Communication Between Researchers and Practitioners: Findings from a Qualitative Evaluation of a Large-Scale College Intervention

Abstract

Community-based interventions often include local coalitions. The focus of these coalitions often diverts to motivation and commitment, not implementing project goals. In the Safer California Universities study, we implemented a very directive approach in our instructions to local liaisons on the participating campuses, setting specific benchmarks for implementing the interventions. This paper reports on the findings of a qualitative study based on interviews with campus liaisons of how the directive approach was perceived on campus. Findings indicate that the strategy was successful in terms of “getting things done” but could be improved in terms of liaison involvement in decision-making processes.

INTRODUCTION

As prevention science matures, we are becoming increasingly cognizant of the enormous gap between science and practice, and of researchers doing their part to close that gap by attending better to issues of prevention implementation per se. In conventional views, a distinction is made between efficacy studies, in which an intervention implemented under close control of the developer or researcher, and effectiveness studies, in which the efficacy of a specific intervention is less an issue than how well it can be implemented “in the field” and outside the direct or close control of an investigator (Holder, et.al., 1999).

This distinction is blurred, however, when conducting efficacy research in organizational or community settings, where implementation of the intervention depends “on persons or groups that cannot be completely controlled by the research team” (ibid). In those cases, we often must rely on best practices (broadly defined) or improvise implementation strategies on the fly. Needless to say, this is not the most efficient method to improve implementation practice and advance the science. In this paper, we wish to directly report on an implementation strategy that some may find unconventional, but which also may expand the options that researchers and practitioners may now consider when working with communities.

This study is an outgrowth of a larger community-based prevention evaluation that sought to reduce alcohol- consumption-related problems across a number of large public universities in California. The success of the implementation was dependent on our setting up an effective relationship with prevention specialists on each campus (our liaisons) and through them, to the coalitions or task forces that would carry out the intervention. Previous experiences putting together coalitions (Berkowitz 2000, Cohen, Baer & Satterwhite, 2002; Hallfors, Cho, Livert & Kadushin, 2002, Wandersman & Florin, 2003) show that they often are not the best tools to guarantee local action but instead can often become inner-directed without pushing for action. Our implementation strategy was aimed at creating and maintaining a focus on the intervention objectives by keeping meetings to a minimum while providing extensive details for the meeting agendas and the associated tasks for moving from planning to action. In short, we took a far more ”directive” approach with our collaborators than is found in the literature or conventional wisdom. Mindful of the risks such a strategy may create, at least according to the literature (Wandersman, 2003), we conducted an assessment of our approach and report on it here. Our hypothesis was that we could be more directive or prescriptive, while still fully “respectful” of our community collaborators (Addiction, 2005). In fact, we argue that clarifying the complementary roles of the researchers, liaisons, and broader community enhances all those relationships.

We begin by distinguishing between coalitions and task forces. A coalition consists of people who are relevant to create the foundation, e.g., make policy changes, make decisions about what kind of interventions to carry out, have authority to make interventions happen, etc., for implementing interventions directed at changing a specific behavior, in this case high-risk drinking among students. A task force, on the other hand, consists of people who can be instrumental in implementing the actual interventions and therefore often consists of community actors, e.g., police officers, social workers, prevention workers, etc. In other words, the coalition is supposed to provide the “community readiness” for the actual interventions, whereas the task force’s job is more narrowly defined as being instrumental in carrying out the interventions (see Treno & Holder, 1997). Hence, the directiveness in this project related to the process of implementing the actual interventions locally.

This distinction between coalitions as foundational vs. task forces as instrumental may explain why so much of conventional wisdom with respect to coalitions is focused on motivation and commitment. In using these terms or others (e.g., ownership; buy-in) emphasis seems to be given to “bringing the coalition along” via processes meant to create consensus and commitment. These processes include attending to coalition membership (being inclusive), making sure that everyone has a chance to speak (and honoring divergent perspectives, as well as finding mechanisms to make decisions that will maximize consent if not consensus.

In our case, however, we chose to pay minimal attention to such issues by limiting the planning phases to 3 meetings, and reaching agreement on the targets in the very first meeting. Although this was a gamble, our aim was to promise a short and fixed planning phase so that senior officials and administrators would agree to attend (and give their stamp of approval). We also made the assumption that the interested parties were already in agreement that college student drinking was a problem needing attention, and that they (and the larger community) would be eager to move directly into the action phase. As it turned out, our basic assumptions seemed correct in this case, and we therefore paid much greater attention to the logistics of implementation rather than spending time on motivating change per se (though it must be said that moving quickly to direct action was, for many, itself an energizing change of pace from common practice).

The directive approach included having the liaisons form a coalition for planning the local interventions. However, unlike conventional community prevention approaches, we provided specific directions for the planning process as well as the implementation phase. We specified the planning phase to include three planning meetings as well as agenda items for each. The planning phase was followed by the implementation phase including three implementation meetings, for which we also provided agendas and specific directions. We fixed the number of planning meetings to three so that senior department heads or administrators would be more likely to attend (as opposed to an open-ended invitation). We made an effort to keep the planning and implementation groups relatively small (e.g., 7–9 people in each group) to avoid the danger of losing momentum and focus that often arise in typically larger task forces or coalitions. In sum, we wanted less discussion of what to do and much more on how to get it done. We asked that there would be a transition from the planning phase to the implementation phase including changing the composition of the working groups, the idea being to transition from a coalition in the planning phase to a task force in the implementation phase.

