Theory of Loosely Coupled Systems: The Implementation of
Federal Youth Employment Policies
Elizabeth Anne VanderPutten
STREET-LEVEL BUREAUCRATS, OUTPUTS AND
The concept of slippage is introduced in this chapter
specifically as it refers to federally supported work and
education programs. Slippage in such programs, it is
suggested, can be measured as the difference between the
service output levels that are promised by a project in its
contract with the federal government and those that are
delivered by the staff of the contracting social service
More important, this chapter seeks to explain slippage, in
part, by the subgoals of the staff who deliver the services.
Subgoals are defined as attainable, short-term objectives
that guide the behaviors of persons where there are no secure
means of linking actions to higher-level goals. The
hypothesis that subgoals influence action is based on work by
Herbert A. Simon.4-1 A brief review of
relevant research is presented, which points to the
appropriateness of focusing on the staff who deliver services
in seeking to explain slippage in ways to be amenable to
Reasons are offered for focusing on federally supported
youth employment policies in a study of the relationships
between subgoals and slippage. This chapter also contains a
discussion of the data to be used; procedures, limitations
and strengths of the methodology; and some implications of
2. Slippage: A Definition
Social scientists often distinguish outputs from outcomes.
In broad terms, slippage is the difference between what
policy-makers want and what actually happens. These wants are
sometimes expressed as outcomes. At other times, they focus
on outputs. Often they specify both.
Outcomes are the consequences of actions. Increased
educational achievement levels of poor or minority students
is one of the expected outcomes of many federal education
policies. Outputs are the means by which the desired outcomes
are to be achieved. Often they are the services to be
provided or a specific technology to be used.
While outcomes and outputs are related, just as ends and
means are related, they are often separated. One result is
that the achievement of one may not be preceded or succeeded
by the other. The adoption of a given technology, for example,
may not lead to higher scores on achievement tests.
Conversely, a desired outcome may be attained by means of
outputs that were, in whole or in part, unrelated to policy
One study, for example, reviewed 140 cases of
technological innovation in state and local governments. It
found that those factors leading to the incorporation of the
technology (e.g., chief executive support, location of
innovation, timely delivery of funds) are different from
those leading to desired outcomes. The study also found that
unless the needs of the staff were considered, they would
simply not efficiently implement the new procedures.6-1
Slippage can be defined as the difference between policy
intent and (1) outcomes (i.e., benefits accrued by intended
clients or recipients), or (2) outputs (i.e., the services
intended to be delivered). This study uses the second
definition. The basic reason is that service providers are
generally held accountable for delivering specific services
or other outputs, and not for broad outcomes.
3. Basic Questions
At all levels, policy-makers, who are concerned with the
problems of slippage, face three basic questions:
1. What are the determinants of policy outputs?
2. What are the relationships between staff behaviors and
3. In what ways can the actions of staff members be
influenced so as to increase congruence between what
policy- makers want and the outputs they obtain?
Even the most laissez-faire policy-maker or administrator
assumes that there will be some connection between his or her
decisions and outputs. To do otherwise would be for the
higher-level official to assume that he or she was at best
superfluous in the organization.
In some organizations, managers have considerable
influence over outputs. A factory foreman, for example, or an
office manager can observe and often measure the actions of
staff. Tasks can be specific and deadlines set. Staff who do
not comply with orders can be fired.
Nevertheless, even in these organizations, staff do not
always do as they are told; outputs do not always follow
policy. Several studies of businesses, for example, found
that the adoption of an office technology is more than simply
purchasing equipment. Unless staff accept the technologies,
are comfortable in their use, and do not feel threatened,
they will generally resist change. Failure to consider the
concerns of staff members can result in unnecessary or wasted
Slippage, in terms of outputs, seems to be particularly
widespread in organizations characterized as "loosely
coupled."7-2 In such organizations,
goals are often unclear and sometimes conflicting, the
ability of managers to control staff behaviors is restricted,
staff are often unsupervised, and the means of evaluating
them are limited. Schools and other social service agencies
are generally typed as loosely coupled.7-3
B. THE SLIPPAGE PROBLEM IN THE
1. Variation in Policy Intent
During the 1960s and 1970s, the federal government
initiated a wide range of policies and programs aimed at
improving the political, economic, and social conditions of
the poor, handicapped, women and minorities.8-1
These outcomes were to be achieved partly as a result of
The Brown decision, in 1954, and subsequent court orders,
sought to reduce some of the inequities in society by ending
de jure segregation. Among the expected outputs were equal
educational services for all students.
