What’s up without WhatsApp? Experimental evidence on social capital and subjective well-being

Last registered on May 02, 2026

View Trial History

Pre-Trial

Trial Information

General Information

Title

What’s up without WhatsApp? Experimental evidence on social capital and subjective well-being

RCT ID

AEARCTR-0017560

Initial registration date

April 13, 2026

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published

April 14, 2026, 9:40 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Last updated

May 02, 2026, 9:49 AM EDT

Last updated is the most recent time when changes to the trial's registration were published.

Locations

Country

Indonesia

Region

Primary Investigator

Name

Pyan Muchtar

Affiliation

Australian National University

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Budy Resosudarmo

PI Affiliation

Australian National University

Contact Investigator

Additional Trial Information

Status

In development

Start date

2026-05-01

End date

2026-09-30

Keywords

Behavior, Other

Additional Keywords

Social capital, Subjective well-being

JEL code(s)

D90, I31, O33, Z13

Secondary IDs

Prior work

This trial does not extend or rely on any prior RCTs.

Abstract

How social ties are formed, maintained, and mobilized has changed with the adoption of digital communication technologies like WhatsApp. This paper estimates the value of communication infrastructure that sustains social networks and examines the short-run causal effects of losing partial access to it on social capital and subjective well-being. We conduct a field experiment in three remote island communities in Indonesia, recruiting approximately 1,200 household heads as participants. To elicit individuals’ willingness to accept a partial daily restriction on WhatsApp usage, we implement the double-bounded dichotomous choice (DBDC) method. Participants are then randomly assigned to a treatment group, in which WhatsApp access is restricted for 4 hours per day over 6 days, or to a control group with no restriction. We compare outcomes across groups along three domains: social capital, subjective well-being, and time use. We further explore the mechanisms underlying these effects by examining changes in time use and daily behavioral patterns during the restriction period.

External Link(s)

Registration Citation

Citation

Muchtar, Pyan and Budy Resosudarmo. 2026. "What’s up without WhatsApp? Experimental evidence on social capital and subjective well-being." AEA RCT Registry. May 02. https://doi.org/10.1257/rct.17560-2.0

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

Participants are randomly assigned with equal (1:1) probability to a treatment or control group. Treated participants have their WhatsApp restricted for 4 hours per day over 6 consecutive days using a third-party app installed on their smartphones by enumerators. Access is blocked starting the next day of the baseline survey. Participants are compensated upon completion of the endline survey. Participants in the treated group are compensated with an amount equal to the last offer they accepted during the baseline willingness-to-accept elicitation. Control participants face no restriction on their WhatsApp usage throughout the study period.

Intervention (Hidden)

Prior to randomization, all participants complete a Double-Bounded Dichotomous Choice (DBDC) willingness-to-accept (WTA) elicitation to determine their minimum acceptable compensation for a one-week WhatsApp restriction. Initial bids are randomly drawn from the set IDR 150,000, IDR 175,000, IDR 200,000, IDR 225,000, and IDR 250,000. Eligibility for the experiment requires active WhatsApp use in the past week and acceptance of at least one bid in the DBDC sequence. Eligible participants are randomized within the survey application immediately after eligibility is confirmed, using a fixed random seed. Treated participants are informed of their assignment, asked to reconfirm consent, and have the Freedom app installed on their smartphones to block WhatsApp for seven days. Compliance is verified at the endline visit by inspecting the app's restriction log. Treated participants who withdraw before endline receive pro-rated compensation. All participants are surveyed at baseline and at endline one week later, with endline surveys administered simultaneously to both groups.

Intervention Start Date

2026-05-03

Intervention End Date

2026-09-29

Primary Outcomes

Primary Outcomes (end points)

(1) Social capital index; (2) Subjective well-being index (Indonesian Well-being Scale, IWS)

Primary Outcomes (explanation)

Social capital index: Composed of three dimensions : bridging social capital (5 items, Williams 2006 scale), bonding social capital (5 items, Williams 2006 scale), and maintained social capital (5 items, Ellison et al. 2007 scale adapted for old friends). Each dimension score is the mean of its component items measured on a 5-point Likert scale. The index is the simple average of the three dimension scores. For robustness, we also construct an inverse-covariance weighted index following Anderson (2008).
Subjective well-being index: Measured using the 20-item Indonesian Well-being Scale (IWS; Maulana et al. 2019), covering dimensions of self-acceptance, basic needs, family relations, spirituality, and harmony with the environment. Each item is measured on a 5-point Likert scale. The index is the simple average of dimension-level scores (each dimension score being the mean of its items), with an Anderson (2008) inverse-covariance weighted index as a robustness check. All items are oriented so higher values indicate better outcomes before index construction.

Secondary Outcomes

Secondary Outcomes (end points)

(1) Life satisfaction; (2) Happiness; (3) social media time use; (4) Time use across daily activities (work, volunteering, exercise, socializing); (5) Daily behavioral patterns (self-reported battery of behavioral change statements)

Secondary Outcomes (explanation)

Experimental Design

This study is a two-arm randomized controlled trial conducted across three remote island communities in Indonesia: Karimunjawa (Central Java), Sumba (East Nusa Tenggara), and Banda (Maluku). Approximately 1,200 household heads are recruited as participants. Prior to randomization, each participant completes a Double-Bounded Dichotomous Choice (DBDC) willingness-to-accept (WTA) elicitation for a one-week WhatsApp restriction.
Eligible participants (those who use WhatsApp actively and accept at least one bid) are randomized with equal probability to a treatment group (partial daily WhatsApp restriction for 6 days) or a control group (no restriction). Both groups are surveyed at baseline and endline one week later. The primary analysis uses intent-to-treat (ITT) estimation via OLS with baseline outcome controls and fixed effects, following the ANCOVA specification. Fieldwork is conducted sequentially: Karimunjawa (1 May - 13 June 2026), Sumba (20 May–29 June 2026), and Banda (18 August–29 September 2026).

Experimental Design Details

Randomization Method

Randomization is performed by computer within the Survey Solutions CAPI application at the time of the baseline interview, immediately after eligibility is confirmed. Equal assignment probabilities (1:1) are applied across all eligible participants within each study site.

Randomization Unit

Individual (household head). Treatment is assigned at the individual level, with treated and control participants interspersed across the same villages.

Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters

Not applicable (individual-level randomization). 3 study sites (islands).

Sample size: planned number of observations

Approximately 1,200 household heads recruited; approximately 960 expected to be eligible (80% eligibility rate); approximately 864 expected to complete both baseline and endline surveys (90% completion rate among eligible).

Sample size (or number of clusters) by treatment arms

Approximately 480 participants treatment (WhatsApp restriction), 480 participants control (no restriction), from the expected eligible sample of 960 participants. Allocation ratio 1:1.

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

Under the planning assumption of 80% eligibility and 90% completion, the effective experimental sample is 864 participants (432 per arm). The cross-sectional MDE for the primary outcomes (social capital index and subjective well-being index) is approximately 0.19 SD (80% power, 5% two-sided significance level, balanced design). Exploiting the panel structure following McKenzie (2012), the panel-adjusted MDE is approximately 0.17–0.18 SD, assuming an autocorrelation of ρ ∈ [0.3, 0.5] between baseline and endline outcomes. Effects are expressed in standard deviation units.

Supporting Documents and Materials

IRB