We were, of course, sensitive to the possibility that being so directive could alienate our collaborators at each of the campuses. Throughout the process, however, we continually checked with them to ensure that they were comfortable with our approach. There was consensus that our providing such detailed instructions was appreciated (with minor exceptions), and a pleasant departure from more traditional technical assistance.

The purpose of this qualitative evaluation was 1. to measure the liaisons’ perception of the strategy presented by the directive approach from the research team, and 2. to obtain information about the success of the approach, i.e. if the strategy had provided a good foundation for carrying through the community interventions and secured a working collaboration among local partners.

Intervention Strategy

The Safer California Universities study is a 5 year multi-site intervention study. The project involves 14 California university campuses which were matched in pairs and randomly assigned to condition as either an experimental site or a control site. Each campus designated one or two employees (usually from either Health Services or Student Affairs) to be the “liaisons” for the project on campus. Those staff members were already designated prevention specialists at their home institutions. The liaisons then participated in the process of planning the overall project and became responsible for carrying out the interventions in their community.

In considering the intervention strategy, it may be helpful to think of its comprising three different levels. In the broadest or most general definition, our objective was to adopt an ‘environmental’ or public health approach to reduce college student drinking problems. In particular, we wanted the experimental campuses to adopt a set of intervention strategies recommended by the National Institute on Alcohol Abuse and Alcoholism’s Task Force on College Student Drinking (NIAAA, 2002) as “Tier 2” strategies - interventions that had been shown to be effective in general population settings, but had not been rigorously evaluated in university or college communities. Nearly all the strategies in the ”tier” represented public health, alcohol control measures such as restricting alcohol access, responsible beverage service, raising alcohol prices and the like.

The second level of intervention strategy for our project concerned the choice of specific intervention components. This was done after campuses selected the setting to target. As it happened, all experimental campuses chose to focus on off-campus parties, not only because our baseline student survey data showed them to be the primary source of drinking problems, but because this also conformed to local staff and task force members’ perceptions of the problem. On several campuses, off-campus parties necessarily included student fraternity and sorority houses (social organizations often sharing a group home) that would be subject to the same interventions.

Once the targeted setting was chosen, planning groups were then given the specific components recommended by the NIAAA task force, supplemented by those of the Institute of Medicine’s report “Reducing Underage Drinking: A Collective Responsibility” (National Research Council and Institute of Medicine, 2004). The primary focus was on a set of alcohol control measures coupled with heavy publicity to give visibility to those enforcement activities. The enforcement included 1) roadside DUI checks; 2) underage decoy operations at retail outlets; and 3) designated “party patrols” that would enforce local and state laws regarding provision of alcohol to minors or disturbing the peace. These operations were to be amplified by the use of local media reports and events to announce the prevention program.

The third level of strategy focused on implementation per se. A major challenge for comprehensive, community prevention interventions is to maintain focus and to coordinate resources within a specific time frame. If the intervention becomes diffuse, the desired synergy among components will not be generated. With college student drinking problems, there is research evidence (Del Boca, Darkes & Greenbaum, 2004) and anecdotal accounts that suggest problems are greater and more prevalent in the first days and weeks of the school year. We used this information to mobilize the intervention effort so that all enforcement and visibility activities would be ready to go on the first week of school and continue through the end of October or early November.

METHOD

After the first year’s interventions on and around campuses, the liaisons on 5 of the participating intervention campuses were interviewed. Two campuses were left out, one because of scheduling problems, and the other one because of no activities in the previous semester for reasons unrelated to the intervention. The interviews were recorded using a digital recorder. The liaisons were given the opportunity to bring to the interview other key personnel that had been involved in planning and implementing the project on their campus respectively.

The main purpose of the interviews was to get the liaisons’ and other key personnel’s perception of the process of the project through the previous year. This knowledge was vital for the research team to make potential adjustments to the procedures for the following year of implementation. The interviews focused on the different phases of the project from the previous year, in particular overall planning meetings at the Prevention Research Center with liaisons from all intervention campuses, the planning phase on each campus, the implementation phase on each campus, materials generated by the research team to guide the liaisons in carrying through the interventions and visibility efforts on campus, and the overall organization of the project, including communication between PRC and the liaisons as well as communication among the liaisons themselves. Each interview was scheduled to last an hour, which turned out to be fairly accurate; most of them lasted between 45 minutes and 1 hour and 15 minutes. The interviews were performed by the lead author of this paper and took place in the liaison’s office on each of the campuses.

The interviews were analyzed using the Qualitative Media Analyzer (QMA) software. This software allows direct coding of audio files, meaning that rather than transcribing each interview verbatim, coding can be performed on the actual sound files. For smaller projects with only a few interviews like this one, this software presents an excellent alternative to more elaborate programs that work with transcribed text. The final product of the coding process is a number of sound segments each attached to a code, for all interviews obtained. After coding, a thematic analysis was performed focusing on salient themes relevant to the aims of this paper.

FINDINGS

Four major groups of codes relevant to the purpose of this paper were identified in the interviews:

1. Organization, comprising information about the suggested planning and implementation of the project locally and centrally;

2. Communication, comprising information about all aspects of communication between PRC and liaisons, among liaisons, and different channels of communication provided during the first year of the project;

3. Networking, comprising information about collaboration internally on campuses and externally that the liaisons initiated, and

4. Recommendations, comprising suggestions that the liaisons had for improvements for the future of the project as well as suggestions that would help make the project applicable in other similar settings.

