What is ANCHORING and why is it messing with my Jeopardy! game??!!
Several years ago my professional musician son, Cameron Kinghorn (http://nookyjones.com/; http://blackmarketbrass.com/index.html), was hired by a local church to “anchor” the tenor section of the church’s choir.
Because he has a powerful and strong tenor voice, when he sang the tenor part in the choir’s songs, the other tenors were more easily able to find their notes and sing out more forcefully and loudly than would have been the case had they not had Cameron to “anchor” them. In this case, Cameron was acting as an “anchor” in the following definition of “anchor” from dictionary.com:
“a person or thing that can be relied on for support, stability, or security; mainstay.”
Anchoring is also a behavioral concept that is defined as:
“the common human tendency to rely too heavily, or "anchor," on one trait or piece of information when making decisions.”[i]
Put another way:
“Anchoring or focalism is a cognitive bias that describes the common human tendency to rely too heavily on the first piece of information offered (the "anchor") when making decisions. During decision making, anchoring occurs when individuals use an initial piece of information to make subsequent judgments. Once an anchor is set, other judgments are made by adjusting away from that anchor, and there is a bias toward interpreting other information around the anchor. For example, the initial price offered for a used car sets the standard for the rest of the negotiations, so that prices lower than the initial price seem more reasonable even if they are still higher than what the car is really worth.[ii]
Apparently, there are three generally accepted potential mechanisms for why people might use what seems to be unrelated bits of information to make decisions that end up being highly correlated to these bits of information (i.e. the decision ends up being anchored to these bits of information).[iii]
- Anchoring may result from “an insufficient adjustment process” by the brain after being exposed to the information.[iv]In other words, the brain is stuck on the bit of information and has difficulty moving away from the bit of information.
- What is called the “attitude change” explanation states that “anchors could serve directly as a cue or indirectly influence the information processing that bias judgments towards the anchors.”[v] Put another way, the information itself suggests the answer.
- The “selective accessibility” theory posits that individuals may actually test whether a provided anchor is in fact the right answer. Briefly, the idea is that when given an anchor, the decision maker will evaluate whether the anchor is a suitable answer. Assuming it is not, the decision maker moves on to another guess, but not before accessing all the relevant attributes of the anchor itself. Then, when evaluating a new answer, the decision maker looks for similarities between the old and the new answer resulting in the anchoring effect.
Michael Jetter and Jay K. Walker recently explored the concept of anchoring in research published by IZA[vi] in August 2016 in a paper titled “Anchoring in Financial Decision-Making: Evidence from the Field.”[vii] Jetter and Walker looked at the 12,596 wagering decisions of 6,065 contestants in the Jeopardy! television game show. Specifically, they looked at how the bets related to the so-called Daily Double were influenced by the dollar amount of the clue associated with the Daily Double.
As probably everyone on the planet knows, Jeopardy! is a game played by three contestants in three rounds: the Jeopardy! round, the Double Jeopardy! round and the Final Jeopardy! round. In the first two rounds, there are six categories of clues with five clues in each category. Before November 26, 2001, the Jeopardy! round clue values were $100, $200, $300, $400 and $500 and the Double Jeopardy! round clue values were $200, $400, $600, $800 and $1000. Since then, the value of all the clues in these first two rounds has doubled.
In the first two rounds, one of the contestants, typically the returning champion, picks a category and the dollar amount of a clue. Then, the host (Alex Trebek), announces a ‘clue’ for that selection in the form of an answer. Whichever of the contestants is the first to press a buzzer has a chance to pose the correct question to which the ‘clue’ is the answer. If the contestant correctly provides the right question, they receive the dollar amount associated with that clue and are allowed to choose the next clue. If the contestant incorrectly answers the clue, the dollar amount of the clue is subtracted from their account balance and the other contestants have an opportunity to answer the clue.
There are three Daily Double clues in each episode: one in the Jeopardy! round and two in the Double Jeopardy! round. The Daily Double clues are hidden meaning that the contestants do not know which of the clues are associated with the Daily Double until after they choose a clue. If a player selects a Daily Double, they are able to place a wager on their correctly answering the corresponding clue. The amount of the wager is up to the larger of either their entire current account balance or the largest dollar value remaining on the current board. So, for example, if a contestant had $3500 in their account with a $2000 clue still on the board and selected a Daily Double clue, they could wager up to $3500 on their correctly answering the clue. Importantly, there is absolutely no correlation between the value of the Daily Double clue and the amount that could of should be wagered.
There is a lot at stake in the game! The contestant with the highest account balance at the end of the game gets to keep their account balance with the average winner earning $19,752 for each episode. Another key feature of the game is that there is both a large live and much larger television audience (estimated to be about 25 million viewers per week!). A further key feature of the game is that there is time pressure to answer clues and as we will see, decide the amount to wager in the Daily Double situations; the game is quite fast-paced!
Jetter and Walker discovered that “an extraordinary number of Daily Double wagers fall closely to the initial dollar amount.”[viii] In fact, although the maximum possible wager averages $5914, more than half of all the wagers fall within $500 and over 76% fall within $1000 of the corresponding clue value.[ix] This means that even correcting for other factors such as the size of the contestant’s account, whether they have and how large of a lead they have over other contestants and whether they have previously answered other questions in that category, the contestants still anchor to the dollar amount of the clue associated with the Daily Double.