The Elementary and Secondary Education Act of 19658-2 authorized funds to increase
expenditures for certain economically and educationally
disadvantaged children. School districts, states and the
federal government were to provide compensatory services
where needed, and to carry out evaluations of the programs.
The Education of All Handicapped Persons Act of 1975, for
example, required schools to provide individualized education
plans for handicapped children, and to educate them in the
"least restrictive environment" suitable for their needs.8-3
The point here is simply to document the case that a wide
range of outputs (remedial instruction, program evaluations,
individualized programs) were specified.
2. Slippage in Outputs and Outcomes
Twenty years of social program evaluation indicate that,
while much was accomplished under federal leadership and with
federal funds, there remains a marked lack of congruity
between the hopes of federal education policy-makers and
outcomes attained. For example, while the educational
achievement levels of disadvantaged children improved, they
still score, on the average, well below that of their more
advantaged counterparts.9-1 At the same
time, income and race remain fairly valid predictors of both
academic failure and of unemployment, as well as of crime and
other antisocial behaviors.9-2
Of more immediate concern here, however, several of these
evaluations also show that there is considerable slippage
between policy intent and outputs. A study of the
implementation of state-administered, federally supported
educational innovation projects reported, for example, that
"the same innovation typically was implemented
differently in different school districts, in different
schools within the same district, and even within the same
school."9-3 While variation by
itself does not necessarily mean that slippage had occurred,
the variation does suggest that some slippage probably took
place. The variation, moreover, indicates that actual
services delivered depended, in part at least, on the staff
of the schools and districts.
The National Science Foundation (NSF) conducted a 4-year
study of its attempts to improve the curricula in science,
mathematics and social science across the country. The study
reported that the basic structure-the scope and sequence --
of these courses had changed little between 1955 and 1975 (the
two principal decades of NSF activity), and, moreover, that
less than 15 percent of the teachers in the country had ever
used the materials developed with NSF grants.10-1
A major study of Title I of the Elementary and Secondary
Education Act (ESEA) found that variation was the norm in the
implementation of those programs. It concluded that there was
no such thing as one "Title I program, but rather 16,000
Title I programs" (i.e., one for each school district in
the country).10-2 It is not surprising
that the study's director stated that, "although the
federal government can influence
the action of state and local governments, it cannot fully control
them" original emphasis.10-3
Faced with the growing recognition that several major
policies were not producing the desired results, federal
policy-makers attempted various means to reduce slippage. In
doing so, however, they faced a complex set of dilemmas. On
the one hand, Congress wished to ensure that the funds it
authorized were spent in the intended manner; on the other
hand, Congress wished to avoid offending states and
localities with overly restrictive regulations that might get
it embroiled in undesirable political disputes. Again, the
federal government wished to hold recipients accountable for
the "proper" use of federal funds (i.e. the
provision of approved services), while, at the same time, it
wished to allow the states adequate flexibility in
implementing programs so that intended outcomes would result.
In general, Congress reacted to the slippage phenomenon in
two somewhat contradictory ways. One was to increase the
specificity of regulations. The other was to give funds to
the states with minimal specifications in terms of outputs or
outcomes. It may be fair to note that neither approach fully
resolved the dilemmas seemingly inherent in slippage.
a. Increased regulation
In broad terms, federal regulations can be categorized as
(1) those aimed at ensuring federal funds are spent on
specific clients, that is, targeting; (2) those intended to
influence the types of services offered to clients, that is,
outputs; and (3) those designed to achieve long-range goals
or outcomes. In part because of their specificity and the
consequent ability of the federal government to hold
recipients accountable, and, in part, because of growing
concern about slippage, Congress during the 1970s focused
increased attention on targeting.
