Let me choose my incentives: A mindfulness intervention

We test the effect of two monetary incentive schemes intended to support the take-up and maintenance of an online mindfulness meditation program. The program, inclusive of incentives, lasts for 36 days and is based on daily meditation sessions which last between 5 and 15 minutes. Subsequently, participants continue to have access to meditation sessions for two additional months but receive no additional payment.
Subjects have access to two types of incentive schemes: (i) Constant scheme, where they receive 2 euro for each complete meditation session; and (ii) Dynamic scheme, where subjects receive 8 euro for each sequence of three completed meditation sessions (3 days in a raw).
Subjects are randomized into 3 treatments: (1) control, (2) exogenous incentives, where one of the two incentive schemes is randomly assigned; and (3) endogenous incentives, where subjects can choose, between the two incentive schemes, the one that they prefer.

2019-11-04

2020-04-30

• Number of completed meditation sessions during the intervention period.
• Indicator variables for assigned treatments (1: Control, 2: Exogenous incentives, 3: Endogenous incentives).
• Indicator variables for incentive schemes (1: Static; 2: Dynamic).

• Standardized self-reported mindfulness level. We will measure mindfulness level at baseline and end-line with the “Mindfulness attention awareness scale” (Brown, K.W. & Ryan, R.M., 2003); and in particular the Dutch version by Schroevers, M., Nykliček, I., & Topman, R., 2008].

• Standardized self-reported stress levels. We will measure stress levels at baseline and end-line with the Perceived Stress Scale (Cohen, S., Kamarck, T., & Mermelstein, R., 1983).

• Standardized meditation motivation measure at baseline. Elicited with six questions extracted from the Intrinsic Motivation Inventory scale (Ryan, R. M., 1982).
• Past meditation frequency at baseline. Self-reported typical number of meditation sessions per week, on a typical week prior to the study.
• Meditation goals at baseline. Desired number of meditation sessions per week for the near future (before knowing about incentives).

• Standardized desirability of control measure at baseline. Elicited with six questions extracted from the Desirability of Control Scale by Burger and Cooper (1979).

• Risk aversion measure (Gneezy and Potters, 1997).

• Beliefs at baseline about future completed meditation sessions, given assigned treatment. Participants in the endogenous treatment answer the belief questions both for the incentive scheme they choose and did not choose. Subjects are asked to guess how many meditation sessions they think they will complete during the study period. Incentivization: Subjects receive 1 euro if their guess is correct.

• Standardized academic self-concept measured at baseline and endline. Elicited with six questions extracted from the Academic self-concept scale (REF) by Reynolds (1988).

• Standardized self-esteem measured at baseline and endline with the “Self-Esteem Scale” by Rosenberg (1965), and in particular the Dutch version by Franck et al. (2008).

• Time preferences measure elicited at endline.

• Past meditation frequency and future meditation goals elicited at baseline.

• Completed meditation sessions in the 2 months following the end of the intervention.

• Demographic questions (age, gender, study program).

We test the effect of two monetary incentive schemes intended to support the take-up and retention of an online mindfulness meditation program. The program, inclusive of incentives, lasts for 5 weeks and is based on daily meditation sessions which last between 5 and 15 minutes. Subsequently, participants continue to have access to meditation sessions for two additional months but receive no additional payment.
Subjects have access to two types of incentive schemes: (i) Constant scheme, where they receive 2 euro for each complete meditation session; and (ii) Dynamic scheme, where subjects receive 8 euro for each sequence of three complete meditation sessions.
Subjects are randomized into 3 treatments: (1) control, (2) exogenous incentives, where one of the two incentive schemes is randomly assigned; and (3) endogenous incentives, where subjects can choose, between the two incentive schemes, the one that they prefer.

We exclude from the sample (and do not count towards the total of 160 subjects per treatment):
- Non-students who accidentally obtain access to the registration link (we will verify the student status at the moment of payment)
- Participants who register and do not reach the end of the baseline survey.

Depending on our ability to reach out to enough subjects on the first recruiting round, we may have to recruit participants in multiple waves.

Randomization by computer software (Qualtrics).

Individual

No

no clusters

about 480 participants

We will have about 160 participants per treatment.
Subjects in the exogenous treatment are further randomized into the dynamic scheme (60%) and the static scheme (40%). The percentages are chosen to match the distribution of subjects in the endogenous treatment from pilot data collected prior to the intervention.