Sunday, October 18, 2009

Using the Advocacy Model to Present Evidence

As I have described before, the Advocacy model is what is used when a work discussion is viewed as a contest and one side is trying to “win”. I feel an obligation to insert the following statement: I think the advocacy model is what is wrong with business communication and I only bring it up here because it is as important for business people to be able to identify what is wrong as what is right. The advocacy model should almost never be used in business, but it frequently is, just because the inquiry model is not well understood by most business people. The Pathfinder Mission is to help change that.

In the advocacy model, the general rule for presenting evidence is: only present evidence that supports your position OR that undermines the opposing position. This makes perfect sense if you are viewing the discussion as a contest. So if there is evidence that would undermine your own position, you would NOT present it in an advocacy discussion, but you WOULD in a collaborative discussion because you would want to build a perspective with the least downside, so you have to include the downside in the “information pool”.

One kind of evidence that is normally presented in a discussion is statistical information. The first attribute of statistical information that you should be ready to discuss is its source. For your perspective, you should try to use information from a well-known source OR one that you can describe as a generally accepted source. Conversely, if your counterpart presents statistical information, attack the validity of the source.

You can ask questions about was the data taken from a suitable range of samples, or were the samples taken randomly. If they answer in the negative or that they don’t know, you can be dismissive of the data because it is not well-developed. Be advised that even among professional statisticians, there is disagreement upon what constitutes an adequately random sample. If they answer in the affirmative, then you can ask them how they know that because that information is rarely supplied with the data. If they prepared the data themselves, then ask if they are a degreed statistician. If they say yes, then ask them if they are a professional statistician.

You get the idea.

Another attribute of statistical data is the notion of the “Average” value. The idea of an “Average” value actually can apply to any one of three different concepts: the “arithmetic mean” (what most of us think of when we say average), the mode (the number that occurs mostly frequently in a group of numbers), and the median (the number from which half the numbers are larger and half the numbers are smaller).

Let’s say you gave the following set of numbers: 19, 25, 25, 30, 60, 80, and 95. The Mean is 47. The Mode is 25. The Median is 30. The minimum is “as few as 19”. The maximum is “as many as 95”. Which number best supports your position?

If the opposing position makes claims based on the "average", ask questions about the data set and use the answers to show how they are trying to be deceptive.

You go into an electronics store that is selling a toaster, normally priced at $30 but this week is $15. You see a television in the same store that sells in other stores for $1000. In this store it is in the same ball park at $985. Let’s say you want to advocate that store to someone. Certainly you could say “I went to the XYZ store and saw a toaster for half off. I also saw a TV at similar savings.” Certainly the $15 off the TV is similar to the $15 dollars off the toaster, right?

The key question to ask is what do they mean by similar savings: percentage-wise or dollar value? You can make differences seem smaller or larger than they really are just by expressing them as percentages or as values.

Next week, I will present more on how the advocacy model twists evidence and how to deal with it.

Insist on great business results! Go to Pathfinder Communication

No comments: