of course, that the effect of a limiting factor is 
exactly commensurate with the number of people it 
prohibits. Whether or not the people it does not 
prohibit from enrolling would actually enroll is the 
concern of the stimulating factors. 

The following simplified example illustrates the 
measure of the effects of limiting factors. 

The Context Pratt College is located three miles 
outside the city of Pratt. It accepts students who have 
completed high school or who have passed an equi- 
valency exam. It offers liberal arts transfer courses, 
a professional nursing program, and an electronic 
technology program. All students must be enrolled 
in one of these programs, which are offered only from 
8:00 a.m. to 5:00 p.m. There is no public transportation 
or school-provided transportation to the campus 
where all classes are held. Tuition is free for residents 
; of the district, $700 per semester for all others. 

that this figure of 54 per cent is true for the 71,500 
potential students described above, as well as for the 
total adult population, this limitation brings the total 
potential student population to 32,890. 

Of the 32,890 potential students remaining, 30 
per cent are from a lower income area of the city of 
Pratt and are unable to provide themselves with trans- 
portation. This limitation reduces the potential student 
body to 23,023. 

The effect of college curricular offerings is difficult 
to categorize as limiting or stimulating. To some stu- 
dents, the nursing program at Pratt College may well 
represent an incentive to enroll in college. To others, 
the lack of, say, an environmental science program 
represents a limitation. Recognizing this ambiguity 
as virtually unrssolvable, we may assign stimulating 
effect to programs offered in the college curriculum 
and assume that some portion of the potential student 
body is restrained from enrollment by a lack of 
appropriate programs. 

The observed effect of recognizable limiting factors 
at Pratt was to reduce the potential student body from 
the total population of the district — 200,000 — to an 
effective potential of 23,023. The magnitude of the 
force of limiting factors may be estimated by noting 
that 88.5 per cent of the citizenry in the district sup- 
porting Pratt are effectively restricted from utilizing 
its services. 

Quantifying Stimulating Factors 

Effects of Limitations. Although the college accepts 
qualified students from anywhere in the world it 
effectively limits enrollment to .residents of the dis- 
trict, ipar^; because of its location, partly because of 
discriminatory : tuition. The total population of the 
district is 200,000 people. 

By accepting. only graduates- of high school or its • 
e^valen^ Prott Collie effectively limits enrollment 
, (with ah occasional exception) to those people in the 
\ district over 'age 17-about l00,(X)0. r Sih<» School ut- 
| teidanreris: co^ age: 16, approximately 

30 per cent of those native to the area have hot com- 
| pleted .High school, leaving 70,000. Of: the 30,000 non- 
| graduates,' 5/'^ex-Jcent have / passed an ■: equivalency - 1 1 

► - /w oTn H *4r1 t-fvVtt -'"1 ^-WW- Ttfifonfro " ghi rf onfo ol /^mnr - 


After having identified and quantified the effect of 
limiting factors on the enrollment at Pratt College the 
quantification of stimulating factors is very simple 
in terms of aggregate effect. From a potential enroll- 
ment (after limitations) of 23,023 Pratt College 
actually enrolls 1,151 students, or 5 per cent. Clearly, 
. not everyone who was regarded as a potential student, 
in fact, became a student. '7;^-.- iT ■ • ^ 

So fkr, the quantification of variables in terms of 
the ; new perspective has not been exceptionally 
difficult,, nor has. it required extensive data collection. 
However, the ambiguities with which the college 
administrator is familiar become quite troublesome 
■;a£Pfii(s 'r stagh "of tthe £ ehroflmeht>ahafy^:-.D 
21^72 ; : potential ^dients who arc not enrolled: r^p»r^- 
sent .lack of force : in the 'stimulating factors operating 
in ' the situation? Or. do these potential students 
: repi^^f dis<hepamd^ the estimation of the effect 
;of limiting: fhctorei (hei-are they - actually excluded by 



Before proceeding with a discussion of forecasting 
procedures appropriate to the new perspective, the 
key terms in lie premise must be defined operationally. 

