alternatives, implementation, the problem, criterion, implementation plan, Tradecraft, writing workshops, Policy Analysis, supporting detail, memo, consequences, clients need, Lydia Thornton, Bob Myrtle School of Policy, Juliet Musso Bob Biller, Gerald Caiden, the writing process, recommended option, George Orwell, implementation schedule, problem statement, brilliant analysis, Carmen Rogers, punch lines, analytical writing, Bardach, explicit goals, policy outcome, Design alternatives, educational outcomes, management problems, status quo, Winners and losers, decision criteria, feasible options, University of Southern California, professional writing skills, fountain street parking, Arbitrary values, professional writing
The Tradecraft of Writing for Policy Analysis and Management1 Juliet Musso Bob Biller Bob Myrtle School of Policy, Planning, and Development University of southern California
February, 1999 1 We are grateful for useful suggestions provided by Yan Tang, Gerald Caiden, Lydia Thornton, and Carmen Rogers. We also thank the many participants in our writing workshops, whose insights, creative approaches to writing, and constructive comments inspired and improved this manual. 1
The Tradecraft of Writing for Policy Analysis and Management In our time it is broadly true that political writing is bad writing.... The great enemy of clear language is insincerity. When there is a gap between one's real and one's declared aims, one turns as it were instinctively to long words and exhausted idioms, like a cuttlefish squirting out ink... -- George Orwell
: Politics and the English Language
Introduction: Teaching "Tradecraft" The following is a "stand alone" manual, intended to assist students in the practice of professional writing skills
. Several premises inform our design of this manual: decision makers
value good writing. Inept writers lose credibility. The activity of writing is excruciating, but becomes less so with practice. A "how-to" manual can assist students in the stressful processes of writing and revision. In designing the manual, we drew in part from a small group of very useful works on policy analysis and writing. (Bardach, 1996; Krieger, 1988; Lynn, 1980; Majone, 1980, 1989; Meltsner, 1980; Weimer and Vining, 1992) We also called upon our professional and teaching practice, to articulate a model of analytical writing that "like a Bauhaus tool or instrument, has form and function nicely congruent." (Krieger, 1988, p. 408) Thus, the manual speaks to analysis and argumentation, process and product. It should not be used in a "paint by numbers" fashion, but rather to help writers practice the logic of argumentation, and to show them that analysis and communication constitute a single endeavor. (Meltsner, 1980; Majone, 1989) The manual was tested and revised during two experimental writing workshops led in 1996-97 at the School of public administration
, University of Southern California.2 Our goal was the development 2 Students in all Masters' degree program
s were invited to take part in a free, no-credit workshop. In 1998, the School of Public Administration and the School of Planning and Urban Development joined to form a new School of Policy, Planning, and Development.
of "knowledge-in-action," skills that "are dynamic and changing, acquired through participation, and oriented toward present and future rather than past. In learning... an individual is taking on a dynamic set of tools for being in and making sense of the world." (Applebee, 1995, p. 20) Each workshop met for three full days over two months and employed these key exercises: · Small group discussion of the "Tradecraft" manual. Students discussed assigned sections, and reported insights and suggestions for change. · In-class writing. Students wrote one-page memos on a simple news article. In small groups, they reviewed the memos and reported on the writing process
and results. · Iterative memo assignment. Students wrote a short memo at home, using one of two cases. Students and instructors reviewed the memos, and prepared written comments for discussion in the next meeting. Students revised in response to comments. Student evaluations of the workshops were uniformly positive; however, both series experienced substantial attrition.3 Students reported difficulty balancing "optional" writing work with other class and work demands. We subsequently substituted a writing component in each of the introductory Masters' degree courses.4 Faculty in these courses use the manual, and assign at least one client-oriented memo that is revised in response to peer and/or instructor review. We have not formally evaluated this approach, but faculty have reported good results, and informal student feedback has been positive. While "Tradecraft" can be integrated into a variety of curricular approaches, we believe that a key to success is the revision of papers in response to detailed peer and instructor comment. Also important is a willingness to coach students who experience anxiety, writing block or other difficulties with the 3 Around 60 students participated; with around 25 finishing an entire seminar. 4 The manual is also now used in Core Course
s for our undergraduate degree program, with reported satisfaction by students and faculty. 2
process (Krieger 1988). Such investment in training writers is labor-intensive for students and faculty alike. We think that the Labor Costs
are justified, that the process produces better writers, and that it helps new public affairs professionals endure the editorial review process employed within most organizations. We welcome comments from faculty who find new ways of integrating the manual into their curriculum. 3
