Anchoring

Anchoring is a cognitive bias based on our tendency to rely too much on our first impression in how we may act, respond and make decisions. Anchoring has the ability influence answers we provide, whether it be conscious or subconscious level.

This post isn’t aimed at how we might defeat our anchoring bias but more so on some observations that I’ve come across in my own life about how anchoring could be used to perhaps influence other people.

Asking the Question

Recently I was asked the following questions one after the other:

How many people do you think is the minimal number of people we should market to in order to build a model?
How many variables would we use when we model? How many variables should we aim for? How many variables should we test in our modelling approach?

Each of these questions are open-ended with no correct solution, since they are all highly contextual (though there would answers which are definitely wrong).

But it was during the discussion in one of these which particularly stuck in me.

If we market too ~1000 people, would using 100 variables in our subsequent predictive model be too many? Too few?

In particular, what concerned me most, was would be answer be different if they went on the other extreme and said, would it make sense to only use and test 10 variables in our model.

This got me thinking afterwards, what kind of statistical heuristics would I use in answering a question like this? Rather than be swayed by an incidental and unrelated comment.

The truth is, anchoring is unavoidable. There are various studies where even overly exaggerating wrong answers will still unfortunately cause an estimate to be smaller or larger than average.

When the Ben Franklin Effect goes wrong…for you!

The Ben Franklin effect is best described as if someone has done a favour for you, they are more likely to do another favour for you compared with if they had received a favor from you. In his autobiography, Ben Franklin describes it in the following way:

Having heard that he had in his library a certain very scarce and curious book, I wrote a note to him, expressing my desire of perusing that book, and requesting he would do me the favour of lending it to me for a few days. He sent it immediately, and I return’d it in about a week with another note, expressing strongly my sense of the favour. When we next met in the House, he spoke to me (which he had never done before), and with great civility; and he ever after manifested a readiness to serve me on all occasions, so that we became great friends, and our friendship continued to his death.

So then, in a practical sense, if you’re starting a new job and you there doing favour for someone. For example, let’s say you’re helping them print something and you complete it in lets say two minutes, they would again expect you to fulfil the same piece of work in two minutes in the future. And then before you know it, you would be continually be doing printing favours for your new colleague. This is the Ben Franklin effect working against you.

Resetting Expectations

Perhaps the best heuristic to remove these act of favours which the other person can complete it, is through providing them opportunity cost. If they realise that by asking you to complete a menial task actually has a material cost, then they would be less willng to ask you to complete that same favour.

In the example above, the situation could work as follows:

  1. Complete the menial task as quickly as possibly (say two minutes)
  2. In the next time it is requested, wait a little bit but still complete it in a timely fashion; reminding them that this is a menial task only worth less than five minutes of time.
  3. Progressively increase the time taken to complete this simple task, reaffirming that this is a menial task which doesn’t take up much of the time.

Hopefully this approach would not come across as passive aggressive, but rather re-affirming that this is a favour, and that there is a cost involved; your time. And that your time is just as important as theirs. The truth is this is a delicate balancing act, but at least by understanding the cognitive forces driving them, we can better equip ourselves with tools and greater understanding to why these relationships and interactions are the way they are.