Definitions and Formula 

1. Enr ollment — head count of students registered. 

2. Educational demands — the aggregate effect of 
stimulating factors operative upon the population 
served; measured by the proportion of potential 
students actually enrolled. 

3. Population served— may be entire population 
of region or district, or any sub-group thereof. 

4. Institutional response— effect of limiting factors; 
measured by the proportion of population not excluded 
by limiting factors. 

. Stated as a formula: A=B (1.0 - C) X D when: 


i B=population (any group from which enroll ees are 

I ' drawn) 

| C=limitation (the proportion of that population 

* which is effectively excluded) ••'/.•V." *.?■}., 

1 D=stimulation (the proportion of the potential 

| students actually enrolled). 4 \ ; k 

I . Outline of Procedures 

tial students as the aggregate effect of stimulating 
factors currently operating. 

5. Identify areas of administrative control (e.g. 
removal of limitations by administrative action). 

6. Identify future events or conditions which may 
effectively stimulate or limit enrollment (e.g. end of 
student deferments). 

7. Determine options for administration action 
(e.g. recruitment of returning ex-servicemen). 

8. Compute enrollment forecast in terms of antici- 
pated future conditions. 

9. Repeat #1-8 for each segment of college enroll- 
ment identified. Aggregate enrollment forecast is a 
summation of individually forecasted enrollment 

li may be apparent from the above steps, that the 
procedure may be seen as three major phases: anal- 
ysis of current enrollment and forces and/or condi- 
tions which have produced it; analysis of future events 
which may affect enrollment; analysis of consequences 
of anticipated future forces and administrative action. 
The collection, analysis, and interpretation of data is a 
part of e ach phase and will be discussed in a subse- 
quent section. The following discussion will treat each 
of these phases and procedures in some detail. 

Analysis ofCurrent EnroIlmentContext 

. AnalyBM of Current Knrollment Context 

v' The procedures to be followed for enrollments ore-: 

easting will require-the following'SteEe: The enrollment.forecast inTIghtof the newp 


DSSSf.f P ■ 

identity of these segments depends upon local condi- 
tions, but illustrative examples might be the segments 
represented by transfer students, career students, 
and non-degree students. These segments would 
function quite differently in response to a change in 
transfer requirements, occupational outlook, and 
class schedules. 

If changes were made in the number of credits and 
specific courses necessary to transfer to a four-year 
institution, transfer students would be required to 
make appropriate responses, possibly resulting in the 
abandonment of the transfer program. 

If a career area, such as allied health occupations, 
were to become overcrowded, many students in career 
programs related to such occupations would undoubt- 
edly turn to other careers. 

If courses of general interest to the non-degree 
student, such as “Law for Laymen,” for example, were 
to be scheduled at 9:00 a.m., it would seem likely that 
attendance and enrollment would plummet. 

2. For each segment specified, identify the parent 
population. Once more, the specific solution depends 
upon local circumstances. However, illustrative 
examples may help to clarify the demands of this task. 

.Arden College has 1,000 students enrolled in various 
transfer programs. These students are virtually all in 
the age group 18-24 and have completed high school. 
Eighty per cent are graduates of Near High School 
and 20 per cent are graduates of Par High School. The 
total population of 18-24 year olds in the district is 
100,000, of which 40,000 live in the area of Near High 
School and 60,000 live in the area of Far High School. 
This total parent population of transfer students may 
be treated as one population of 100,000 or two popula- 
tions of 40,000 and 60,000 respectively. 

At Arden College, students in career programs are 
virtually all in the 18-24 age group. They are also high 
school graduates, with approximately 80 per cent from 
Far High School and 20 per cent from Near High 
SchooL This group also may be divided into sub- 
popui^onsL^ i^,000 ; ;60,0M to reflect high 

school: backgrpimd and/or- residence. 

The non-degree students at Arden.. College are 
drawn almost" exclusively fi»m .the 25-50 -age. group, 
hi. th6Vco™ mm, ^y there are approximately 50,000. -in . j 
this age group. 

tion or an equivalent certificate. Since 30 per cent of 
the population in this age group did not finish high 
school or take an equivalency exam, they are excluded 
from enrolling. 

Because the college campus is located three miles 
outside the city and no public transportation is avail- 
able, approximately 40 per cent of the 70,000 high 
school graduates ages 18-24 are effectively excluded by 
not having their own cars. 

Of the 42,000 population r em ain i ng, nearly 50 per 
cent must hold full-time jobs during the day. Because 
the college does not schedule career-oriented classes 
at night, an estimated 21,000 potential students are 

Other limitations imposed by the staff and the pro- 
gram are difficult, if not impossible, to quantify. For 
the sake of simplicity, only those limitations indicated 
above will be considered in the example. By measuring 
the effect of limiting factors in terms of the propor- 
tion of the target population excluded, the number 
of potential students remaining is 21,000 — 79 per cent 
having been excluded. 

4. Compute the ratio of students enrolled to potential 
students as the aggregate effect of stimulating factors. 
Of the 21,000 potential students identified by the two 
previous steps, 1,050 are currently enrolled in Arden 
College career programs. The proportion 1,050:21,000 
or .05 is the estimated aggregate strength of stimulat- 
ing factors operating in the situation. 

Because career program students are usually goal- 
oriented, the prospect of employment is unquestion- 
ably a major stimulant to their enrollment. Other 
factors migh t include parental pressures and peer 
influence. In general, it may be assumed that students 
are motivated to enroll in college: (1) because they 
like to go to school, (2) to learn a specific skill or 
trade, (3) because they are vitally interested in learn- 
ing a particular subject or area, (4) to achieve status 
and/or credentials, (5) because alternatives to college 
are less favored, or (6) a combination of several 
reasons. :■}''■ V : 

As a means of estimating the effective force of these 
reasons in a particular student body, a simple survey 
might be made. The results of such a survey should 
give hmts : abotit the effects of anticipated future 
events: The total strength of the stimulating factors 
currentiy operating, however, 3s: m&sured by the 
proportion of potential students currently enrolled. 

above) excluded ■ btf- limiting factors.: (Note: For the . Jo the example, this figure is .05. . 
sake of?simpiK3^ -A A "' 4 :' ' ” • ■ ■ ' •" " V - • 

WBMi W*s made previously that the college administrator 

probably not subject to the discretion of the local 
administrator, although it is possible for potential 
students to be encouraged through availability of 
appropriate classes to become eligible through an 
equivalency exam. 

The second limitation noted— campus location — is 
subject to change instigated by the administrator. 
While it is true that the removal of the campus to a 
more convenient location is beyond the bounds of exe- 
cutive action, the administrator can effectively remove 
the limitation in two ways. Campus extension centers 
can be established in out-of-reach areas of the district 
and, college-supported transportation can be made 
available to bring students from far points. 

The limitation easiest for the administrator to re- 
move can be done by the scheduling of classes for 
career programs during the evening and night. This 
will help the would-be student who must be employed 
full-time during the day. 

In the area of stimulating factors, the college 
administrator probably has most effective control over 
program offerings. Assuming, in our example of career 
students, that the principal motivation for the students 
being enrolled in college is to learn a specific skill or 
trade, the administrator may increase the force of that 
stimulation factor by expanding the career program 
curriculum. Although every administrator undoubtedly 
is familiar with the phenomenon of expanding curricula 
and stable enrollment (students merely shift to 
another program), that phenomenon may well be 
attributable to lack of public communication, con- 
tinuance of limitations, or several other factors. 