The Tradecraft of Writing for Policy Analysis and Management I. Why should I bother to read this manual? You should read this manual because it will make you a more successful public policy
analyst or manager. The most important skill anyone can take to a job is the ability to write clearly and persuasively. Bad writing is long-winded, pointless or factless, studded with technical jargon or tongue-tied with passive constructions. Good writing is decisive, compelling and brief. If you write well, and sincerely, you will get a well-deserved reputation as someone who "speaks truth to power."5 Your rewards will be political credibility, interesting assignments, and the opportunity to influence public affairs. In this manual, we discuss how to develop "action memoranda," which are decision- oriented, analytic papers directed toward a specific audience. Your measure of success is the extent to which your audience acts upon your advice or directions. In turn, success depends heavily on analytic quality, the way you frame the issue, reason it through, and support your reasoning with strong facts and examples. Research, analysis, and writing constitute a joint endeavor to express judgment, backed by evidence. 5 The authors are much indebted to Wildavsky (1987) for his vision of policy analysis as the Art and Craft
of analyzing "public problems that must be solved at least tentatively to be understood." (p. 15) 4
II. What is "professional writing?" To write professionally is to engage in problem solving
activity. It is to clarify the problem, design alternative means of improving the situation, and explore the Advantages and Disadvantages
of each important option. Good professional writing is relevant, client-oriented, action-oriented, and analytic. It is inextricably linked to your ability to frame an issue, gather and analyze information, and structure a helpful response. Client orientation: Do not write for God, write for your client. You cannot cover all possible effects on involved parties, all values, or every conceivable alternative. Frame your paper with attention to the jurisdiction of your client, and the institutional and political limitations of his or her position. Measure the costs and consequences of particular policies against the singular values of your client. You will thereby establish a basis upon which values can be triangulated by your client and other readers. Action: Do not tell your clients what you know, tell them what they need to do. Do not write for "understanding" alone. A policy memo or issue brief should reveal how better understanding is to be expressed in action. Even if you cannot make a specific recommendation, clarify the tradeoffs, values, or risks that your client should consider in making a decision. Tell your clients what they might do next, even if the next thing to do is to take responsibility for a difficult choice. Your role: Do not define yourself by what you know, define yourself by what you can find out. Information is always of uncertain quality, transient and situational. You do not want your self-definition to depend on what you know about an issue at any particular time. You write good policy papers not because of what you know when you start, but because of what you can find out in the process of writing them. Honesty: Speak in your own words, and give credit where due. Seek originality in your writing, and do not rely excessively on quotations or paraphrasing of other work. State arguments in your own 5
words, and above all, avoid plagiarism! Follow standard rules of attribution when you rely on the ideas or words of another. III. Tell me more about how I can engage usefully in problem solving... Policy and management problems are often "wicked" in that the "cause" is neither obvious nor agreed upon, politics are powerful, and the future is risky or unpredictable. (Rittel, Horst, and Weber, 1973) In such a setting, the most important professional skill is not salesmanship, but design, finding an "opportunity for improvement" (Dery, 1984), something that might work better than the status quo. You must clearly communicate the costs and consequences that you expect to follow from each option, including the current policy. Analysis and writing are an inseparable process through which you (1) frame the issue, to clarify and constrain your identification of the problem; (2) articulate the goals your client is trying to achieve; (3) design feasible options; and (4) join theory and evidence to judge the likely consequences of different actions.6 Frame the issue: Disentangle which claims of harm (or missed opportunities) the client might have power to address. As Bardach (1996) describes, many "problems" come wrapped in "issue rhetoric," the "raw material" of political discourse. One useful way to clarify the problem is to identify which individuals (human or otherwise) may be hurt by the current situation. Then consider the match between this "claim of harm" and the jurisdiction within which your client has authority to act. Of particular importance are the following: · Pay attention to jurisdiction. Do not frame your issue so broadly that you develop "options" which your client is powerless to influence. · Do not preordain one solution. Do not define a problem that is actually a solution in disguise. (Bardach, 1996) For example, the problem definition, "classes are too big" can only be solved by 6 Bardach (1996) provides an excellent conceptualization of an "eight-step path" for working through this analytic process. 6
reducing class size, ignoring the possibility that improved teaching or an improved curriculum might
improve educational outcomes at less cost. The problem definition, "students are not learning
enough," encourages more open-minded exploration of alternatives.
