Supplementary Reading
Evaluation Procedures[4]
There is no one policy evaluation technique that may be learned and applied in all circumstances.Because a policy is like a theory，all the methodologies and techniques available to test theoretical hypotheses may be brought to bear in policy evaluations.Controlled social experiments can be run;statistical analyses using historical or comparative data can be undertaken;mathematical models can be built;citizens and policymakers can be interviewed;qualitative assessments can be made.Which methodology or set of methodologies should be used will vary with the situation and with the time evaluators are given to do their job.
In the end，learning to do sound policy evaluations is the same as learning to do sound research.The trick lies in putting together a sensible research design，in asking the right questions in the right way.But what does it mean to put together a sensible research design?From the foregoing discussion we can distill several steps:
1．Identify clearly the policy to be evaluated．This means understanding both the goal，or intent of the policy，and the instruments of policy.What was the policy supposed to accomplish?Precisely how was it to be done?
2．Devise valid measures．Restate policy goals and instruments in ways that can be measured.Remember that most policies—to attract support and minimize opposition—state goals only in verygeneral terms that cannot be measured directly.You must translate broad aims like“urban redevelopment”into measurable variables，such as occupancy rates，employment，and income.In doing so，make sure that your measures are valid.
3．Collect empirical data．For each of the measures of each of the variables in your policy hypothesis，collect the most reliable information you can find.This may mean sifting through government statistics or conducting interviews or other field research.
4．Analyze your data．Does there appear to be a causal relationship between the independent variable(the policy)and the dependent variable(s)(the outcome)?That is，is a change in the former variable associated with a change in the latter?
5．Consider competing theories．This is actually part of Step 4，but it is so important that it deserves special emphasis.In analyzing your data，consider whether the changes you observe could be attributed to other causes.Is the redevelopment we have seen a function of tax incentives or of something else?To answer this question，it will be necessary conceptually to start over at Step 1 and follow all the procedures outlined.Before you can say that a policy has worked，you must be prepared to reject all plausible alternative explanations.This is not always easy to do.(https://www.daowen.com)
6．Look for unintended consequences．Even though a policy may be aimed at one target，it may hit another accidentally.Urban enterprise zones may(or may not)stimulate economic activity，but they may(or may not)also displace poor people and contribute to the problem of homelessness，create new pockets of poverty as businesses shift areas of operation，reduce state and local revenues beyond levels anticipated，and so on.Sound policy evaluations look for unintended as well as intended consequences.
Policy evaluations can and should serve as valuable feedback to the policy system.We can learn what we are doing right and what we are doing wrong.
【注释】
[1]Source:Schultz，David ed.，Encyclopedia of Public Administration and Public Policy，New York:Facts On File，Inc.，2004，pp.350-351.
[2]Source:Excerpt from Jay M.Shafritz and E.W.Rusell，Introducing Public Administration(2nd Edition)，New York:Addison Wesley Longman，Inc.，2000，pp.49-58.
[3]efficiency:Competence as well as speed in performance.Americans have historically been suspicious of a too-efficient government，feeling that a truly efficient administration of public affairs could eventually eat into political liberties.
effectiveness:The extent to which an organization accomplishes some predetermined goal or objective;more recently，the overall performance of an organization from the viewpoint of some strategic constituency.
[4]Source:Huddleston，Mark W.and Dresang，Dennis L.，The Public Administration Workbook，New York:Pearson Education，Inc.，2007，pp.312-313.