There are various procedures used to target funds. "Income
qualification tests" are intended to ensure that only
members of certain income groups receive services.11-1 "Set-aside provisions"
require that funds be specifically earmarked for certain
purposes or groups.12-1 "Maintenance
of effort" requirements are designed to ensure that
states and localities continue to spend their own funds at
the same level they did before receiving federal funds.12-2 "Comparability"
requirements are aimed at ensuring that districts spend as
much local funds on schools receiving federal monies as they
spend on other schools.12-3
In general, studies found that these provisions have
strong effects on state, district and school behaviors. For
example, studies of Title I of ESEA reveal that few
ineligible students get aid.12-4 Other
studies show that, where federal programs require specific
targeting of funds, state and local government agencies are
likely to comply.12-5
Federal attempts to influence the types of services
offered have had mixed results. While the government rarely
attempted to define the exact services to be offered, it
often sought to determine the procedures by which such
services were selected. The mechanisms included regulations
such as requirements for (1) increased coordination among
social service agencies,13-1 (2)
detailed planning processes,13-2 (3)
involving parents13-3, and 13-4 (4) appointing coordinators
responsible for sex or racial equity. Evaluations show
generally that, after the rules are known, state and local
officials comply. At the same time, studies often found
little relationship between technical compliance and actual
outputs. A study of the planning involved in vocational
education, for example, reported that:
The 15 case study states were unanimous in their
belief that the primary purpose for the preparation of
State plans was to serve as a basis for measuring
compliance ... As a result, the planning documents
produced are rarely used at the State level. States speak
of the "real" plan, which usually is a detailed
set of operating procedures, sometimes written down in a
single document, sometimes not, and the "compliance"
one which is sent to Washington and not referred to
thereafter. Nor are the full annual plans used much at
the local level.13-5
In reviewing the effects of these plans, the House
Committee on Education and Labor concluded:
It seems that what is occurring now is that the Office
of Education is demanding a great deal of paperwork and
detailed data from the States and local school districts
but then there is no follow-up... In other words, a blind
concentration on seeking compliance on paper with the
process has led to a total neglect of trying to seek the
results the process was created to achieve.14-1
A study of the Youth Employment Demonstration Projects Act's
requirement that at least twenty-two percent of the funds be
set aside for in- school youth, for example, reported that
the provision "set in motion the forces necessary for
genuine collaboration between the education system and the
employment and training establishment."14-2
The study found, however, that financial incentives alone
were not sufficient to bring about substantive change within
The various mechanisms adopted to ensure that outputs
conform with federal intent not only do not always work, but
they frequently engender local hostility as well. In fact, as
early as the late-1960s, the cumulative effect of federal
regulations was recognized as a problem, and, by the late
1970s, such cumulative effects were perceived to be of major
proportions. As Samuel Halperin, a prime mover behind much
federal education policy during the Kennedy-Johnson era,
admitted a few years ago, "The weight of federal
guidelines and regulations may soon outweigh the beneficial
impact of relatively meager federal funds."14-3
b. The "no-strings attached"
Another approach, that of turning federal funds over to
state and local governments with minimal regulations, has
also proved not to be satisfactory, at least from the
viewpoint of ensuring that certain services are offered. One
argument for delegation of federal authority is that state
and local governments are closer to the problems and more apt
to know what needs to be done. Revenue sharing, for example,
provides funds to states to be used within several broad
categories as the states determine to be most useful.15-1
If the goal of a federal policy is to provide general-purpose
funds to other levels of governments, the "no-strings
attached" approach works. If the purpose, however, is to
ensure that federal funds supplement local monies, or that
particular services are provided, the approach does not work well.15-2
4. Slippage: A Continuing Issue
There is no imperative that federal social policies aim at
particular outputs or outcomes. Within certain political
constraints, it is possible for the government to determine
its policies on purely political grounds, or to support a
position based on national security or economic needs or any
one or combination of purposes. It may decide that the only
purpose is to give funds to states to be used for whatever
purpose they choose.
To the extent, however, that specific results are desired,
slippage will be an issue. Where existing regulatory measures
have not been successful in ensuring compliance or avoiding
local opposition, other policies and mechanisms are needed.
C. AN OVERVIEW OF PREVIOUS RESEARCH
ON SLIPPAGE IN FEDERAL SOCIAL PROGRAMS
Research has done much to explain why federal social
service programs have had limited effects on the
characteristics of services offered at the local level.
Slippage has been shown to occur at all levels in the federal
social policy development and implementation chain. Generally,
it is seen as undesirable. Frequently it is taken as a sign
of inefficiency and wastefulness.
Research suggests that slippage in federal social policies
is a function of multiple factors, no one of which alone is
totally responsible. Among these are the law-making process
and the often resulting vagueness of goals and means, the
loosely coupled nature of the federal system and the federal
policy implementation chain, and the street-level bureaucrats
who ultimately are responsible for providing the service
outputs at which policies are directed. Increasingly,
research points to the local level as a potentially fruitful
area of interest, particularly because it is there that
service outputs are produced, and because the local level may
be susceptible to federal policy influence. Ultimately, many
analysts suggest, federal social policy outputs are a matter
of local commitment. What research has not done well is to
develop theories that explain the nature of local commitment
and the mechanisms by which it operates.