Organization

As part of the first year of implementation of interventions, two meetings were arranged for liaisons at PRC in Berkeley. The first meeting took place in January of 2006. At this meeting, the research staff laid out the data of a student survey from the participating campuses performed in the previous fall that showed that the setting for most problems experienced by students on all participating campuses was off-campus parties. The purpose of this meeting was to present the data and reach consensus about the setting for interventions for each of the campuses. The second meeting was held in May with the purpose to prepare the liaisons for the transition from planning the interventions to the actual implementation.

The interviews focused on a number of questions regarding these meetings. First of all, we were interested in knowing if the meetings had been helpful for the campuses in planning and implementing the interventions as well as helping the liaisons to get a better understanding of the project in general. In the interviews we also focused on issues such as the organization of the meeting, the relevance of the topics, and recommendations for improving the meetings.

Overall, there seems to be consensus among the liaisons that the first meeting that took place in January was very helpful and clear in its message:

• We saw our data, and we saw all the other campuses’ data, and I thought that gave us the true picture, and, you know, was evidence based about what direction we were going in.

And a little later in the same interview:

• I thought that meeting was very organized and very clear versus the May meeting.

• (Campus liaison, CSU campus)

The presentation of data seemed to play an overwhelming role for the success of the January meeting. Another liaison described it this way:

• I know that the January meeting was very helpful because I think up to the January meeting I was kind of wondering at what point we were going to get data in such a way that would really help us make a determination about either how to intervene or where to intervene or whatever. And I think it was a little bit fuzzy up to that point even if we were in communication with [research staff] what the implementation was gonna look like.

• (Campus liaison, UC campus)

The same picture is repeated in this liaison’s estimation of the importance of the data presentation in the first meeting:

• I think that – at least for me – that the structure, being able to … the visuals and the very thorough nature of that data presentation was very helpful and to be able to take that information back to the campus, that was really powerful. Because we would often focus on “Oh, the Greeks” or “Oh, the Resident Halls”, but the off-campus parties, which for us was so significant and as it was for almost everyone, really opened peoples’ eyes and, I think, allowed us to think differently about where we were putting our energy, and then have strategies associated with that for each of the different areas of concern was very helpful.

• (Campus liaison, CSU campus)

Whereas the group of liaisons had to be regarded as a somewhat homogenous group by the research team in terms of planning the PRC meetings, the liaisons on the involved campuses were far from joining the project with the same background. Among the liaisons were huge differences in level of experience carrying out projects, preferred approaches to do alcohol-related work among students, and the level of experienced problems on campus. This created some problems in terms of bringing up issues with relevance to everyone in the group. It came through in some of the interviews in which it was pointed out that some of the information presented at the meetings was not relevant for these particular liaisons but it may have been for others with a different background:

• I felt like the May meeting was very … every school was coming from a different place, everybody’s readiness was different, so the questions that were being asked were all over the place. And then the lecture or demonstration to us about media for some was very beneficial but for others of us who have been doing this kind of work for a long time, you know, we know how to do all that, so I didn’t understand why we were going through all that.

• (Campus liaison, CSU campus)

Another issue that was raised in the interviews was the proximity of the May meeting to summer. Two main reasons for these difficulties mentioned in the interviews are the difficulty for some liaisons to get work done during the summer months because most of them are gone or important partners on campus may be gone, and deadlines for having various materials needed for the interventions approved in time for the “campaign”:

• I thought it was somewhat of an unrealistic timeline to hold a meeting in May when, for instance I am off the whole month of July, a lot of people are on a 11 or a 10 (?) schedule, which renders the summer months pretty unproductive. If you want, let’s say, copy proof for a party guide by all the key players, associated students, student participation, you want police, campus police, you know to joggle vacation, people you want to have involved. Hold a meeting in May sent us off, right as we approach finals where everybody is crazy anyway, and want us to have a plan in place before school starts in August. We were already sort of too late at that point in some respects. We pulled it together and got it done but I think it would have been more useful or helpful to have that happen earlier.

• (Campus liaison, CSU campus)

Even though the overall evaluation of the PRC meetings was positive, there were suggestions to how the meetings could have been improved:

• It was good to have the data behind it because we were going to represent that and I understood the depth of … it was felt that it was required. But I remember thinking “Wow, couldn’t we have read this beforehand and come with our questions instead of feeling like ….” I remember feeling like I had to fight and I felt that it kind of frustrated some people like I was interrupting the process but the more that happened the more I got less interested.

And a little later in the same interview:

• You know there was hardly time to speak to each other about what are our worst fears and best hopes.

• (Campus liaison, UC campus)

The overall impression from the interviews concerning the PRC meetings is that they were viewed positively by the liaisons. They felt that the presentations by the research team included high quality information, and despite some feelings of frustrations about some issues, e.g., timing and the lack of time to exchange information among liaisons, the organization of the meetings and the relevance were viewed positively.

Directiveness

Another issue concerning the organization of the project was the question of directiveness by the research team. The directiveness pertained to the planning and implementation process locally. The research team had laid out the meeting schedule for both phases, including number of meetings, agendas for meetings, how much time should be spent on each issue, report forms from the meetings, and delegation of responsibilities to members of the implementation task force. The reason for this directiveness was two-fold; 1) the research team wanted the meetings to be very focused on the issues of the project, and 2) by limiting the number of meetings it was assumed that it would be easier to involve relevant influential people.

