Internal & External Evaluation Collaboration Questions

Dear Colleagues,

I was recently asked by a major client to provide them with the pre-test data set from an external evaluation of their program, just as I was concluding the post-test data collection. I was hired to provide an external evaluation (with contract). I am a small evaluation firm, a sole proprietor, and I did expect to provide them with all of the data after the conclusion of the study (after the evaluation reports are submitted and approved and final payment is received). I was hired to conduct the evaluation based on a federally funded grant that requires an evaluation, but not necessarily an external evaluation. The external evaluation was part of their accepted application to the federal agency. One other note, the director who hired me has recently resigned and the program coordinator with whom I have been working with now for over a year, was hired to replace the director.

The back story is that they planned to increase their research capacity, which I supported, and consequently, hired a very competent research analyst, who now has the time and competence to conduct their own internal evaluation. Without getting into the politics and personalities, which, I acknowledge plays an important part in this little drama, I am curious about the approach and opinions of other evaluators who may visit this blog.

We live in an urban environment in the Northeast where the poor outcomes of children in public education is a very serious matter, and the stakes are high for the program and the agency which runs the program. The agency has a very positive reputation in the community. With one year left on the evaluation, and with the limited structure of my research firm, I am tempted to "collaborate" (assuming I have a choice in this matter), and hand over the pre-test data set to the in-house evaluator for preliminary analyses. (There are many sub-evaluation questions we planned for in which in-house data and external data would be combined anyway.) But I also am obligated to protect the agency and program from potential criticism of the results of the study, in light of the challenge to objectivity that may emerge from an arrangement in which the in-house evaluator has prior access to pre-test data.

Assuming for a moment that the contract allowed or was passive on data access and integrity, what advice would you offer, if you were inclined to collaborate with your client. What advice would you offer if you were not so inclined? What safeguards or protections would you impose in a written agreement with your client so that some mutually beneficial compromise would emerge?

I am sending this post to the EVALTALK listserv, and later I will post some of the comments I receive.

Thank you in advance for your feedback.

David

Differentiating Program Evaluation Logic Terms

Thanks to EPA Administrator and blogger, Marcus Peacock, and the EVALTALK contributors, here is an interesting and useful way to distinguish between the terms used in many common forms of logic models (The Flow of the River, http://flowoftheriver.epa.gov/my_weblog/2007/10/getting-juiced-.html)

Nectarine Juice Production Program

Performance Measurement

• The inputs are nectarines and a hammer
• The activity is smashing nectarines with a hammer
• The output is how many nectarines got smashed
• The outcome is how much juice resulted
• The impact is the improved health of the people who drink nectarine juice

Program Evaluation

• How well is the program producing nectarine juice using the current inputs?
• Is a hammer really the most effective intervention for producing nectarine juice? What are the alternatives?
• What is the nectarine juice production program’s specific contribution to the improvement of national public health?

I'm not a devoted fan of juice, but the example helps me understand the way people may use the terms. The program "Juice Production Initiative" further illustrates that programs have both resources (objects) and activities (processes) to organize and make use of for a desired end.

Selecting an Evaluator - A Primer

Notes Regarding Professional Evaluation Work and Evaluation Reporting

David S. Robinson
www.evaluationhelp.com

If I were interested in employing an evaluator or evaluation firm, I would be interested in how responsive the evaluator has been to his clients with respect to evaluation needs.

Evaluator Checklist
• Can he/she complete an evaluation report in a timely manner?
• Is he/she respectful of the program and the complex environment or context in which it operates?
• Does he/she listen or is he single minded about the way evaluation must be done?
• Does he/she understand the concepts behind the program, the program theory or logical connections between the intervention and the outcomes of the intervention?
• Does he/she write well? Is he conscientious and detail oriented?
• Does he/she stay focused or wander in his questions and contributions?
• Does he/she search out seminal articles that are research oriented and share them with the program staff as appropriate?

Brief Background
I was the Director of Research and Evaluation for MSPCC for 24 years. I have been teaching research and practice evaluation since 1992 and I continue to assist the Simmons College School of Social Work in several ways. I have had a federal grant from the Head Start Bureau (1998 - 2001) and was the principal investigator on the study. I helped design the practice evaluation course at Simmons College. I consult to community organizations in the Boston area as part of the School of Social Work Research Institute. I started DSRobinson & Associates in 2004 and currently consult to medical training programs, public health initiatives, and child welfare programs in Massachusetts and Rhode Island.

My Evaluation Practice
I am trained in Empowerment Evaluation, and have been a member of the American Evaluation Association since 1990, and a current member of APA (American Psychological Association). Here are some points that I would consider noting about my experience and work.