Express explicit goals and values: Be overt about the values that structure your analysis. If
you judge potential outcomes using explicit criteria, you will be more likely to confront all of the strengths
and weaknesses of each option. Moreover, well-defined analytic criteria facilitate understanding of
tradeoffs and allow different readers or stakeholders to reevaluate your findings based on their values. In
identifying values, consider your client's perspective on the following questions:
What principles should guide government intervention? Only when the market fails? Any time a
group is suffering harm?
What standards should distinguish "better" from "worse" policies? Costs? Effectiveness?
Efficiency? Fairness? How do you--and your client--define values?
What are the appropriate measures of these standards? Inputs? Outputs? Outcomes?
It is not that you must believe equally in the values of your client, but rather, that you be clear
about how those values are affected by the alternatives you are describing. It is not that your own values
are not relevant. They are. However, if you wish to demur, or suggest an alternative, do so directly rather
than by indirect means or shaded language. When you and your client are in essential value agreement,
you need to seek value dissent from opponents to help clarify your understanding. When you and your
client are in essential value disagreement, you need to recognize and act on the potentials for voice
(engaging the client in debate) or exit (resignation). (Weimer and Vining, 1992; Hirschman, 1970)
Design for implementation: Design alternatives that are legal, politically feasible, and that will
be implemented in the world. Above all, do not commit the cardinal sin of comparing a deeply flawed
status quo to an ideal alternative that works only in theory. Be clear about the behavioral or organizational assumptions required for your recommendation to succeed. Do not imagine that thinking about implementation is something you do after a policy is identified, analyzed, or adopted. If you first devise what should be done, leaving until later consideration of how, you will almost certainly fail. Focus on implementation at the outset. Ask, as you define the problem and consider how to fix it: "If this is a good idea, how could we accomplish it?" A successful policy change always involves the negotiation of new sets of relationships, and focusing on implementation allows you to discover how those relationships might be forged. Always consider the status quo as a baseline option against which you judge other alternatives. Ask yourself, what will happen if no action is taken? Will the situation get better by itself? Will it get worse? Question theory and evidence: Be aware of the behavioral theories that structure your arguments, and support your conclusions with evidence. In doing so, cultivate the possibility that your theory is wrong, or that the evidence is faulty. Entertain counter-arguments with an open mind, consider the sources of your information with a fair amount of cynicism, and be very explicit about possible weaknesses in the assumptions underlying your analysis. IV. How will I ever find the answer? Clarify the tradeoffs: Analysis is the art of designing plausible improvements. Do not imagine that you will discover a "best" alternative to recommend. You use the major resources available to you (time, information, and analytic reasoning) to learn your way into a better solution to a problem or issue. You want to learn as much as possible about probable outcomes rather than to depend on pure speculation. 8
You are not a calculator that spits out a single answer while disguising the decision circuitry and
intermediate calculations within gray plastic. Be transparent about the process that you used to weigh
alternatives, and about any uncertainty you feel. Support your judgment with:
a clear and richly descriptive statement of likely outcomes and risk factors
an honest reckoning of the extent to which a choice involves value tradeoffs.
Help your clients understand what they are giving up in order to get something else. Your
responsibility in the analysis is to compare the consequences of all relevant alternatives against a common
set of criteria. Do not apply criteria selectively in order to bias judgment in favor of your preferred
Present an alternative-criterion matrix: A useful tool for summarizing argumentation is the
alternative-criterion matrix, which tabulates your evaluation of the likely consequences associated with
each alternative. Table 1 exemplifies a matrix for the placement of an off-leash dog park. Notice that the
agency criteria are shown as column headings, and the options for sites are row headings. The criteria are
the cost of creating and maintaining the park; its suitability for off-leash use (size, interesting things for
dogs to sniff, lots of trees, water); and spillover effects on neighbors (noise, traffic, parking congestion).