The interviews give a mixed picture of the success of this approach. For some of the participating campuses, the directiveness worked very well for others it was seen as an obstacle. The interviews show that liaisons with many years of experience in prevention tended to see the directiveness as an obstacle as opposed to liaisons who were relatively new to the field. Often people with more experience had developed a specific way of “doing things” whereas “new” people were more receptive to the ideas laid out by the research team. An example from a CSU campus highlights the latter:

• I think it forced us to have those meetings when, you know … according to schedule. Because I remember thinking: “Oh, there is another meeting like in … you know, next month” or, you know … it was very laid out, and I liked the structure. I mean, another meeting but it is almost like it forced us to look at it again and do that, and by having them scheduled the way you did it, I don’t know exactly how, but I am sure you gave them “OK, wait three weeks and then tell them about another …” and so on. It kind of forced us to have a reaction and to do something.

• (Health Director, CSU campus)

For the more experienced liaisons, the directiveness seemed to put strains on their ability to “work as usual”:

• I: Back to the planning phase a little. As I mentioned, one of the … the way we had laid that out was very directive with agendas and how many minutes you could spend on each item. How did that work out for you? Did you feel that was a good idea or was that too directive for you?

• R1: Well, being a product of the sixties that was too directive.

• R2: Ha ha ha, we didn’t want to … (?)

• R3: We didn’t follow all of the (?)

• R2: Oh oh, admitting that on tape. Yes, we tailored that part as well.

• (CSU Campus)

Another part of the question of directiveness is the issue of materials provided to the liaisons by the research team. Throughout the planning phase, materials relevant to the planning and implementation of the interventions were developed by the research team and forwarded to the liaisons. These materials included clear instructions about the content of the suggested interventions as well as how to implement them, agendas and report forms for local meetings, instructions for transitioning from the planning phase to the implementation phase of the project, and a check list including timelines for the implementation of enforcement activities and related visibility activities. Once again, the interviews show a mixed picture of how these materials were received by the liaisons. Along the same divide as above, the materials were seen as more helpful by relatively inexperienced as opposed to more experienced liaisons:

• R: … but in terms of “your goal by the tenth week of school is to have had on your campus these six things with this much, you know, with this much visibility” or “four of these six things or three of these six things”. And I feel like, not that it wasn’t there, but I feel that I couldn’t have told you what that was, you know, in the middle of welcome week where I try to rush around thinking “well, who is my CL (communication leader), who is my ICL (internal coordination leader) for that”. And so I was a little bit … the tools perhaps were a little bit too, the-forests-for-the-trees-kind-of-thing, there were so many tools to use, I felt like I lost sight of what was the actual outcome that was supposed to be out of all of this.

• (Campus liaison, UC campus)

The research team had developed a GANTT chart describing in detail the tasks that had to be fulfilled and within what timeframe. The GANTT chart was intended to work as a check list for the liaisons to help them plan the different phases of the project. However, it was received with varying enthusiasm at different campuses; once again depending on the experience of the liaisons:

• R: The GANTT chart, I would say, was not helpfull.

• I: The GANTT chart was not helpful. And why was that?

• R: It was too confusing, none of the dates jived with academic year

And a little later in the same section of the interview:

• R: I was gonna say, to tell me to get a microphone is insulting, to tell me to get a podium is insulting. I mean that’s what we do all the time, and I don’t need to be told that.

• (Campus liaison, CSU campus)

On the other hand, the GANTT chart was viewed as helpful by others:

• I: For the timelines we developed the GANTT chart. How did that work for you?

• R: It worked well.

• R2: Which one was that?

• I: That was the one with all the tasks and all the deadlines.

• R2: And who was responsible?

• I: Yes, who was responsible for what. It was 6 or 7 pages long

• R: It was a chart that had like the different tasks and it had different timelines …

• R2: There was one that was very specific.

• R: Yeah, that’s it.

• I: It had like 215 different tasks into details like “get a microphone”

• R2: Yeah, yeah. I think it kept … to me it was like a “OK, so we didn’t forget anything if we could go and check these things off.

• (CSU campus)

As already mentioned, the question of fidelity vs. tailoring was also touched upon by the specificity and directiveness of the materials that were provided by the research team. Given the directive approach it seemed to be difficult for some of the liaisons to find out how much they could tailor the instructions to fit their own reality on their own campus:

• R: The (?) information we have been getting over the last year has been a template and without a lot of either clarity provided by PRC or permission provided by PRC and confusion on my part about “well, you say you can tailor but how much can I tailor?”

• I: So that was unclear?

• R: Yeah, and again it wasn’t so much at the first meeting that it came up but it was later “what if I don’t want to have three meetings? What if I want to have one meeting or two meetings or five meetings?” I am sure that if I had asked that it was OK, they would have said yes and yet I still get the e-mails about you have not given me your notes from meeting 3.

• (Campus liaison, UC campus)

Overall, the interviews show a varied picture of the liaisons’ perception of the organization of the project. There seems to be a clear divide between liaisons who have many years of experience and relatively inexperienced liaisons. Whereas the inexperienced liaisons found most of the organization helpful, a lot of the more experienced people felt restrained by the directiveness and the general organization. Hence, the interviews reflect the difficulty of finding a common denominator for building collaboration with many different entities with different experience and skill levels.

Communication

The issue of communication has several aspects. It pertains first of all to the channels of communication provided by PRC for efficient communication between PRC and the campuses as well as among the liaisons at the participating campuses and also to the use of these channels. In addition, issues pertaining to the communication between the liaisons and local entities, i.e. the police departments, university offices, and local neighborhood organizations, etc. are also included in this section.