1. Before beginning to write an evaluation plan, I tend to ask many questions, or to re-state what I have heard or read about the program or similar programs, always trying to clarify what the program director (or key leader) is attempting to do. I may offer up prior research articles in similar areas, or I may contribute program evaluation articles to clarify the kinds of things that evaluators do, define new terms, suggest articles or people working in the same or similar area. All of this preliminary work, which may take more than one meeting, is designed to explore with the director an evaluation approach that will be helpful, answer important questions, help strengthen or improve the program, contribute to scientific knowledge if the field, and be a positive learning experience for program participants.

2. After learning as much as I can about the program theory and actual intervention activities, I try to establish a logic model or conceptual model of the program graphically and in text so that all of the key participants can comment on it, revise it, think about the intervention elements that are linked to outcomes, "evaluate" how accurately I have summarized the program and context. This phase of the evaluation is participatory and often empowering for program directors, staff and participants. Sometimes this is experienced as frustrating to some – “Enough discussion, already!” may be expressed. A little bit of frustration with the speed of evaluation planning is healthy. If I hear it repeatedly, I quickly make an initial plan on paper, before I feel ready to do so, to help the program members feel that progress is being made, and to give me a little more time to assess evaluation readiness.

3. Evaluation Design. The formulation of the evaluation design is a key topic with me, and, I think, sets the tone and conditions for the most important evaluation work ahead. I know that most program directors want their program to be accepted by their colleagues as "evidence-based" and they are aware that achieving this level of reception is rare and challenging. I try to be forthright with them that achieving the status of a "science-based" program is a long and arduous road with many obstacles in the real world. Firstly, I emphasize the strongest possible experimental design that will fit the context of the program, and discuss the trade-offs that will accompany each decision to move toward more quasi-experimental or exploratory designs. Secondly, I help the program leaders think about the best design for their objectives, and try to give them the tools (words and attitudes) to help them create a culture of self-reflection and professionalism worthy of the scientific method. But I also emphasize the value of qualitative methods best used to capture the larger context of the program, and the peculiar circumstances of implementing the program in its setting. What do participants say about the program? How does the program appear to the interveners? Where did the intervention vary from the plan? These are some of the questions I raise as I incorporate the answers into the evaluation design.

4. I am careful to research standardized measures for the outcomes envisioned by the conceptual model of the program, but I am open to consider indicators suggested by the program participants. I will give my advice about why one approach is more explainable, or will be more acceptable to the scientific community, but I am flexible on this. I believe that many of the most powerful and innovative aspects of programs are not adequately captured by existing standardized instruments. I am willing to create new items and modify them according to the needs and circumstances of the setting and the characteristics of the participants.

5. I carefully develop data collection methods that fit with the participants, and make use of innovative collection methods that are consistent with the characteristics and values of the participants. I will use computer-based on-line methods or paper and pencil methods - to ensure that all participants have an opportunity to respond. I often make suggestions for incentives, because I believe that participants' time is valuable and should be compensated whenever I ask them to do some work. I am prepared to write applications for human subject protection at the sponsoring institutions (IRB Committee applications vary but essentially are consistent with the federal IRB guidelines located at http://www.hhs.gov/ohrp/ , and will write and revise informed consent documents as needed. I will prepare responses to outside commentaries by experts who have suggestions for improvement - and make those changes even at the last minute when called upon.

6. Reporting Results. I am careful to check the quality of the data before initiating analyses. I use a step-wise analytical strategy, first analyzing frequencies for each question, and then move to more and more complex analyses. I then review each evaluation question, and attempt to answer each evaluation question using appropriate statistical analyses, often combining questions (or items) into compound constructs (combined variables) to allow for more sophisticated analyses appropriate for the question. (I have found that the simplest appearing evaluation questions are often the most complex.) I write readable reports tailored to the audience, professionally documented that include background, organizational context, literature review, methods, results, conclusions and recommendations, and abstract or executive summary. I try to include text and graphic representations of the results whenever possible to make the report attractive and understandable to different kinds of readers.

Four Cornerstones of Assessing Evaluation Needs

Four Cornerstones of Assessing Evaluation Needs
David S. Robinson, Ed.D.
October 1, 2006

Beginning Evaluators are often unsure about how to begin to assess an organization and determine evaluation needs. This is especially true when the student of evaluation is inside the organization - an employee of the organization that is the subject of an evaluation initiative. So what are the four keys to beginning an assessment of an organization and its evaluation needs? Here are my suggestions.

1. Be positive and confident that there are relevant articles, web sites, reports and digital information available and accessible out there to begin to explore the major mission of the organization. Many students report that after an "extensive" literature search, they cannot find anything relevant to their evaluation question or issue. My experience is that this more often results from an initial skepticism or negative attitude about the topic which is fulfilled while searching. Keep an open mind and be sure to examine the biases (through self-reflection) that reduce the likelihood of finding relevant literature about the topic. Key one: be positive and confident that you will find relevant literature about the organization and its mission, programs and services, goals and objectives.