2. Research on Slippage in the
Policy-Making and Policy-Administration Processes
Some early research showed that a considerable degree of
slippage could be explained by the law-making and
administration processes. Work by Bailey,17-1
for example, investigated the processes by which a social
policy bill became a law, and documented the tradeoffs
involved that resulted in vague and ambiguous statements of
intent. Such statements could clearly account for much of the
discrepancy between intent and the results obtained.
Succeeding work by Bailey and Mosher,17-2
among others, examined the processes by which a social
service law is translated into programmatic terms of goals
and objective, rules, regulations, and guidelines. Because of
the law's vagueness, which permitted varying interpretations,
and because Administration officials frequently had different
agendas than those of law-makers, members of administrative
agencies had a good deal of freedom in interpreting the law
and deciding on its administration.
This line of research indicated that the existence of
vaguely worded laws, coupled with different understandings
and motivations on the Administration side and the relative
independence of Administration officials from congressional
control, could permit considerable alteration in the intent
of the law. Slippage could therefore be explained, in part,
by the nature of the federal system and by the differing
agendas of important actors in the law making and
3. Loosely Coupled Systems and Street-Level
Organizational theories, according to researchers such as
March and Simon,18-1 can be divided
among three groups: (1) those that assume that members are
essentially passive units,18-2 (2)
those that assume members are basically reluctant participants,18-3 and (3) those that
assume that members are fundamentally decision-makers.18-4 Each type of theory implies a
different type of organization, different assumptions about
and explanations of member behavior, and different analyses
Organizations where members are decision-makers are often
called "loosely coupled." Members at any given
level are largely immune from control by members at any other level.18-5 The federal system and
social service agencies are typically classified as loosely
Slippage in federal social policies can then be explained,
partially at least, in terms of the relatively autonomous
nature of members at different levels. The loosely coupled
nature of the federal system may permit Administration
officials to modify Congressional intent, as well as allow
front-line service providers considerable discretion in
decision- making to choose which services to deliver and at
what level. Slippage in federal social policies can then be
seen as a function of the loosely coupled nature of the
Recent research on slippage in federal social policies
tends to focus on policy implementation at the local level
and on those service providers who have come to be known as
These are the social workers, teachers, health care providers
and others, who are responsible for the day-to-day provision
of services to clients. A police- man is illustrative. It is
generally recognized that there are more laws on the books of
any town, city, state or national government than can be
enforced by any single individual at one time. Often the laws
are vague or ambiguous and sometimes even mutually
contradictory. As a result, the policeman must be a daily,
moment-to-moment, on-the-spot decision-maker, deciding to
enforce some laws and not others.19-2
Using what is sometimes called a "mapping backward"19-3 approach, researchers have
investigated the phenomenon of slippage in federal social
policies by starting with the street-level bureaucrat and the
local con- text. Two findings have become almost axiomatic.
One is that implementation of federal policies is ultimately
a matter of local commitment. The second is that local
service providers, the street-level bureaucrats, are central
to the federal social policy implementation process. The
questions become, then, why some street-level bureaucrats are
committed to policy subgoals, and how they translate this
commitment into action.
4. Individually Held Subgoals
Given the multiple, vague, ambiguous and of ten
conflicting nature of federal social policies; the loosely
coupled nature of the federal social service policy
implementation system; and the centrality of local commitment
and the role of the street-level bureaucrat, some of the
basic questions are: What is the nature of "local
commitment?" What is meant A. by "local?" Who
are the significant "local" actors? What is the
essential nature of local "commitment?" What are
the mechanisms by which such commitment operates?
In loosely coupled systems, where members are decision-makers,
and, according to Herbert Simon, where there are no certain
ways by which a member can tie individual behaviors to long-range
goals, individually held subgoals may be important.20-1 Subgoals are relatively short-term,
specific, and discrete objectives, the means of achieving
which are perceived by the person to be under his or her
control. They can be: (1) professional, such as increasing
students' reading ability; (2) organizational, such as
promoting a positive image of the member's school in the
community; or (3) personal, such as protecting one's job.
This line of investigation posits that individual subgoals
do not derive from broad policy goals, such as increasing
profit or decreasing youth unemployment. Their formulation,
rather, seems to depend on the "knowledge, experience
and organizational environment of the decision maker...It
can also be influenced in subtle and not so subtle ways by
his self interest and power drives."20-2
While various lines of research on the local
implementation of federal social policies have been
productive in explaining many of the factors involved in the
disparity between policy intent and local outputs, the
continued existence of slippage suggests that the remedies
implied are either insufficient, or are not followed, or both.