The alternative-criterion matrix should communicate in "natural units" or "word pictures." Each
cell of the alternative-criterion matrix should tell a richly descriptive story about how you expect a policy
outcome to measure up against a particular criterion. Arbitrary values such as "high/medium/low" or
"good/bad" do not convey sufficient information to support reasoned debate about tradeoffs.
Table 1: Dog Park Alternative-Criterion Matrix
Desirable Featuresnegative effects
4 wooded acres; small stream; natural pond
100 feet from homes; limited parking
Land's End $150,000, plus
1 acre, muddy in winter, 1 mile from homes; parking lot,
$10,000 to $30,000 few shade trees, drinking left-turn lane eases traffic
Splitsville $200,000 plus
1/2 acre, very steep slope 150 feet from homes with
$20,000 to $30,000 with erosion; no shade or history of noise complaints;
No Exit $100,000
1 wooded acre with duck South fence adjoins 20 homes;
pond; drinking fountains, residents vocally oppose park;
1/2 acre, rolling hills,
Entrance 400 feet from "dog-
some trees, ornamental friendly" commercial uses; off-
Note: Highlighted cells indicate best performance on respective criteria.
Compare consequences: The comparison of options is usually complicated by immeasurable
and conflicting goals. Consider using the following methods to communicate the alternative-criteria matrix:
Highlight extremes. Help the client see the range of costs and benefits
, by highlighting cells to
indicate particularly "good" or "bad" performance. (Playland is the most suitable, while No Exit is
the least expensive).
Winners and losers. Tell the client about alternatives that are clear "winners," better on each
criterion than all other options, or "losers," worse on each criterion than at least one option.
(Splitsville is a "loser" relative to Land's End).
· Must-meet goals. Highlight performance on "must-meet" or "primary" criteria. This will help
your client to filter alternatives that may be on the table, but which do not meet primary
objectives. Practical criteria, legality, political feasibility, and implementability (Bardach 1996),
should be considered "must-meet" criteria.
Seek better, not best: Do not imagine that you must design a clear "winner" that leads to an
"optimal" outcome. The attempt may immobilize you. Concentrate on improving the situation, in hope
that improvements accumulate into substantial gains.
IV. So, the first step is Data Collection
Wrong. Research, analysis, and writing are not sequentially ordered, separate steps. Think and
write before starting research, and continually modify your product. Discuss what you are doing with
colleagues, and seek constructive criticism of drafts. From the beginning, plan for deadlines. A close
deadline of necessity limits the complexity of your analysis.
Work backwards rather than forwards: Writing and thinking should precede information
gathering, not the reverse. Most people are in the habit of immersing themselves in library research, taking
copious notes on reading, and then attempting to organize these notes into a paper outline. Only at the end do they think about the point they are trying to make. This is a waste of time. Before you do any research, prepare a "back-of-the-envelope" analysis. Rough out a tentative framing of the issue, a set of decision criteria, and a sketch of options for addressing the problem. Rely initially on your "ordinary knowledge" (Lindblom, 1974), conventional wisdom, common sense, and basic facts. Clarify what you need to find out, and discuss the initial cut with the client or a colleague before you invest heavily in research. Sketch a full first-draft policy memo as early as you possibly can. Try to review this draft with your client (on the same day if possible). As you continue the research, revise your findings and fill in missing information. Write simultaneously rather than sequentially: Do not try to get your facts straight before you identify your conclusions. Instead, write something on all elements of your outline, even if you use potential alternatives and dummy data. Prepare as many iterative drafts as the available time permits. Attend to time: You will never have enough time to do an analysis "perfectly." Any "small" issue could take an infinite amount of time if you had it, and a "large" issue may require a nearly immediate decision. Sometimes your only chance to improve the quality of the decision may be a "quickand-dirty" analysis. You must understand the time frame of your client, and write in a way that responds to that time frame. A late analysis is worse than no analysis (because it cannot affect the outcome, but you have spent resources on it). A timely but imperfect analysis that improves the quality of decision is almost always better than no analysis at all. If time is short, be sure to identify gaps and questions that you could fill in if you had more time. 12
Use multiple sources: Do not become mired in a research rut. Use multiple sources of information. Become facile at moving between government data sets, interviews, policy reports, and Social science research
