Effects of Childcare-Experience Training for Managers of SMEs: A Randomized Controlled Trial in Japan

Last registered on July 27, 2026
View Trial History

Pre-Trial

Trial Information

General Information

Title
Effects of Childcare-Experience Training for Managers of SMEs: A Randomized Controlled Trial in Japan
RCT ID
AEARCTR-0019239
Initial registration date
July 24, 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
July 27, 2026, 6:59 AM EDT

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

Locations

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Primary Investigator

Name
Shintaro Yamaguchi
Affiliation
University of Tokyo
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Mari Tanaka
PI Affiliation
Graduate School of Economics
Contact Investigator
PI Name
Hiroko Okudaira
PI Affiliation
Graduate School of Business, Doshisha University
Contact Investigator
PI Name
Chang Yang
PI Affiliation
Graduate School of Economics, The University of Tokyo
Contact Investigator

Additional Trial Information

Status
On going
Start date
2026-05-16
End date
2028-03-31
Keywords
Firms & Productivity, Gender, Labor
Additional Keywords
Child Care, Labor Management, Training, Time Allocation and Labor Supply
JEL code(s)
J13, J16, J22, M53, M54
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
This study evaluates the effects of a management training program combined with a one-day childcare-center immersion experience on managers at small and medium-sized enterprises (SMEs) in Fukushima Prefecture, Japan. The project is implemented jointly by the University of Tokyo, Doshisha University, and ten municipal governments in the prefecture, in partnership with the NPO "Fathering Japan (FJ)."

The intervention has two linked components: (1) an online management-training session providing concrete knowledge and practical recommendations on family-friendly workplace management; and (2) an in-person childcare-center immersion in which managers directly observe and interact with toddlers. The guiding hypothesis is a causal chain: hands-on childcare exposure heightens managers' recognition of the practical constraints of childcare (raising their perceived burden of caregiving); this shifts their support for family-supportive workplace management and their attitudes toward fathers' caregiving and gender roles; and these shifts translate into actual family-supportive management behavior. The childcare immersion is hypothesized to be the key driver of the translation from knowledge/attitudes into behavior.

Municipalities recruit firms, each enrolling up to two managers (executives or department heads). After a baseline survey of managers and firm HR contacts, firms are paired in order of enrollment within each municipality and randomly assigned within pairs to one of two cohorts: Group A receives the program in 2026 (treatment), Group B receives the identical program one year later in 2027 (delayed / waitlist control). All participants and firms are surveyed again roughly 1–3 months after their program (Follow-up 1) and before the next cycle. A subset of employees is surveyed as a complementary, non-incentivized, employer-blind data source. The between-arm comparison in the first period (A vs B in 2026, while B is untreated) is the primary design, maintained under a no-interference (SUTVA) assumption (see the PAP); the two-cohort structure additionally permits a staggered-adoption / event-study analysis using both periods.

All survey data are collected via Qualtrics using anonymized participant IDs not linked to personal identifiers or employee ID numbers; individual responses go directly to the University of Tokyo and are never shared with participants' employers.
External Link(s)

Registration Citation

Citation
Okudaira, Hiroko et al. 2026. "Effects of Childcare-Experience Training for Managers of SMEs: A Randomized Controlled Trial in Japan." AEA RCT Registry. July 27. https://doi.org/10.1257/rct.19239-1.0
Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
Experimental Details

Interventions

Intervention(s)
Treatment has two linked components delivered to participating managers by Fathering Japan: (1) an online management-training session on supporting employees' work-family balance; and (2) a one-day in-person childcare-center immersion experience (observation of and interaction with toddlers, followed by a structured reflection with childcare staff). Group A receives both components in 2026 (training in July; childcare-center experience August–October 2026). Group B receives the identical program one year later and serves as the control group during 2026.
Intervention Start Date
2026-07-27
Intervention End Date
2027-11-15

Primary Outcomes

Primary Outcomes (end points)
The following 4 variables measured at the individual manager level.
- PO1. Recognition of childcare constraints
- PO2. Support for family-supportive workplace management
- PO3. Attitudes toward fathers' involvement in childcare and gender roles
- PO4. Family-supportive management behavior
Primary Outcomes (explanation)
PO1. Recognition of childcare constraints — 5 quantitative-estimate items (manager survey D2.a–D2.e: a caregiving parent's available free time alone / with a partner present; frequency of daycare sick-child call-outs; time to feed and clean up after a 2-year-old; number of 2-year-olds one adult can safely supervise). Coded so that higher = stronger recognition of the time constraints, unpredictability, and labor-intensity of infant/toddler care; standardized by the control-group baseline mean and SD and averaged with equal weights. *Directional hypothesis: this population underestimates the practical burden of childcare at baseline, so the intervention is predicted to increase recognition of constraints.

PO2. Support for family-supportive workplace management — 4 items (manager survey B3.2, B3.4, B3.6, B4.5), coded so higher = more supportive; standardized (control baseline) and equal-weight averaged.

PO3. Attitudes toward fathers' involvement in childcare and gender roles — 5 items (manager survey B4.1–B4.3, B4.6–B4.7), coded so higher = more supportive of fathers' involvement / less traditional; standardized (control baseline) and equal-weight averaged.

