Last August at the ICSOM conference in Cincinnati, I gave a presentation regarding preparing for bargaining (“Bookends of Bargaining, Part 1”). Of the topics I covered, orchestra surveys sparked the most feedback. Many delegates have asked if they could share that part of the presentation with their colleagues, so I would like to expand upon that here as orchestras start preparing for next year’s negotiations.
Orchestra surveys are critical in preparing for bargaining. At a basic level, they help answer the question every bargaining committee considers: “What do my colleagues want?” In certain ways the answer is obvious—more salary, better healthcare, better working conditions. But knowing that your colleagues want improvements isn’t of tremendous value to the committee in the bargaining process. As everyone knows, you don’t get everything you want. The committee is forced to make choices—often, very hard choices—and to prioritize. A good survey will go beyond the obvious, and give the committee the knowledge it needs to make those choices when the time comes.
The first task in drafting a survey is coming up with topics, i.e., the issues you expect will be on the table in bargaining. Start by identifying broad categories, like health care, orchestra complement, retirement benefits, or work relief. Then get more specific with respect to issues that have come up in your orchestra. The source of those issues can be varied: disputes that have come up with management; issues raised in orchestra meetings or smaller group meetings with the committee; or issues emerging in the orchestra world (your ICSOM delegate is a good source for those). Often, the committee will solicit suggestions for items to be put on the survey. That can be helpful, but use discretion; not everyone’s pet issue should make it onto the survey.
In terms of what to ask and how to ask it, there are a number of guidelines I typically advise, and covering them all would make for far too long an article. But here are what I see as the three most important:
Don’t ask questions to which you already know the answer. This may seem obvious, but I often come across survey questions like, “Do you want better health care?” or “Should nine-service weeks be eliminated?” When those questions are asked in isolation, you know the answer will always be “yes.” Don’t bother asking.
Other guidelines flow from this. For instance, don’t ask questions about improvements that would benefit only a small number of musicians, or an individual (e.g., assistant principal pay). You know that those who benefit will want it, and those that won’t benefit probably won’t care (unless it is a point of principle that has universal appeal). Note that does not mean the committee shouldn’t seek those kinds of improvements in bargaining; rather, the committee will use its own independent judgment as to whether to seek such improvements and how hard to push. But there is no need for a survey question.
The same applies to pretty much any type of salary question. It does no good to ask your colleagues how much of a raise they want, or will accept. The answer is obvious: as much as possible. In bargaining, the committee will try to obtain the largest increase management will agree to, period. Then the committee and, at ratification, the bargaining unit will need to determine whether the salary component of the deal is acceptable in the context of the entire agreement and the circumstances of the situation. So whether your colleagues want a certain percentage raise, in the abstract, is simply irrelevant.
Ask your colleagues to make choices. This may be the most important survey guideline of all, and the hardest one to get right. As noted, the bargaining committee invariably must make difficult choices, and the most difficult are when the committee must decide whether to drop a particular proposal the committee knows is of great importance to the musicians, in favor of pursuing something else the committee determines is more important. When making such choices, it won’t be enough for the committee to know that the musicians really want those improvements, or even how badly they want them; rather, the committee needs to know which improvements the musicians want more than others.
Consider, for example, questions about typical subjects like health insurance, orchestra complement, and work relief. Commonly, surveys will ask “how important” each is, and provide answer choices such as “very important / somewhat important / not important” or offer a numeric scale of importance. That is only somewhat helpful, as it doesn’t answer the question, “as compared to what?” Such questions give only an idea of how strongly someone feels about an issue, but not that person’s priorities. It remains unclear whether that person would be willing to forego work relief in exchange for adding a musician to the complement—or whether that person would accept any less of a salary increase, or any concession on health care, as the price of obtaining that additional musician. That is the kind of information the committee needs.
The easiest way to fix this is with a ranking question—list a few items and ask the respondent to rank them in order of importance. That will indeed identify priorities. But ranking questions suffer from some common mistakes. For example, I often see nine or ten items on the ranking list, ranging from salary to doubling to getting more Mondays off. That’s not at all helpful, for several reasons.