.7 V. How can I build credibility? Avoid straw men: Do not frame any "straw man" alternatives. If an alternative is not credible, it should not be in your analysis. The exception is the status quo. Even if you do not consider the status quo an acceptable outcome, analyzing it allows you to demonstrate its costs and consequences. Avoid surprises: Do not try to "sell" an alternative simply by demonstrating its good points. As you explore the pros and cons of the alternatives, describe clearly and specifically the disadvantages of your recommended alternative. Also clearly display the advantages of the status quo. By these means you really demonstrate that you are trying to meet a no-surprise test as far as your clients are concerned. Your biggest risk as a policy writer is not to have your recommendations ignored (though that is not an acceptable professional standard over the long haul). It is to have your client follow your advice, only to be surprised by a negative consequence that you concealed. Pay early: Do not defer engagement with opposition to an action that you are considering. Opponents are always the best teachers, because they have an incentive to look hard for what is wrong with your proposal. Seek out collegial working relationships with opponents, and exchange drafts that clarify differences and alternatives as early as possible (the only time you can do something about them). While you do not have a choice about whether you "pay" to overcome opposition, you can choose whether to pay early or late. Paying early tends to present smaller bills spread out over a more reasonable payment period, than does paying late (after the recommendation is fully set). 7 See Bardach's (1996) discussion of "the strategic aspects of policy research." (p. 83) 13
Build support: Do not imagine that you will be able to build support after you complete a policy
paper. Build support as you design plausible alternatives and craft an implementation plan. One reason
for an iterative drafting process (i.e., the circulation of successive drafts, improving as you go) is that you
can put the following kinds of questions to key stakeholders:
Does this disentangle our goals and concerns on this issue?
Is the right alternative being recommended, and is it shaped about right?
Can you see ways to improve this outcome?
Can you support this outcome?
Can you do what the implementation schedule suggests would be needed?
Proofread your work: And then proofread it again. Remember to double-check that your tables
sum up properly. Minor errors can undermine the credibility of the most brilliant analysis.
VI. I'll never be able to organize this stuff! Build a pyramid: Your argument should be structured like a pyramid, with the point on top, and supporting detail spreading out below. The subject line should describe your recommendation. The first section should be an executive summary of the major issue and your conclusions or recommendations. You want to ensure that the reader understands the recommendation after a thirty second reading. The end of the memo should tell your clients what action steps they, and others, will need to take. Fit lock and key: Do not imagine that your completed analysis is a series of separate "things" such as a memo title, a Problem Statement
, the comparison of some alternatives of which one is recommended, and an implementation plan for that recommendation outlined. All of these elements, together, should constitute the same thing. The problem statement, for example, poses a question that the recommended alternative answers. The implementation sequence clarifies what would have to be done (and by whom and when) to make that recommendation real. The subject line of the memo is a description of the recommended alternative. This "lock and key" test may suggest ways of understanding a problem differently at the end of an analysis (once you have figured out how to make progress in handling it). When you have a final implementation sequence, review the memo to ensure that the implementation plan gives form to your recommendation, that the recommendation improves the problem you define, and that the subject line accurately describes the recommendation. Select supporting detail: Make supporting documentation digestible by presenting it selectively and visually, and guiding the reader through it. Use charts and graphs to illustrate or emphasize important points. Delegate technical detail to appendices. 15
Avoid extraneous information. It is not neutral! It imposes costs! If one or two reasons for your
recommendation are sufficient, keep the other reasons in reserve. Resist the impulse to include
everything you know about the topic.
Do not allow contextual detail to diffuse the momentum. A lengthy contextual section may merely
slow the reader down: Consider whether you can integrate contextual detail into the problem
definition, or the analysis of options.
Select appendices wisely. Only include information that is essential to deepen understanding of
what you are recommending, and why. This will almost always involve a selective draw from data
you have found or created. Simplify the format to allow its support of your analysis to emerge
easily and overtly.
Create punch lines: Help the clients to interpret appendices, charts and graphs. Include
highlighted "punch lines" at the bottom of an appendix page to state clearly what inference the data
support. Similarly, make certain that the point of each graph or table is evident at a glance, and that its
title states that point ("Figure 1: Budget deficit will grow fifteen percent over three years"). Design
section headings as punch lines that communicate the main point of each section ("Problem: Second-hand
smoke causes cancer").