Research on the relationships between individually held
subgoals and policy outputs seems to offer a promising and
relatively new approach to the continued problem of slippage.
D. FEDERAL YOUTH EMPLOYMENT POLICIES
This study compares and contrasts the implementation of
ten youth employment-education projects authorized by the
Youth Employment Demonstration Projects Act (YEDPA) and
administered by Youthwork, Inc.
Any number of federal, or, for that matter, state or
corporate, programs could have been selected for a study of
the relationships among outputs, slippage and subgoals.
Federal youth employment policies in general, and Youthwork
projects in particular, have several advantages.
First, the problem of youth unemployment is serious and it
is not open to easy technological solutions. In 1981, over 16
percent of the white youths and over 35 percent of the
minority youths were unable to find jobs.21-1
The unemployment rate reflects lack of basic skills,
knowledge of job availability, job experience and jobs themselves.21-2
No one action or set of actions has proven consistently to
alleviate the problem. Moreover, there is considerable debate
about the nature of the problem. Is youth unemployment a
problem because an unemployed youth lacks money? Has nothing
meaningful to do? May turn to crime and other anti-social
behavior? With ambiguous goals and lack of strong evidence
that a given service will help, it is unlikely that staff
will be able to make decisions about what services to offer
through a process of rational decision-making. It is more
likely that they will be motivated by their individually held
Second, the federal government has a long interest in
employment policies for youth. The Manpower Development and
Training Act of 1962 (MDTA), while primarily a retraining
program for workers displaced by technology, also supported
basic skills centers and the Job Corps, a residential program
for poor urban and rural youths.22-1
The Comprehensive Employment and Training Act (CETA)
consolidated a number of programs for youths and adults while
turning authority for running the programs over to prime sponsors.22-2 The Basic Educational
Opportunity Grants authorized by the Higher Education
Amendments of 1972 (now called the Pell Grants) gave funds to
low-income youths that could be used at training schools as
well as colleges and universities.22-3
The Youth Employment Demonstration Projects Act (YEDPA)
provided funds for five programs. These included the Youth
Incentive Entitlement Pilot Project (YIEPP) which was
designed to test the efficacy of guaranteeing work to youth
as an incentive for remaining in school. The Youth Employment
and Training Program (YETP) provided funds for education and
work experience for disadvantaged youths, and called for
formal cooperation between CETA prime sponsors and local
education agencies. The Act also continued the Job Corps and
an older Summer Youth Program (SYP).
Finally, the Act gave the Secretary of Labor discretionary
funds to study alternative means of providing education and
employment services. The Youthwork projects were supported by
these latter funds.23-1
The above litany of federal involvement in employment
services is not intended to be exhaustive. It is rather
intended to suggest that it is likely the federal government
will continue to have some interest in supporting federal
youth programs with more or less specific intended outcomes
Third, the system that provides youth with federally
supported employment-related services epitomizes the concept
of "loose coupling." Federal funds flow to prime
sponsors, usually through the states, and are then contracted
to various local agencies, such as schools and community-
based organizations. Decisions must be made at all levels.
Schools and community-based organizations, as well as private
sector employers, who give jobs, are subject to numerous
policies and regulations and have many goals, only some, if
any, of which are related to youth employment. The prime
sponsor has no authority over these agencies, except insofar
as they accept funds.
Fourth, the Youthwork projects provide a natural
laboratory to study how different organizations implement the
same policy. Youthwork -- a non-profit organization
established to help the Department of Labor administer YEDPA
-- gave forty-eight awards to schools, community based
organizations, private employers, and CETA prime sponsors to
test alternative methods of providing employment related
services to youths. As is discussed in detail in Chapter III,
all awardees were to offer some combination of basic skills
instruction, academic credit for work experience, and job
experiences. In addition, they were free to select other
services to be offered. All were required to ensure that
federal funds were spent only on economically disadvantaged
Finally, and not of little importance, data were available
for the study. As part of its "knowledge development
effort," Youthwork awarded a contract to Cornell
University, under the direction of Dr. Ray Rist, to conduct
case studies of each site. The studies were conducted over a
period of a year. Detailed field notes were kept, as were
copies of relevant documents. These data were available for
E. RESEARCH ASSERTIONS
Because this study is exploratory and seeks to develop a
model for further testing, rather than testing a developed
theory, it is guided by research assertions24-2
rather than technical hypotheses. Research assertions are
what might be considered "best guesses" about what
is expected. The assertions that guided this study were:
1. Persons with different educational and professional
backgrounds have different individual subgoals.