In order to assure the best possible communication among participants, a number of channels of communication were established. A listserv was established for communication among intervention campuses alone and another listserv was established to be used for and by all participating campuses. Regular conference calls between PRC and the liaisons took place, and in periods without these calls, the conference call service was available for the liaisons to use in case they wanted to discuss issues among themselves. Furthermore, a web-site was developed that contained all the materials that had been developed for the project as well as bulletin boards, news, and available updates from the participating campuses. Despite these efforts, very little communication took place utilizing these channels among the campuses:

• I: Would you prefer to have more unstructured conference calls or is the structure very helpful for you? I mean, the structure apparently has been very helpful for setting up the whole project for you – I mean there are diverse opinions about that of course.

• R: I don’t know at this point what would be …

• R1: Well, it might be that the conference calls usually feel like PRC has some kind of information that you need to get – either need to give or receive from us – so … and honestly that is just fine. So I prefer the structure if that is the nature of the conference call. Again, it is just that I know that it is hard to get … seriously, it feels like if people are not with you in person, they are not going to make the time for it if they don’t have to. So, I don’t feel the need for more conference calls, structured or unstructured. I am also personally a connector, so in-person is so much better for me because then I can show people what we are doing, we can engage, whereas the conference calls feel so much less connecting.

• (CSU campus)

However, often communication took place in more informal settings:

• I: How about communication with other participating campuses, have you had a lot of that?

• R: Uhm…

• I: Apart from the conference calls that we have initiated.

• R: I know I talked to [other campus] a couple of times just because, you know, we happen to see each other a lot and exchange information and exchange ideas. I think what I find happening is that whenever we bump into each other at any other type of event or function we end up talking about “How is this going with your grant activities”, you know and things like that. So I have communicated a lot more with people just to find out what their challenges are, and if they can help me with anything with my challenges.

• (Campus liaison, CSU campus)

And in another interview, an accidental informal meeting leads to exchange of information and ideas:

• R: I had discussions because I met separately with some of the people [liaison] from [other campus] and the people from [other campus] just randomly. We actually realized that we had shared some problems that we didn’t know each other had, and we had solutions for each other but that was informal, and that would be something that would be hard to put over the listserv.

• (Campus liaison, UC campus)

In general, the interviews show that there was a desire for more exchange of information and ideas among the participating campuses. However, despite the available channels of communication in most instances it did not happen. Many of the liaisons expressed that this exchange should have happened at the meetings at PRC, so there are indications that it is easier to exchange information and ideas in person rather than using electronic media to do so.

For the most part communication between liaisons and local entities went well. However, there were incidents of difficult communication particularly based on some confusion about what information was needed, who would provide the information, and when it was necessary to produce the information. The following quote from one campus illustrates some of these problems, and how important it is to build networks with local entities that are instrumental in carrying out local interventions and related activities:

• R: It is a tremendous amount of work for people who I think in a lot of our cases are not media trained. And so, like the DUI checkpoints for example, because I couldn’t find out from our law enforcement, you know, the couple of days before the city of [location] police person working on the checkpoint tells me that there is not going to be any campus police involved. So I have no campus police involved, because it took so long to find out what the date is we still have students scrambling with the coupons [certificates for non-DUI drivers], they didn’t know if they would have enough coupons to hand out. So I had been talking to our campus media, and he is like “What is the campus story?”, and I am like “Yeah, pretty much at this point, I can’t tell you for sure that there is one.”

• (Campus liaison, UC campus)

Even if communication in general was considered to be working, obviously there were problems that were damaging to the basic idea of the project; enforcement and visibility of enforcement. This example shows that there was a lack of awareness of how the different pieces of the program were supposed to work together, and therefore led to problems implementing the project on this particular campus.

The general picture generated by the interviews is that despite a few problems or obstacles here and there, the communication efforts as laid out in the project have been successful. A lesson to be learned from the interviews about communication is that in setting up different channels of communication, research staff needs to be more aware of existing communication patterns among collaborators. Ideas that may seem good and workable to research staff may not be efficient for the people who are going to be the users of the communication channels that are provided.

Networking

The issue of networking is related to some of the aspects of communication that were described above. However, it also pertains to the local organization of the project on each campus. Hence, segments of the interviews describing the liaisons’ cooperation with local entities to carry out the interventions and the visibility of these interventions are of interest. The research team had suggested a specific local organization of the project that included decentralizing some of the responsibilities of the project to local organizations or people. The idea was that different entities with responsibilities would be more committed to the project, the implementation of the interventions, and the success of the efforts. In practice, no campus followed these recommendations. The interviews show that most of the liaisons were uncomfortable handing responsibilities for parts of the project to others because they felt that they needed to be in control of the progress:

• R: In the beginning of the planning and implementation phases I think sometimes just our office ended up doing a lot of the other roles as well – it just ended up being a little easier that way ‘cause … it is a little bit scary, I would say, or nervous to rely on someone else to do all that and not know if it will be done properly.

• (Campus liaison, CSU campus)

In general, it appears from the interviews that communication between the liaisons and local entities necessary to carry out the project went very well. This may in part be due to some already existing connections or, as some of the campuses expressed it, "serendipity" that the campus was just ready for the risk management approach to try to prevent alcohol- consumption-related problems among students. However, there are also some indications in the interviews that the organization of the project played a major role in this success as well as being part of a large scale research project including a number of other campuses:

• R: We have used … community complaints to target some areas, and we have also been … this is really exciting, and this project has really prompted this, is that we have in an independently owned apartment complex that houses at least 80%, uhm, are students. And we have been working with the manager. We started off this semester, we gave them our “Wanna Party” brochure, our “Social Host” brochure, and they put it into every move-in packet, and they do a package for every … every student has an independent contract with them.