Keep it simple and straightforward: Make your memo readily comprehensible to someone who
does not know or care about your organization and its problems. Use plain English, avoiding bureaucratic
jargon, and keep technical background in footnotes or appendices.
Document adequately: Always date and document your work. Keep a comprehensive list of
documentary and interview sources. Develop and extend this list as you continue to work the issue. As
you draw conclusions, be sure to document the source of information for future reference. Date each draft
so that you can identify the most current version.
Use a readable format: Organizations have preferred formats that should be respected and used.
These are the way that people in the organization expect to absorb and communicate information. In
some cases, however, you will have some discretion regarding the format of a writing product, and you
will need to a develop the format that best fits the problem at hand. A sample memorandum format is
attached as an Appendix.
Use bold section titles and aggressive indentations.
Use a reasonably large type font.
Use aggressive spacing to invite written comments.
Create wide margins on the side of your client's handedness, to allow comments.
Start a new page for each section.
VII. How do I increase the chance that my "clients" will act on my recommendation?
Plan for implementation: It is not enough to suggest that something is desirable. You must also
demonstrate how the desirable state of affairs
is to be created. What steps need to be taken, by whom,
and when? You do not have a responsibility to develop an implementation plan for all credible
alternatives, but you certainly must for your recommended option. Be sure your plan identifies each step
required, an action date, and the person responsible for accomplishing it. The clients' scan of an
implementation sequence will confirm whether they understand your recommendation.
Consider your clients' clientele: While you write for particular clients, remember that they are,
in turn, constrained by others. The clients' use of your work will be immeasurably aided if it is clear
enough to be shared. Have your clients' clients in mind as you finish a policy paper. For example,
include appendices that your clients may need to circulate as they seek counsel in deciding whether to follow your recommendation. Anticipate what the clients need to take action: Identify what your clients will ask for next, and provide a good draft of it in advance. For example, if you recommend amendment of a bill, provide suggested language. If you think the agency director should circulate a directive to staff, attach a draft of the directive. Add an action block: Do not allow your analysis to go easily into a "pending" file. Make it easy for your client to take some action on your paper immediately. Provide a short menu of important choices or actions that can be checked or initialed as a way of getting action implemented (even if that action is a return of the paper to you for a rework based on marginal notes from your client). Consider change strategies: Do not assume that the only change strategies you have are positive or "push" factors. Some of the most successful strategies employ negative or "pull" factors. For example, showing the negative costs and consequences of continuing with a present policy can be a powerful motivation for change, even if it does not reveal the strengths of a particular change itself. When confronted with a boulder you want to move, trying simply to push harder in a particular direction may be greatly helped by digging sand out in the direction you want the boulder to move. Delegitimizing an existing policy has a similar effect. IX. Can you provide the "top ten" tips on tradecraft? 1. Tell your clients what they need to do, not what you know. 2. Write before and as you research, and revise and revise, until you reach the deadline. 3. Be overt about your clients' values and your own. 4. Design for implementation. 5. Anticipate objections and identify tradeoffs. 18
6. Build support as you develop your analysis. 7. Begin with a summary (your conclusions and recommendations) and end with a beginning (implementation plan...action block). 8. Make the memo simple and interesting to read. 9. When appropriate, ignore tips 1-8. 10. Tell the truth! 19
Appendix: A Model Memorandum
(May use date of proposed release)
SUBJECT: "I recommend Action X to solve Problem Y."
The memo in a nutshell. One or two sentences that state the action, then the problem.
SUMMARY: "Why Action X will solve Problem Y, and key implementation needs."
An executive summary of the problem, the recommended action and its basis, and the key implementation
needs. In a very short paper with a good subject title, this is optional.
PROBLEM: "The `real' issue is Problem Y, which has these characteristics."
A brief description of how you "frame" the issue for your client. Identify how the status quo is in
discordance with major organizational or policy objectives. Consider claims of harm (or opportunities for
improvement) and evaluate the magnitude of the "problem."
CONTEXT: "These are some additional crucial facts about Problem Y."