2. Staff members' individually held professional,
organizational and personal subgoals are one of the
explanations of outputs of social programs.
3. The extent to which professional subgoals influence
output may be strongly limited by personal subgoals.
4. The strength of the relationship between professional
subgoals and outputs is influenced by the community,
scarcity of resources, and the focus of the project.
5. By considering the subgoals of the front-line service
deliverer, higher-level authorities can reduce the
likelihood of slippage.
F. RESEARCH PROCEDURES
While this study involves a secondary analysis of
qualitative data, the procedures used are essentially similar
to those used by qualitative researchers in general. These
include defining the problem, reviewing the related
literature, formulating hypotheses or research assertions,
developing data collection instruments, collecting and
analyzing the data, and drawing conclusions and
For the purposes of this research, slippage was limited to
the difference between the levels of service outputs
specified in the contract between the project and Youthwork
and the actual output levels delivered. A sample of ten cases
was selected from a database consisting of forty- eight
Cornell University/Youthwork case studies.
Subgoal levels were determined using both open- and closed-ended
questions that were asked about participants in the Youthwork
ethnographic protocols. The relationships between these
subgoals and outputs were analyzed in terms of three broad
concepts: scarcity, community, and focus.
G. RESEARCH LIMITATIONS AND ADVANTAGES
Secondary analysis of qualitative, multi-site case-study
material is relatively new. The related methodologies are
discussed in detail in Chapter III. It should be noted here,
however, that there are distinct advantages and disadvantages
to such analyses. As with other types of qualitative research,
statistical generalization to a larger population is not
possible. Because such a study depends on data collected by
other researchers, who may have been guided by somewhat
different conceptual designs, certain important data may not
On the positive side, the use of existing qualitative data
is cost effective, and permits a single researcher the
opportunity to examine data from several sites. The closer
the hypotheses or research conceptualizations guiding the
studies are to each other, the more likely it is that the
needed data will be available. Because of similarities in
conceptual design and purpose between the Cornell University/
Youthwork studies and this one, most needed data were
H. RESEARCH SIGNIFICANCE
As has been suggested and is discussed further in Chapters
II and III, much has been done to demonstrate the conditions
under which slip- page at the local level is likely to occur.
These include vagueness of goals, loosely coupled systems,
layers of decision-makers and decision points, and scarcity
of resources. Relatively little is known about the mechanisms
that link these factors to outputs. By examining the idea
that the professional, organizational and personal subgoals
of street-level bureaucrats may have a strong influence on
final outputs, this study may contribute to an understanding
of these mechanisms.
The findings of this research may be significant for
policy development in several ways. This study suggests that
considering the subgoals of staff might lead to greater
congruence between policy intent and outputs. Staff, whose
subgoals are similar to the subgoals of the policy, are more
likely to deliver services that are consistent with the
policy. Where personal subgoals are reasonably fulfilled, the
study suggests that more attention will be paid to
professional subgoals and, consequently, outputs should
This study suggests that, if slippage is to be minimized,
policy-makers and administrators should consider the extent
to which the policy contributes to -- or detracts from -- the
individually held subgoals of street-level service deliverers.
The study also suggests that there is an interaction between
the professional subgoals that staff bring to a project and
the project's subgoals. That is, over time, the professional
subgoals of staff may become modified in reaction to the
project. One of the implications of the study is that care
should be taken in designing policies and administrative
procedures so that subgoals are reinforced.
Implications of this study for methodology are in two
areas. The first involves the use of secondary analysis of
qualitative data and the second involves federal policies
concerning social research. Secondary analysis of qualitative
data is a new field. This study explores one method for such
analysis. More important, perhaps, it suggests that secondary
analysis in general may be an effective and efficient means
of enhancing the use of qualitative data.