PO4. Family-supportive management behavior — 8 binary items (manager survey C5.2–C5.9), each 1 = took the action in the past 3 months, 0 = did not; index = share of items answered "yes." For effects in SD units, the composite share is standardized by the control-group baseline SD (individual binary items are not standardized before averaging).

Secondary Outcomes

Secondary Outcomes (end points)
- SO1. Work–family support information-search index
- SO2. Support for employees with childcare responsibilities
- SO3. Communication and support-activation index
- SO4. Individual work-accommodation index
- SO5. Work redesign and organizational-resilience index
- Additional firm survey outcomes: male and female employee childcare-leave uptake rates; availability and male usage of non-statutory childcare-related leave; availability and generosity of family-friendly HR policies (shortened hours, flextime, telework, sick-child leave, overtime exemption); overtime/long-hours measures; implementation of general work-style-reform initiatives.
Secondary Outcomes (explanation)
- SO1. Work–family support information-search index (firm survey; Q16.1–Q16.2).
- SO2. Support for employees with childcare responsibilities (firm survey; Q16.4, Q16.7). Analysis population: firms with at least one employee with a preschool- or elementary-school-age child at baseline.
- SO3. Communication and support-activation index (manager survey; C5.2, C5.7, C5.8).
- SO4. Individual work-accommodation index (manager survey; C5.3, C5.4). Analysis population: same baseline childcare-employee restriction as SO2.
- SO5. Work redesign and organizational-resilience index (manager survey; C5.5, C5.6, C5.9).
- Additional firm survey outcomes (firm survey): male and female employee childcare-leave uptake rates; availability and male usage of non-statutory childcare-related leave; availability and generosity of family-friendly HR policies (shortened hours, flextime, telework, sick-child leave, overtime exemption); overtime/long-hours measures; implementation of general work-style-reform initiatives. The male childcare-leave uptake rate is a policy-relevant behavioral outcome but is expected to be substantially underpowered at the establishment level.

Experimental Design

Experimental Design
A firm-level randomized rollout with a delayed-treatment (waitlist-control) design. Within each of the 10 participating municipalities, enrolled firms are paired in order of application. Within each pair, one firm is randomly assigned to Group A (program in 2026) and the other to Group B (identical program in 2027, serving as a waitlist control during 2026). Because Group B eventually receives the same intervention, this is a delayed-treatment (staggered-adoption) design; the primary between-arm contrast is the 2026 first period, when B is untreated (maintained under a no-interference assumption; the PAP pre-registers a spillover/anticipation diagnostic).
Experimental Design Details
Not available
Randomization Method
Within each municipality, firms are ordered by application sequence and grouped into consecutive pairs (1st–2nd firm = pair 1, 3rd–4th = pair 2, etc.). Within each pair, one firm is randomly assigned to Group A and the other to Group B. No other stratification variables (e.g., firm size, industry) are used. Randomization is executed centrally by the research team at the University of Tokyo using an archived R script with a recorded random seed (for reproducibility); municipalities do not execute or influence assignment. Assignment is determined at the point of firm enrollment — before the firm's baseline survey is answered — and proceeds on a rolling basis as firms enroll. If a municipality has an odd number of enrolled firms, the final unpaired firm is assigned singly by an independent random draw (having no pair-mate, it is handled via the municipality-fixed-effects variant in analysis; see the PAP). At registration, allocation is still in progress.
Randomization Unit
Establishment (firm for single-establishment firms).
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
~200 establishments.
Sample size: planned number of observations
~350 individual managers (1–2 per firm); ~200 firm-level respondents (1 per firm); ~400 employees (complementary survey).
Sample size (or number of clusters) by treatment arms
Group A (2026): ~100 firms / ~175 managers. Group B (2027, delayed): ~100 firms / ~175 managers.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Prospective calculation (alpha = 0.05, power = 0.80, two-sided). For the individual-level (manager) primary outcomes (standardized indices; ~175 per arm, small within-firm clustering), the MDE is approximately 0.31 SD without covariate adjustment and approximately 0.22–0.26 SD with baseline covariate adjustment (assuming the baseline predicts the follow-up outcome with R2~ 0.3–0.5); with ~20% attrition these become roughly 0.25–0.29 SD. Establishment-level secondary outcomes (~100 per arm) and the **male childcare-leave-uptake rate** (few eligible male employees per SME per year) are materially less well powered (MDE ~> 0.31–0.44 SD for the firm indices) and are interpreted accordingly. The research team's own recruitment effort is exhausted, though the municipalities continue enrolling firms up to the intervention deadline (so the final N may still rise modestly); these figures are treated as design constraints, not targets, and will be updated with the realized sample. Final MDEs will be updated using the realized baseline standard deviations, intra-cluster correlation, R², and attrition rates once baseline data collection is complete, before any follow-up data are observed. Detailed assumptions are in the attached PAP.
Supporting Documents and Materials

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
IRB

Institutional Review Boards (IRBs)

IRB Name
The University of Tokyo, Graduate School of Economics, Research Ethics Review Committee.
IRB Approval Date
2026-05-12
IRB Approval Number
E2026ALS032 (application receipt no. E26ALS0289).
Analysis Plan

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information