A brief discussion of historical, political, and/or technical context that provides a deeper understanding of
the issue as you frame it. Only include information likely to be needed by the audience (clients and their
clients, other interested parties)
ALTERNATIVES: "This is my assessment of the likely outcomes of key options.
An alternative-criterion matrix, and a summary of advantages and disadvantages of the status quo, your
recommendation, and at least one credible alternative. In a short memorandum, the alternative-criterion
matrix may be attached as an appendix.
Option V Option W Option X
Table: Comparison of Options for Addressing Problem Y
(Detailed evaluation, using Criterion A, of Option V's likely outcome)
1. Headline/brief description, typically of "maintain status quo" Each option has a brief title and description. The first option is usually some version of "maintain status quo." Advantages: (brief bulleted items; good performance on alternative-criterion matrix) (effectiveness, benefits, political advantages, etc.) Disadvantages: (costs, risks, foregone benefits of continuing present policy) 2. Headline/brief description, typically of credible option(s) At least one credible but not recommended alternative, provided to triangulate the probable consequences of your recommendation against something else, and against the status quo. The set needs to be small to be understandable. Advantages: Disadvantages: 21
4. Headline/brief description, typically of recommended alternative Advantages: Disadvantages: (Be certain that this section is particularly strong, and will clearly meet the "no surprise" test if your recommendation is accepted) RECOMMENDATION: "I recommend Action X, because..." A summary comparison of alternatives, and justification of choice. Include an honest assessment of forgone benefits and any uncertainties which may affect outcomes. IMPLEMENTATION PLAN: "What people need to do, and when, to make X work." A sequence of steps, ordered in time, showing what needs to be done, by whom, and when, to realize the recommended alternative--typically one sentence each. ACTION: "These are possible next steps on my recommendation." Possible action steps to be initialed or checked. These may be contained in a highlighted box: See me; respond to comments; discuss with __ and revise; prepare implementing memo for my signature; no because... APPENDICES: "These additional findings support
my analysis because..." List numbers and titles of appendices at the front, and begin each appendix on a new page. Each appendix should have a punch line box at bottom that answers the question: what critical finding, necessary to the analysis, do these data support? Be selective. 22
References Applebee, Arthur N., 1996. Curriculum as Conversation: Transforming Traditions of Teaching and Learning. Chicago and London: The University of Chicago
Press. Bardach, Eugene, 1996. The Eight-Step Path of Policy Analysis (A Handbook for Practice). Berkeley Academic Press. Dery, David, 1984. Problem Definition in Policy Analysis, Lawrence, Kan.: University Press of Kansas. Hirschman, Albert O., 1970. Exit, Voice and Loyalty. Cambridge, MA: Harvard University
Press. Krieger, Martin H., 1988. "The Inner Game of Writing." Journal of Policy Analysis and Management, 7: 408-416. Lindblom, Charles E. and David K. Cohen, 1979. Usable Knowledge: Social Science and Social Problem Solving. New Haven and London: Yale University Press. Majone, Giandomenico, 1980. "An Anatomy of Pitfalls," in Majone and Quade, eds., Pitfalls of Analysis. NY: John Wiley and Sons
. ----- 1992. Evidence, Argument and Persuasion in the Policy Process. Yale University Press. Lynn, Lawrence, E., Jr. "The User's Perspective." in Majone and Quade, eds., Pitfalls of Analysis. NY: John Wiley and Sons. Meltsner, Arnold. 1980. "Don't Slight Communication: Some Problems of Analytical Practice." in Majone and Quade, eds., Pitfalls of Analysis. NY: John Wiley and Sons. Patton, Carl V. and Sawicki, David S., 1993. Basic Methods of Policy Analysis and Planning, 2nd ed. NY: Prentice-Hall. Rittel, Horst W.J. and M.M. Weber., 1973. "Dilemmas in a General Theory of Planning," Policy Sciences, 4: 155-169. 23
Stokey and Zeckhauser, 1978. A Primer for Policy Analysis. New York and London: W.W. Norton and Company. Weimer, David and Aidan Vining, 1992. Policy Analysis: Concepts and Practice, NY: Prentice Hall. Wildavsky, 1987. Speaking Truth to Power. New Brunswick, New Jersey
: Transaction Books. 24
J Musso, B Biller, B Myrtle