Given the current costs of supporting case-study research,
federal and private research agencies might consider adopting
requirements that encourage secondary analysis. Such
procedures could be as limited as requiring that field notes
be made available to other researchers, as was required by
This chapter introduced the idea that slippage is a major
problem for higher-level policy-makers, particularly those in
loosely coupled systems. Evidence was offered that it is a
serious problem for federal social policy-makers. It has
suggested that the nature of social service agencies and the
types of services they provide make it difficult to reduce
slippage by increasing regulations. In loosely coupled
systems, street-level bureaucrats have considerable
discretion in implementing policies. This suggests that,
since outputs are the immediate and direct results of actions
by street-level bureaucrats, an analysis of the reasons that
such service providers do what they do may contribute to an
increased understanding of the mechanisms that lead to
slippage. Individually held professional, organizational, and
personal subgoals are suggested as one of the explanations of
the behaviors of staff in social service agencies.
The reasons for selecting youth employment policies for
analysis in this study were discussed. More detailed analysis
follows in Chapter III.
Table of Contents |
Top | Search
Chapter I: Footnotes
4-1. Herbert A. Simon, "Rational Decision-Making in
Business Organizations," American Economic Review 69
(September, 1979): 493-513.
6-1. Robert K. Yin, Karen A. Heald, and Mary E. Vogel,
Tinkering With the System (Lexington, Mass.: Lexington
7-1. John P. Kotter and Leonard A.
Strategies for Change," Harvard Business Review
57 (March-April, 1979): 106-113.
7-2. For an often cited essay on the nature of loosely
coupled systems, see Karl E. Weick, "Educational
Organizations as Loosely Coupled Systems," Administrative
Science Quarterly 21 (March, 1976): 1-19.
7-3. Margaret Davis, The Structure of Educational
Systems: Explorations in the Theory of Loosely Coupled
Organizations (Stanford: Center for Research and
8-1. For a comprehensive review of the policies and laws
passed during the decade, see Henry A. Levin, "A Decade
of Policy Development in Improving Education and Training
Programs for Low-1ncome Populations," in A Decade of
Federal Anti-Poverty Policy: Achievement, Failures and
Lessons, ed. Richard Haverman (New York: Academic Press,
1977), pp. 123-89.
8-2. P.L. 89-110.
8-3. For example, of the $15 billion the federal
government spent on elementary and secondary education during
fiscal year 1980, only $4 billion was administered directly
by the federal government in Department of Defense Schools
and on Indian reservations. Victor Miller, "Federal
Support for Education in Fiscal Year 1980," A special
analysis, prepared for the U.S. Department of Education,
Office of Planning, Evaluation and Budget (Washington, D.C.: n.p., October, 1981). Mimeograph
9-1. Condition of Education 1981 (Washington, D.C.:
National Center for Educational Statistics, 1981), Chart 2.27,
9-2. Ibid., Chart 5.14, p. 256.
9-3. Paul Berman and Milbrey Wallin McLaughlin, Federal
Program Supporting Educational Change Vol. VIII: Implementing
and Sustaining Innovation (Santa Monica: The Rand
Corporation, 1978): p. 37.
10-1. Stanley Helgeson, Robert E. Stake, Iris Weiss, et al.,
The Status of Pre-College Science, Mathematics, and
Social Studies Educational Practice in U.S. Schools: An
Overview and Summary of Three Studies (Athens, Ohio:
Ohio University and the University of Illinois, 1978).
10-2. Administration of Compensatory Education (Washington,
D.C.: National Institute of Education, 1977), p. v.
10-3. For a discussion of the tightening of Title I of
ESEA see Michael Kirst and Richard Jung, "The Utility of
a Longitudinal Approach in Assessing Implementation: A
Thirteen Year View of Title I, ESEA," Education and
Policy Analysis 2 (September-October, 1980): 17-34.
11-1. ESEA Title 1, the Pell Grants (The Higher
Education Amendments of 1972, and the National School
Lunch Act (P.L. 79-326), for example, require recipients
have family incomes below a certain point.
12-1. The Youth Employment Demonstration Projects Act
(YEDPA), (P.L. 95-93) required that 22 percent of funds be
spent on in-school youth pursuant to agreement with local
education agencies. The Higher Education Amendments of
1976 required that 10 percent of the funds spent on
teacher centers be set aside for higher education
12-2. Required by ESEA Title 1.
12-3. Required by ESEA Title 1.
12-4. For a current review of the effects of Title I see H.
Carl Haywood "Compensatory Education," Peabody
Journal of Education 59 (July,1982):272-300.
12-5. Mary E. Vogel, "The Distributive and Fiscal
Effects of Grants Consolidation on States and School
Districts," in Grants Consolidation: A New Balance
in Federal Aid to Education, eds. C. Phillip Kearney and
Elizabeth VanderPutten (Washington, D.C.: Institute for
Educational Leadership, The George Washington University,
1980), pp. 95-112.