• (Campus liaison, CSU campus)

And even shorter with the police in a different interview:

• I: What did the police chief think of the layout for the interventions that we had proposed?

• R: They were very supportive.

• (Campus liaison, CSU campus)

And at another campus, the same picture:

• R: … Resources available within the law enforcement that they said “Pick your date, tell us when you want us to go out and do this.”

• (Campus liaison, CSU campus)

However, at some of the campuses there were initial problems establishing a good working relationship with the police departments or other campus offices. These problems pertained to the fact that the police were asked to do enforcement interventions on specific dates. Some police departments were not able to comply with the exact dates or were not willing to share the information of when their enforcement activities would take place. Another problem that emerged early in the process was how to establish collaboration with the police. Some liaisons felt uncomfortable “telling the police how to do their job”. In most cases, these problems were solved relatively fast and led to very productive relationships that are expected to last after the end of the project.

A third problem that occurred on a few campuses was the fact that it came as a surprise to some of the liaisons to find themselves having trouble collaborating with campus offices. In particular, a few liaisons had problems including the campus public relations office in the visibility effort on their campus.

In terms of establishing collaborations with local entities, the interviews describe many success stories. Some of the liaisons were successful in establishing contact with local key persons who had been involved in public voluntary work for a long time or with local neighborhood organizations that became involved in the project:

• R: So within the […] community we have a great community organizer who is part of our advisory board and we have worked very very close with her. She has met with every city council member, she has met with every member of the Board of Supervisors and she has (?). She is in a landscape that looks at [the community’s] alcohol and drug problems, and alcohol is the number one drug, and now she talks with them.

• (Campus liaison, CSU campus)

Lastly, a number of liaisons talk about the great contributions from students who have been involved in the project:

• R: And I should point out that our students were a big support in this as well. When I said on-campus individuals I think I meant staff and faculty mostly but as far as the students go, I feel very strongly that their support and their work on this project has been a huge contribution to the success.

• (Campus liaison, CSU campus)

The interviews show a fairly consistent picture about networking. Despite a few initial problems experienced by some of the liaisons, it seems that they all were able to establish good working relationships with local entities. A clear example of the success of this project in establishing a broad local support for the interventions and the whole program was described in an interview from one of the CSU campuses:

• R: A huge advantage from this grant for us is that we have really good working relationships with the city and specifically with law enforcement, and, you know, the key people who were involved in making things happen. I mean, being able to present and having the city manager there and the mayor there and the chief of police …

• (Campus liaison, CSU campus)

In particular, collaboration with local police and campus police seems to have been successful. A few liaisons were able to establish collaboration with local neighborhood organizations or community organizers. These relations, when established, turned out to be of great value to the planning and implementation of the project locally. Finally, students were involved on all campuses in planning and implementing the project.

Recommendations

One of the aims of the face-to-face interviews with the campus liaisons was to give them an opportunity to share their perspective on the project and their experiences from the first year of actual interventions. After the directive approach in this first year, the research team wanted to evaluate the process and make adjustments for the second year according to the liaisons’ experiences and recommendations. Even if the perspective of asking for recommendations for adjustments is focused on the future, these recommendations also give an extra opportunity to review the past experiences in the light of improving the next year’s interventions and visibility efforts.

The first issue that was raised in the interviews was the issue of process- vs. outcome-orientation. A lot of the planning time in the first year was dedicated to secure that the interventions were carried out with the proper visibility attached. This naturally led to an emphasis on the process of planning and implementing the suggested interventions. However, some of the liaisons found this to be frustrating because the heavy emphasis on the process took away some of their focus of the actual interventions:

• R: You need to, you know, be sure that you have someone that can do the media stuff, and so much was focused on how do you get the people on board and convince them to do this job. I felt that I little bit lost sight of what the outcome was going to be – not that it wasn’t there on the GANTT chart or all these complicated things but in terms of “Your goal, by the tenth week of school is to have had on your campus these six things with this much, you know, visibility” or four of these six things or three of these six things. And I feel like, not that it wasn’t there, but I feel I couldn’t have told you what that was.

• (Campus liaison, UC campus)

This liaison also raised the issue of “simplistic” messages that did not take into account the specificities of each campus:

• R: I felt a little bit like … then again, this is more of a generic comment overall, so I can’t remember if that particular meeting triggered this feeling or if it was the whole experience. Here is this formula, and we know it’s gonna work, and you can tailor it but as long as you just follow these 10 steps and get buy-in, then you’ll get … buy-in. And I remember thinking “Man, we have tried doing this work on this campus for 20 years", and I am not saying that some of this is not new and helpful but I have my doubts that it is that easy. Not that I think that Rick or Bob or whoever presented this portrayed it as being easy but there was a little bit of a sense of “If you just grasp the concept of how this is gonna work, it’s gonna work” and meanwhile you are thinking “I have tried to get the chief of police to a meeting for 10 years, and you are telling me that as long as I give her this agenda that will get her buy-in, and when I get her buy-in it will be smooth. Well what if she has an emergency" – it felt in some ways … simplistic is not the correct word because I don’t think it is a simple model … I don’t know, it just felt kind of frustrating.