It may be noted that administrative options or 
courses of action like limitin g and stimulating factors 
are related to local conditions. The alert and astute 
administrator will have little difficulty identifying his 
own areas of effective action^ 

6. Identify future events or ccmditions which may 
effectively stimulate or limit enrollment. The second 
major phase of enrollment forecasting is one which is 
subject to greatest error. ®ven M^Uy-airelmmed seds 
and proplietS/?are5cffc& 

years much attention has been given to forecasting the 
future, and te<^mqp^ for } reducangl ^ the frtargm of 
error in' speculation have been developed. Perhaps 
: j the m<^ of - these newly 

^developed techniques is known as DelphL 

Although ^ basic Delphi i 

technique/have . been developed for -specific applica- 

The :basic;DeIpM procedure Is: (1) to identify indi- 

aBbw eaclf a-t relatively nn- 

^■structured: questions e£the: future^ 

questions: (Round^:^ 

... . ■ - ^ lc " * T- >33 ^ 



develop more specific structure to questionnaire for 
Round Efc (4) continue to refine responses, round by 
round, until some pre-determined goal is reached. 
(This goal varies and is the cause of variant procedures.) 

An example of Delphi procedures applied to edu- 
cational planning in two-year colleges is a study called 
“Focus Delphi.” Focus Delphi was conducted during 
the academic year 1969-70 by Delayne R. Hudspeth 
and others for the New York State Education Depart- 
ment's Bureau of Two-Year College Programs. 

Although the traditional Delphi study deliberately 
attempts to elicit consensus, the purpose of Focus 
Delphi was to discover where, within the social system 
affecting the two-year college, consensus exists. By 
selecting participants on the basis of role-function in 
the system. Focus Delphi sought to delineate areas 
of agreement and areas of conflict among the partici- 

“Knowledge of the differences of opinion held by 
those who serve different roles within a system be- 
comes valuable planning data for the policy maker in 
that it suggests one strategy for an event having high 
probability of occurrence (all sectors having agreed as 
to when an event might occur and to its potential 
value) and another strategy for an event where one or 
more sectors clearly disagree.” (Hudspeth, p. 2) 

After the participants in the Focus Delphi study 
were identified, each was sent a questionnaire 
designed “to elicit from each respondent 10 events 
that he considered as plausable occurrences for the 
future.” (Hudspeth, p. 7) 

When all questionnaires were returned, the events 
described were collated in terms of similarity, and 
edited. Events which would enter into further rounds 
were chosen on the basis of pre-determined criteria 
of frequency, emphasis*, interest, and impact on 
electro-mech anical technology and education (the 
subject of the study). 

The second round of Focus Delphi elicited from the 
participants estimated, dates of occurrence for each 
event. Participants were asked to restrict their esti- 
mates to the period of the next 15 years, with options 
of conjecturing that the event would occur “later” 
or “never,” as well. Two dates were requested: the 
earliest possible date and the most likely date. 

Round HI sought to refine the estimated date by 
£ allowing respondents to change their estimates in 
K^t of other estiihates. Also, Round III elicited value 
statements r^arding tbe incfividual and society as a 
; rwheie^ ! wi^ ^x^gEird': ■■ to - the effect of each event. ^ 

. ] The fourth \ahd : final round of Focus Delphi re- 
jporfa e results of previous 

mrm filunte On 20; events,. 

• r pgraW l t/> their positivedr negative value 

of Ojxurrence 

3 at I aihaiTOwl;y ^specified gathered in 

Round IV included participants’ perceptions of the 
“power group” related to each event’s occurrence and 
the strategy the individual, as a member of his own 
group, would follow with regard to each event’s 

Although this simple description of Focus Delphi 
indicates nothing about the conclusions of the study, 
it may, perhaps, serve to illustrate a procedure which 
the college administrator might adapt to his own 
forecasting needs. By choosing participants judi- 
ciously and directing their attention to the impact of 
events on the college, he can describe the future with 
some precision and, hopefully, with some accuracy. 