First, there are simply too many items. It is difficult for anyone to keep track of more than three or four items simultaneously, and that’s what you want—comparisons that are as direct as possible. Second, it does no good to combine dissimilar items, like economic vs. non-economic or “big economic” issues vs. minor ones that affect only a few people. It is both too difficult and too easy to prioritize between wholly different items, or between items of entirely disparate magnitude. Plus, you probably already know the answer: the orchestra as a whole will almost certainly rank salary as more important than Mondays off, or retirement benefits as more important than raising the doubling rate. So, instead, group like issues together—or, to look at it another way, those that you think could actually be a toss-up. It’s fine to have multiple ranking questions, each with three or four items.
Note that I also often see ranking questions where the order of the items implicitly shows the bias of the drafters—for example, putting salary at the top and Mondays at the bottom. That also pretty much tells people how to answer. Instead, mix it up. SurveyMonkey is great for that, as it has a feature that randomizes the order for each respondent.
Another way to force the identification of priorities is to flat-out ask which of two issues is more important—or, even more uncomfortably, ask whether the respondent would be willing to forego one improvement if that was the only way to obtain the other. For example: “Would you be willing to accept less of an increase in salary if that were the only way to obtain an increase in complement?”
It’s at this point that I typically hear objections. “Everything is important—we should ask for it all!” “Management will see this—they’ll think we’re willing to give stuff up!” “This looks like we’re already bargaining against ourselves!” I get it. But there is an easy cure: explain the methodology, usually in a preamble to the survey. Explain that the survey should not be read as an indication of what the committee may or may not propose at the bargaining table; that questions asking for a choice between two items does not mean the committee does not intend to seek improvements for both, but instead is merely a tool to identify priorities; and that there will be an opportunity for further comment in the survey. As for management seeing the survey questions, you have to assume that is a possibility. But asking a question about an issue does not necessarily signal a willingness to make a concession—and if management takes it that way, you can always say in bargaining that you surveyed it, and your colleagues said, “hell no!”
Ask your colleagues what they want and how they feel, not what the committee should do. I sometimes see questions that ask, “should the committee try to [obtain X]?” That’s the wrong way to ask the question. Deciding what to ask for in bargaining isn’t a decision the orchestra makes as a whole—they elected a committee to do that. The orchestra workplace is a representative democracy, not a direct one; the committee doesn’t make decisions by referendum. That’s actually to the musicians’ benefit, because then management is forced to deal with the folks at the table (as opposed to trying to negotiate with the group as a whole). Not all musicians understand that distinction, which requires some education. Avoiding survey questions that ask what the committee should do can be part of that education.
Similarly, don’t put specific proposals on the survey. For example: “Should nine-service weeks be eliminated?” Apart from asking this in isolation (see above), another problem is that it is presented in the form of a solution rather than the problem that needs to be solved. A better way would be to ask questions about whether the musicians feel overworked, and what the source of that feeling is. I’ve had several situations where the bargaining committee was initially certain that nine-service weeks were the problem, but it later came out that the real issue was the increasing number of programs per week. The committee needs to know what the problems are and how people feel about those problems; but it is up to the committee to come up with possible solutions.
Finally, I’m often asked whether survey results should be shared with the orchestra. The answer always is “no.” For one thing, management will find out. It’s not the end of the world for them to see the questions, but obviously the answers are another story—you don’t want management to know the musicians’ priorities and bargain accordingly! Even if you take steps to try to keep the results within the bargaining unit—for example, stating the results only orally in an orchestra meeting—rumors may get around in inaccurate or exaggerated ways. Also consider that if the musicians see what has been identified as top priorities, they may expect to see those improvements when the committee brings back a tentative agreement for ratification. If the agreement doesn’t line up with those expectations, there may be a problem—even if it’s an acceptable agreement overall.
Note: The author is ICSOM Counsel.