• (Campus liaison, UC campus)

Similarly, other liaisons felt that if the objectives of the different meetings that were suggested for the planning and implementation phases had been better laid out earlier in the process, a lot of time could have been saved to focus more on carrying out the interventions:

• R: The only reason it would have been helpful to have the three meetings, like the three planning meetings, laid out is, again because we already had our team identified and ready to go, we probably could have covered all of that in one 2-hour meeting, and then moved right … And in fact, that is probably almost what we did. So, knowing the objectives of all three meetings could have cut down on the time factor.

• (Campus liaison, CSU campus)

Another major issue that was touched upon in the interviews was specific campus culture. There was a genuine feeling that the program was too generic and not sensitive enough to specific issues on different campuses. This was expressed in uncertainties about how much the program could be tailored to fit each campus (as described above), and it was expressed in uncertainties about how to work with the local university organization:

• I: Is there anything in that phase that could have been done a little differently that would have improved that process? – Any ideas?

• R: One thing that I think could have helped out is if maybe there could have been some type of contact with the legal departments of the different campuses to see where the campus would stand on certain issues like a social host policy … because sometimes with the request to have us do something like that if the campus is like “We don’t think so”, you know, we are in like a tug-o’-war about how to get it … find a common place where everybody can be happy.

• (Campus liaison, CSU campus)

Another liaison takes it a little further in expressing the need for an organizational change to carry out projects like this one on a campus:

• R: I was willing to give it a try but … I don’t know, it just felt that it was presented in a way, maybe that was like “Well, as long as you just follow these steps …” because here is the reason because you want to get buy-in. I don’t know why I want to get buy-in, just getting everybody in the room and sign a piece of paper isn’t the same thing as getting the buy-in, and that was the issue because you are not just talking about individual people’s buy-in but an organizational shift, and what catalyzes that is not three well-organized meetings … I mean I’m telling you this now in retrospect, of course at the time, I don’t think I knew this, but I knew that this change that we actually have achieved, so obviously something worked but it was deeper than the tools that were being provided, and I felt we spent a lot of time on the tools and the method and not a lot … even acknowledging that the underlying level exists.

• (Campus liaison, UC campus)

In the same line of thinking, this same liaison suggested more preparatory work on campus cultures or specificities to make more sound decisions about how to implement the project on each campus:

• R: … maybe if we had had some tool in January or in the semester before that just to do our sort of assessment of where do we think our campus already is in meeting some of these goals, and, you know, maybe if you are [other campus] you are here or if you are [other campus] you are here, and that means what you need and what your expectation is will be a little bit different … maybe you are in a small town where what happens on college is like the newspaper’s biggest story.

• (Campus liaison, UC campus)

The last major issue that was raised under this heading is the issue of timing. The PRC meeting to prepare the campus liaisons for the implementation of the interventions and the related visibility activities was held in May. However, for some of the liaisons this scheduling presented major problems, as many of their collaborators were gone for the summer, leaving them only very limited time to pull a local implementation plan together for the fall activities:

• R: I came back in July and no one else is here, and then we have like two weeks to pull everything off. So, this again is another thought that it probably would have been better to push some of that outcome-oriented … intervention-oriented expertise earlier on in the process because then I could have had … you know, at the same time when I was not filling out my agendas for the three meetings, I could have made consulting with this PR person on how to build that capacity on our campus.

• (Campus liaison, CSU campus)

Most campuses experienced some kind of difficulty with the timeline particularly during the summer months. Even if it worked well on one campus, it turned out to be a hassle on most of the other involved campuses.

Overall, the campus liaisons had valuable suggestions for changes in the project protocol. The segments from the interviews may seem somewhat negative but they are representative of local experiences and are presented with the purpose of improving the overall project and make it more accessible for other campuses to replicate.

DISCUSSION

The analysis presented in this paper adds an important element to the methodology of carrying out community based interventions to prevent alcohol-related problems. For many years, the discussion formed around different approaches known as bottom-up or top-down strategies to guarantee the biggest impact of interventions (e.g., see Allamani et al. 2000, Larsson & Hanson 1999). However, neither of these theoretical approaches seemed promising or capturing the reality of implementing community-based interventions. Instead, a new literature has emerged on different approaches to structure working relations between researchers and local key personnel in community based work. In this literature (e.g., see Winer & Ray 2005), different kinds of working relationships are presented that are based on the tasks at hand and what roles different parties may play in fulfilling the tasks. The point of departure is the question of intensity, defined as a participant’s risk, time needed, and opportunity. From this perspective, three different forms of working relationships are defined; 1) Cooperation, 2) Coordination, and 3) Collaboration in ascending order. A cooperative relationship is characterized by a great deal of autonomy and a structure, in which each organization retains authority and keeps resources separate. A coordinating relationship is characterized by a higher degree of planning and division of roles, and it opens communication channels between participating organizations. Finally, a collaborative relationship is characterized by a formation of new structures consisting of the participating organizations and with full commitment to the common mission (Winer & Ray 2005).

From the get-go of the Safer California Universities project, the different participants shared a common goal, that the interventions on each campus would lead to a decrease in risky drinking situations among students. However, it was also clear that the focus of the study for the participating liaisons differed from the focus of the research team. Whereas the liaisons naturally emphasized the impact of the interventions on their own campus, the research team emphasized the overall impact on all campuses in order to be able to demonstrate that the ”model” could be shown to have an effect. In order to be able to show an effect of a particular model, it is necessary to have a certain degree of fidelity in the implementation of the different elements of the model. At the same time, as was mentioned in the introduction of this paper, the model needs to be flexible enough to adhere to specific needs of the participating communities. The strategy chosen for this project attempted to be flexible to local needs, yet still warrant a high degree of fidelity to the overall model of implementation. This qualitative study represents an evaluation of this effort from the perspective of the participating liaisons.