7. Determine options for adTmmstratxve action. 
After developing a “scenario” for the future of the 
college with the help of Delphi or other techniques, 
the administrator must make an analysis of his alter- 
natives or options. The consequences of exercising 
those options may be quantified in terms of their 
effects upon enrollment. Consideration of the con- 
straints of budgets, etc., may make certain decisions 

8. Compute enrollment forecast in terms of antici- 
pated future conditions. Having analyzed the current 
structure of circumstances, or the context in which 
his college functions, and having forecasted its future 
context (including the impact of administrative 
decisions), the administrator can now compute the 
anticipated enrollment in the segment of the college 
for which the analysis was made. In the formula 
Enrollment = Population x Effect of Limiting Factors 
x Effect of Stimulating Factors, he can substitute 
estimated values. 

9. Repeat for each segment of college enrollment 
identified. The sum or aggregate enrollment figures 
will be derived by adding all segment forecasts. 


The data for enrollment forecasting following the 
procedures outlined above are no more extensive than 
those required for enrollment projection procedures. 
In either case, both institutional and demographic 
data are required. 

Institutional data are those associated with the 
institution — enrollment history, enrollment by pro- 
gram, admission policies, faculty employed, etc. 
Demographic data are those related to the population 
of the district-residential locations, population density, 
transportation facilities, employment data, etc. 

The source of institutional data is, of course, the 
college itself. AH the information required is collected 
by the college and is usually fairly accessible. 

Demographic data have several possible sources. 
Basic information about the population, suck as popu- 
lation maps denoting residential, commercial, or 
industrial areas, socio-economic conditions within 
certain areas, can usually be found in the city or 
regional planning office, or may be obtained from the 
zoning commission. Manpower studies by the U.S. 
Department of Labor, economic studies by the U.S. 
Office of Economic Opportunity, school census, 
diennial Federal Census, high school records — all 
are sources of information about the community in 
which the college operates. 

In the main, the problem for the college adminis- 
trator is not finding data but using data m eaning fully. 


Campbell, Bex R. Population and Higher Education, in Missouri, 1960-1975. 
Columbia, Missouri: Missouri University, 1967. ~ 

Hodspetii,Deilayiie R. A Long-Range Planning Tool for Education: The Focus 
; JJeijpfci. Syracuse, New York: Syracuse University itesearch Institute, 1970. 

: Methodologyof ETtroUTnentPix^eclionsforCollegesanidUnirier- 

sities. 1 Axngncsxii - Association of Collegiate Registrars and 

Admissions Officers, 1960. ' - 

Minnesota Higher Education Coordinating Commission. Population and 
SindentEnroRmentsinMiKnesotii EigJierEducaiion. St. Paul, Minnesota: 
Report Nb. 2,1968- - - / 

Sim<m,Kenneth "A. andFu!lam,Marie0.i*rc^eirfic>wsq ' 

ii fo .2976-^ -1967 Edition. Washing Center for Educational 



The funds for the development of this publication came from the Danforth 
Foundation through the New Institutions Project of the American Association 
of Junior Colleges. The author was assisted by the director of the project and 
by a group which met in a spring 1971 conference. The group included: 

Lords Bender 

Professor of Higher Education 
Florida State University 
Tallahassee, Florida 
Johnnie JRuih Clark 
Assistant Dean of Academic Affairs 
St. Petersburg Junior College 
St. Petersburg, Florida 
Douglas Conner 

Director, American Association of Collegiate 
Registrars and Admissions Officers 
Washington, D.C. 

Aikm Connor 
Research Specialist 

Washington, D.C. 

. UJL Department of Commerce 
Bureau of Domestic Commerce 

Ben Gold 

Director of Research 
Los Angeles City Colleges 
Los Angeles, California 
Lawrence E. Grog 

Chief, New York State Education Department 
Bureau of Two-Year College Programs 
Albany, New York 

Horace Griffitt 

Director, Institutional Research 
Tarrant County Community College 
Fort Worth, Texas 
George Hodson 

North Country Community College 
Saranac Lake, New York 
Virginia R. Keehan r 

Robert GeU 

Director; Institutional Research $ 

Director y •• v. 

Columbus College 
Columbus, Indiana 

kv31e^. Maryland Mchmond,VIrginia 

. Former Director AAJC ;