“Anchoring” influenced customer decisions
GambleAware, an independent charity that commissions problem gambling prevention and treatment services in the United Kingdom, has published a report containing findings from a study on online gambling deposit limits. The study, conducted by the Behavioral Insights Team (BIT), found that pre-set deposit limit selections could result in gamblers setting higher deposit limits for themselves.
The test, as BIT tweeted on Wednesday, was to see if giving players large, pre-filled deposit limit options in a drop-down list “anchored” them at those higher limits:
By removing such pre-sets and instead allowing customers to enter whatever amount they wanted, BIT found that online gamblers slashed their own deposit limits by as much as 46%. The difference, the researchers concluded, is likely the result of the psychological phenomenon of “anchoring,” which has to do with allowing prior information to influence players’ choices.
Experiment structure
BIT invited 45,000 bet365 customers who had not set deposit limits to participate in the study. A total of 1,731 participated and voluntarily set a deposit limit. They were split into three groups. The first was “business-as-usual” (BAU), which used bet365’s normal system, choosing a deposit limit from a drop-down list ranging from £5 ($6.83) to £100,000 ($136,629). The second, “Lower Anchors,” was similar to the BAU group, except the highest pre-set number was £250 ($342). The third test group, “No Anchors,” did not have a drop-down list, instead just giving players a blank text box in which to freely enter their limit.
The raw data made it a bit difficult to settle on a conclusion, as some customers setting very high deposit limits – especially in the control BAU group – skewed the results. For instance, the median daily deposit for BAU customers was £14.30 ($19.54), while the mean was £1,601.10 ($2,187.57). The median and mean for the Lower Anchor group were £8.30 ($11.34) and £230.81 ($315.35), while the figures for those in the No Anchors (text box) group were £7.10 ($9.70) and £308.78 ($421.88).
Test groups deposited almost 50% less
To smooth out some of the variation and draw better comparisons, BIT used log transformation and regression analysis. As a result, using a baseline deposit limit of £1,601.10 for the BAU group, researchers estimated that this same theoretical gambler would set a limit of £866.50 ($1,183.89) if given the Lower Anchor drop-down list. If put in the No Anchors group and given just a blank text box, this same customer would set almost exactly the same limit as they would in the Lower Anchor group, £866.55 ($1,183.96).
those who set higher deposit limits had greater differences in the control group
BIT determined that the 46% and 45% lower deposit limits were statistically significant, despite the small sample sizes of each group. Researchers also found that those who set higher deposit limits had greater differences in the control group compared to the two treatment groups (Lower Anchor and No Anchor).
The team did a similar statistical analysis on the total amount of money players deposited. The control group was estimated to have deposited an average of £445.96 ($609.31) over the course of a month. The average for the Lower Anchor group was 4.4% lower (£426.37/$582.55) and the average for the No Anchor group was 18% lower (£360.78/$492.93). The importance here is that the difference between the No Anchor group and BAU group is about £1,020 ($1,394), which is close to the amount that families need to cover unexpected costs in a year. BIT stressed that these findings were not statistically significant and thus could just be the result of chance.