13-1. This was one of the stated purposes of the "twenty-two
percent set-aside" of the Youth Employment Demonstration
Projects Act. The Vocational Education Amendments of 1976
(P.L. 94-482) and the Comprehensive Employment and
Training Act (P.L. 93-203) required that activities
funded by these programs involve at least statewide
13-2. Perhaps the most detailed planning requirements of
any social service program were in the Vocational
Education Amendments of 1976 (P.L. 94-482). States were
required to plan according to economic conditions needs of
students, and participation of women and minorities.
13-3. The Education of the Handicapped Act (P.L.
94-142) required that an individualized education plan be
developed for each child in cooperation with parents and,
where appropriate, other social service agencies.
13-4. The Vocational Education Amendments required that
sex equity coordinators sign off on state plans.
13-5 Vocational Education Study: Final Report (Washington,
D.C.: National Institute of Education, 1981), IV-41.
14-1. U.S. Congress, House Committee on Education and
Labor, The Vocational Education and National Institute of
Education Amendments of 1976, H.R. 1085, 94th Cong., 2nd Sess., May 4, 1976, p. 17.
14-2. Joseph Coleman and Gregory Wurzburg, Involving
Schools In Employment Programs for Youth (Washington, D.C.:
National Council on Employment Policy, 1979).
14-3. Samuel Halperin, "Is the Federal Government
Taking Over Education?", Compact (Summer, 1976),
15-1. Nathan, for example, argues that revenue sharing
funds were used often for tax reduction rather than increased
services, and benefited general purpose governments with
elected officials over special service units designed to
provide specific services. See, for example, Richard P.
Nathan, Where Have All the Dollars Gone? (Washington,
D.C.: National Institute of Law Enforcement and Criminal
Justice, 1976), and Richard P. Nathan, The Record of the
New Federalism: What It Means for the Nation.
17-1. Stephen K. Bailey, Congress Makes A Law (Syracuse:
Syracuse University Press, 1965).
17-2. Stephen K. Bailey and Edith Mosher, ESEA: The
Office of Education Administers A Law (Syracuse:
Syracuse University Press, 1968).
18-1. This typology was developed by March and Simon. See
James G. March and Herbert A. Simon, Organizations (New
York: John Wiley, 1958).
18-2. 1n these organizations, workers are viewed as "units"
to be managed, and it is management's task to make the system
18-3. 1n these organizations, it is management's job to
instill pride in the company in staff, to maintain morale,
and to develop a sense of comradeship.
18-4. 1n these organizations, staff are sometimes
independent of management efforts. Management must influence
the way decisions are made.
18-5. Karl E. Weick, "Educational Organizations as
Loosely Coupled Systems." Administrative Science
Quarterly 21(March, 1976): 1-19.
19-1. See, for example, Michael Lipsky, "Toward a
Theory of Street-Level Bureaucracy," in Theoretical
Perspectives On Urban Politics, ed. Willis B. Hawley (Englewood
Cliffs: Prentice Hall, 1976) pp. 196-213.
19-3. Richard F. Elmore, Complexity and Control: What
Legislators and Administrators Can Do About Implementation
(Washington, D.C.: National Institute of Education, 1980).
20-1. Simon, "Rational Decision-Making," p. 511.
20-2. 1bid., pp. 511-512.
21-1. Economic Report of the President (Washington,
D.C.: U.S. Government Printing Office, 1981), p. 269.
21-2. For a discussion of the general causes of youth
unemployment, see Report of the Vice-President's Task
Force on Youth Unemployment (Washington, D.C.: U.S.
Government Printing Office, 1980).
22-1. P.L. 87-51.
22-2. p.L. 93-203.
22-3. p.L. 92-318, Sec. 401.
23-1. P.L. 95-93.
24-1. The procedures followed by the Cornell researchers
will be discussed in detail in Chapter III. For a brief
review of the methods see Ray C. Rist, "Qualitative
Policy Analysis: A Case Study of an Emergent Methodology,"
A Paper Presented at the American Sociological Associations
Annual Meeting, New York, August 30, 1980.
24-2. The use of research assertions in case study
research, particularly for evaluation purposes, is rather
common. For a discussion, see Jerome T. Murphy, Getting
the Facts: A Fieldwork Guide for Evaluators and Policy
Analysts (Santa Monica: Goodyear Publishing Co., 1980).
| Table of Contents
of Page |