2. Pay attention to how the organization describes itself in written and published literature, or in interviews with key stakeholders. How is the organization, program or service described by administrators, board, staff and clients? Is it only positive, sticking to the bright and polished aspects of the organization, or does it include the scuff marks and challenges? Is it balanced or one-sided? Key two: Shine or scuffed - members of an organization will tell you how ready they are to evaluate their programs and services by how open they are to objective observations.

3. Pay attention to the bright and shiny aspects of the organization, program or services that members repeatedly talk and write about. These are the best places to begin, because they are the easiest to recognize and an initial evaluation will more likely meet with openness and generous sharing in areas that are commonly recognized as positive and successful. Key three: Pay attention to the low hanging fruit.

4. Review the budget and table of organization and observe the different groupings of programs and services within an organization - the different beachheads will vary as to size and scope, and these groupings should match the budget. Pay attention to these groupings because they will lead you to ask questions like: Does the allocation of resources (budgeted programs) match the organizational goals and staffing? Do some groups receive a disproportionate amount of resources? Do the groupings and staffing resources match the strategic goals and objectives? Key four: What are the different "beachheads" and "shores" of the organization? How do those beachheads compare with the resources, staffing, mission, goals and objectives?

An evaluation initiate can use the four keys to evaluation needs and find his or her way to a beginning dialogue with leaders and colleagues thereby starting an evaluation that fits the organizational profile of evaluation readiness.

Web Sites for Learning About Best Practices

I thought I'd pass on a useful list of web sites for finding out more about best practices. This list was compiled by Karl Hamner and forwarded to the Evaltalk Listserv. Hope you find it useful. David


On-Line Best Practice Resources

Mental Health & Substance Abuse: SAMHSA’s National Registry of Effective Programs (NREP) provides effective substance abuse and mental health programs for every community at http://modelprograms.samhsa.gov/.

The Department of Justice has a wide variety of youth violence prevention resources on its well-organized site at http://www.usdoj.gov/youthviolence.htm.

The National Institute on Alcohol Abuse and Alcoholism (NIAAA) is committed to helping colleges and universities reduce alcohol-related problems on their campuses, protect students from harm, and improve quality of life for the entire campus community. To achieve this, NIAAA has created the resource Reducing Alcohol Problems on Campus: A Guide to Planning and Evaluation.
This is available free at http://www.collegedrinkingprevention.gov/Reports/Handbook/default.aspx.
Other related and useful publications can be found at:
http://www.niaaa.nih.gov/publications /guides.htm.

The Department of Education (DOE) has published Identifying and Implementing Educational Practices Supported By Rigorous Evidence: A User Friendly Guide. This guide provides K-12 educational practitioners with user-friendly tools to distinguish practices supported by rigorous evidence from those that are not. It is available at http://www.ed.gov/rschstat/research/pubs/rigorousevid/index.html. More related resources from The National Center for Education Evaluation and Regional Assistance (NCEE) are at http://www.ed.gov/about/offices/list/ies/ncee/index.html.

The What Works Clearinghouse web site, http://www.whatworks.ed.gov, by the U.S. Department of Education's Institute of Education Sciences, gathers together current best practices in education. The introduction reads: “On an ongoing basis, the What Works Clearinghouse (WWC) gathers studies of the effectiveness of educational interventions (programs, products, practices,
and policies). We review the studies that have the strongest design, and report on the strengths and weaknesses of those studies against the WWC Standards so that you know what the best scientific evidence has to say.”


Karl Hamner
www.kmhconsulting.com

Do You Know What a Summative Evaluation is?

Robert Stakes

"When the cook tastes the soup, that’s formative; when the guests taste the soup, that’s summative."

Summative Evaluation: The assessment or judgment of the value of a program, product, service or educational intervention in relation to similar programs, products, services or educational interventions. The judgment of value is also based on the expected changes in the target population and based on a theory about how the program, product, service or educational intervention is supposed to work. (David S. Robinson, 2004)

Short List to Build Evaluation Culture in Organizations

There is nothing harder than starting out to evaluate a new program. Well, actually it is easier to evaluate a new program than to evaluate a program that has already been running for a time. Here is a short list of things to consider when beginning to evaluate a program - something to get you started:

1. Start with the leaders and gain their approval - as high up as you can go.
2. Then double back and gain approval from the line staff, the personnel who do all of the work, and who will probably provide most of the information for the evaluation.
3. Learn as much as you can about the work culture and the details of the activities that people do.
4. Use some form of structured logic modeling to organize what you learn about the organization, and detail the goals and objectives of the program, the resources used, the activities enacted to meet the goals and objectives, the direct products (or outputs) of the activities, and the immediate, intermediate and long-term outcomes.
5. Brainstorm what would be an observable indicator for each objective and outcome.
6. For each indicator establish a data source and collection method with a schedule and responsible personnel identified.
7. Share with leaders and line staff and revise, revise, and revise until everyone is sick of it or agrees it is the best plan.
8. Congratulations, you are well on your way if you have gone this far.