Gamified Savings as a Problem Gambling Intervention

The intervention focuses on a soccer prize-linked savings product – a combination of a savings account and a lottery based on soccer outcomes. Typical prize-linked savings account functions as follows: potentially win a huge jackpot, without having to risk losing money. This intervention may appeal to bettors with an appetite for lottery-like products, and provide a mechanism for increasing savings and reduce gambling expenditure. We argue that betting is a bundled product that comprises of both fun and the opportunity for skewness of payouts. Gamification comes from making predictions on certain matches, which is the fun component in the game.

2023-08-01

2023-11-30

Gamified soccer prize-linked savings and gambling expenditure

Actual savings and gambling expenditure

This is a lab-in-the-field experiment.

Participants will be randomly assigned into one of four treatment groups as follows:

1. Low Variance (Non Skewness) – No Gamification: Simple interest savings product (at the prevailing market rates). This is the control group
2. High Variance (Skewness) – No Gamification: Prize-linked savings product with the same expected return as the first group, but the potential for a large prize is determined by a random drawing. The principal amount is guaranteed but the interest on the principal goes to the lottery instead of being added to the principal.
3. Low Variance (Non-skewness) – Gamification: Soccer-based prize-linked savings product where there is potential for a small prize that is determined by the outcome of the individual's predictions on upcoming soccer matches.
4. High Variance (Skewness) – Gamification: Soccer-based prize-linked savings product where there is potential for a large prize that is determined by the outcome of the individual's predictions on upcoming soccer matches.

Not available

Randomization will be done in the office by a computer.

Individuals are the unit of randomization.

No

None.

2000 individuals (1000 from Kenya and 1000 from Uganda).

1000 control, 1000 treated (split in half for each country i.e., Kenya and Uganda)

Based on a conservative power calculation, this study will use a sample size of 2000 individuals split evenly between Kenya and Uganda. The power calculation assumes a 5 percent level of statistical significance and a statistical power of 80 percent. The power calculation further assumes a monthly gambling expenditure of KES 1720/Uganda Shillings [UGX] 53,000 (i.e. US$14). These benchmark values for the outcome indicators are from the 2021 Financial Access Household Survey conducted in Kenya that included monthly gambling expenditure from a sample of sports bettors. This sample size will enable the study to detect a minimum of KES 305/UGX 9,500 (i.e. US$3) decrease in monthly gambling expenditure, and an increase of an equivalent amount for monthly savings. The formula used for determining the sample size is as follows: Parameter Value Definition α 0.05 Significance level β 0.8 Desired power of the test Tail 2 One-tailed or two-tailed test n_min 1000 The minimum sample size decided beforehand to determine t values σ_y 1720 The pooled total standard deviation of the estimated effect on the outcome variable P 0.5 The proportion of the study that is randomly assigned to the treatment group δ 305 Minimum detectable effect t_1 1.96 T-value corresponding to the desired significance level of the test t_2 0.84 T-value corresponding to the desired power of the design n 1000 The size of the study sample