The Safer California Universities study was focused on implementing a risk management approach to alcohol problem prevention on California campuses including a strong emphasis on enforcement of existing laws and regulations and a media component to enhance the visibility of the increased enforcement. This strategy has not been implemented in this setting before, but based on our experience with the Community Trials Project (Holder, et.al., 2000) we could anticipate some organizational obstacles during the planning and implementation of the project. Our adopting a very directive approach to the process of planning and implementing the components of the study was a logical extension of a similar strategy that proved successful in the case of Community Trials (ibid).

If we compare the actual approach that we implemented in this study to the model described above, the best match seems to be a coordinating relationship between the research team and the liaisons on each campus. The research team and the liaisons on the participating campuses were involved in a long-term interaction around a specific program; the implementation of the risk management approach to preventing alcohol problems among students. There was a great deal of coordination around planning and the "actors" took on different roles. Furthermore, a number of communication channels were opened to assist in the coordinating effort. However, this project included a number of actors from different campuses that were geographically separated, making the whole coordination more complicated than in the model. In coordinating a project involving two partners, differences may more easily be expressed and overcome than in a project involving multiple partners. As we heard in the interviews, some of the liaisons felt that they were not able to vocalize questions or problems during joint meetings, leaving them with unanswered questions that made them frustrated later on in the process. This may partly be explained by the coordinating approach that was chosen for this project. Furthermore, the directive approach, chosen by the research team also seems to have contributed to these frustrations. It is important to keep these frustrations in perspective, however. Much of what we have reported here had to be specifically elicited from our collaborators in light of our concern about potential ”backlash.” None of the issues brought to light were sufficiently serious to threaten implementation.

Another issue that was raised in the interviews relevant to this discussion was the issue of “up-front’ tailoring of the project to each campus. Some of the liaisons expressed some frustrations that the guidelines or directives did not match the reality on their specific campus, and that it would have been helpful if the research team could have collected information about some of the specificities of each campus prior to setting up the planning and implementation process. Again, the strategy chosen for this project did not encompass this kind of working relationship. It was foreseen that the liaisons would be the ideal entity on campus for working with the organization that they were a part of, rather than having the research team as an outside entity manage the project locally. It does require having relatively sophisticated liaisons, however.

Overall, the interviews left us with the impression that the liaisons showed great appreciation for the project, the strategies, the planning, and the support from the research team. However, they also expressed some criticism of the project and in particular of the timing of different steps in the logistics and some “insensitivity” to specific campus cultures. These problems might have been prevented if closer communication between the research team and the liaisons had taken place earlier in the process, i.e. in the general planning of the project. If such communication had been launched, the involved parties’ expectations could have been touched upon and dealt with earlier and been incorporated into the logistics of the project. Such an approach would have been closer to a collaborative model as described by Winer & Ray (2005). However, in this project with 14 participating campuses as well as the research team at PRC, such a model did not seem feasible. Instead, a model that would give the campus liaisons more room for tailoring the campus specific process of implementing the interventions and the related visibility efforts seems more appropriate.

Glossary

Efficacy study

A study measuring if an intervention can work, i.e. under best possible conditions.

Effectiveness study

If an intervention works in a real world setting.

Coalition

A group of local people who are relevant to create the foundation, e.g., make policy changes, make decisions about what kind of interventions to carry out, have authority to make interventions happen, etc., for implementing interventions directed at changing a specific behavior, in this case high-risk drinking among students.

Task force

A group of people who can be instrumental in implementing the actual interventions and therefore often consists of community actors, e.g., police officers, social workers, prevention workers, etc.

Community readiness

the interest and ability of a community to begin implementing a program.

Biographies

Peter Nygaard received his Ph.D. in psychology from the University of Copenhagen, Denmark, in 2000. He has worked as a researcher in the alcohol field since 1988. His first employment was with the Institute of Psychology at the University of Aarhus, Denmark, and he continued his career at the Centre for Alcohol and Drug Research at the University of Aarhus from 1991 to 1998. In January 2000 he joined the Prevention Research Center (PRC) in Berkeley, California, as an associate research scientist. In 2001 he became a research scientist with the center. In 2007 he joined the Norwegian Institute for Alcohol and Drug Research in Oslo but still maintains ties to PRC. Dr. Nygaard’s main research interest is in how social networks influence individual drinking behavior. His dissertation was an intervention study within groups of social drinkers, in which he was able to demonstrate an impact on the consumption level of members of social networks with elevated alcohol consumption. Dr. Nygaard has also been involved in international collaborations in community action research. Recently, his research has started focusing on how to involve social networks more actively in the screening and brief intervention field.

Robert Saltz received his Ph.D. in sociology from the University of Massachusetts in 1980 and has been involved in research on alcohol and other drugs since then. Recruited by the Pacific Institute for research and Evaluation that year, he then became a founding researcher at the PIRE’s Prevention Research Center (PRC) in Berkeley, California, where he is now a senior scientist. Dr. Saltz’s main areas of interest have been in the contextual and situational influences on alcohol consumption. His evaluations of responsible beverage service programs represent one outgrowth from this topic. Work that is more recent has addressed community-level prevention interventions, having been a collaborator on PRC’s Community Trials Project, extending into his current research on prevention of college student drinking problems. This project is leading to a growing interest in the theory and practic