Obviously, Jeopardy! is a television game and not necessarily representative of “real life.” But, interestingly, Jetter and Walker point to research done by Antonovics et al[x] that shows that results obtained in laboratory experiments (is Jeopardy! a lab experiment?? Would the contestants consider themselves to be lab rats in an experiment?) are comparable with those from “real-life” only when the stakes in the lab are greater than $50 per person. Of course, the average winnings of over $19,000 exceed $50. So, there is likely some real-life lessons that can be learned from Jeopardy!.
The key lesson learned by Jetter and Walker was that pressure affects anchoring. In Jeopardy! there is time pressure (the fast paced nature of the show forces rapid decision making) and pressure because a large studio audience and typically about 25 million television viewers are watching each episode. Because of either or both of these forms of pressure, the contestants displayed marked anchoring tendencies. Of course, if the time pressure and scrutiny was removed, the players would likely make better decisions. But, the excitement and fun of the game would then be compromised! Jetter and Walker suggest that people who often face these forms of pressure (e.g., CEOs and major decision makers in companies, gamblers, negotiators or those that participate in online auctions) take heed of this anchoring phenomena and take steps to mitigate its effect on their decisions.
Because Jeopardy! sometimes had children, adolescent and college student contestants, Jetter and Walker were also able to study whether and to what amount anchoring affected these contestants. They found little evidence of anchoring in children under 13. But, starting at about puberty, the contestants developed more and more anchoring tendencies until they reached college age where the anchoring tendencies approximated more “seasoned” adults.
I personally participated in an excellent example of anchoring albeit in a non-real-world situation. As part of my need to fulfill continuing legal education requirements for my law license, I attended a negotiation workshop put on by the noted scholar Charles B. Craver, now at The George Washington University Law School. (If you can attend one of his workshops, I HIGHLY recommend that you do! Otherwise, pick up his book Effective Legal Negotiation and Settlement.)
At one point in the workshop, Prof. Craven gave the group a few basic facts about a mythical dog bite incident where a man, out for a walk, was bitten by a dog, sustaining a few thousand dollars in medical damages, pain and suffering, lost work and rehabilitation. He then had the entire group of a few hundred attorneys count off by “one,” “two.” The “ones” he declared to be plaintiffs and the “twos” he declared to be defendants. He had his assistants pass out sheets of paper to each group of a “one” and a “two.” We were instructed to draw a line down the center of the paper and write a “P” above the column to the left of the line and a “D” above the column to the right of the line.
The plaintiffs were instructed to write an initial settlement offer at the top of the left column. Thereafter, the defendants were to write their counteroffer at the top of the right column. The plaintiff and defendant were to go back and forth exchanging counteroffers until a settlement was reached whereupon the agreed-upon settlement amount was circled. All of the negotiations were to take place without any words being exchanged. In other words, all the negotiations were to take place by just writing down proposed amounts in the respective columns.
After we finished this exercise, the papers were gathered and statistically processed while we took a break. (Those negotiations were tough! Ha!) When we returned from the break, Prof. Craver informed us that the average settlement value was around $50,000 as I recall. But, more importantly and related to the concept of anchoring, the larger the initial settlement offer by the plaintiff, the larger in a statistically valid way was the final agreed to amount! The opposite was true as well. If the plaintiff started with a low opening bid, the final settlement number was also lower. In either case, the opening bid established an anchor from which all subsequent negotiations were tied. The message to us was clear: if you are a plaintiff and wanted a large settlement for your client, start with a large initial offer. On the other hand, if you are a defendant, be aware that anchoring would likely take place and consciously try to remove this bias during the negotiating process.
So, what are the “take aways” from this research and anchoring in general?
- Anchoring appears to be a real behavioral phenomena. So, be aware of it and deal with it!
- Pressure in TIME and in VISIBILITY seems to increase anchoring. So, if you can, try to minimize time pressure and the visibility of your decisions. In other words, try to avoid snap decisions and making important decisions that are highly visible. If you can’t do either or both of these things, be aware that you may be influenced by unrelated data and consciously try to identify such bits of data and be sure they aren’t influencing your decisions.
- Anchoring appears not to be an inherent behavior. Instead, it is an acquired or developed behavior that develops from about 13 until it is fully developed at about college age.
There is obviously much more to anchoring and more particularly to its application in economics. A Google search on “the economic effect of anchoring” will provide many good references to explore! If you are wise, you can avoid its negative effects and use it to your advantage!
Check out our website by clicking on the button below!
[ii] https://en.wikipedia.org/wiki/Anchoring referencing Anchoring Definition, Investopedia, retrieved September 29,2015 (http://www.investopedia.com/terms/a/anchoring.asp?lgl=no-infinite)
[iii] Jetter, M. and Walker, J. K. (2016). Anchoring in Financial Decision-Making: Evidence from the Field. Retrieved from http://legacy.iza.org/en/webcontent/publications/papers/viewAbstract?dp_id=10151. p 4.
[v] Furnham, A. and Boo, H. C. (2011). A literature review of the anchoring effect. Journal of Socio-Economics, 40(1):35-42, p. 37
[vi] The Institute for the Study of Labor (German: Forschungsinstitut zur Zukunft der Arbeit) or IZA is a private, independent economic research institute and academic network focused on the analysis of global labor markets and headquartered in Bonn, Germany. https://www.iza.org/
[viii] Id. at p. 11.
[x] Antonovics, K., Arcidiacono, P. and Walsh, R. (2009). The effects of gender interactions in the lab and in the field. The Review of Economics and Statistics, 91(1):152-162.