Perceived Costs and Benefits of Participating in Further Training

Last registered on March 06, 2024
View Trial History

Pre-Trial

Trial Information

General Information

Title
Perceived Costs and Benefits of Participating in Further Training
RCT ID
AEARCTR-0012479
Initial registration date
February 21, 2024

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
March 06, 2024, 3:08 PM EST

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

Locations

Country
Germany
Region

Primary Investigator

Name
Armando Miano
Affiliation
University of Naples Federico II
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Silke Anger
PI Affiliation
Institute for Employment Research (IAB) and University of Bamberg
Contact Investigator
PI Name
Ute Leber
PI Affiliation
Institute for Employment Research (IAB)
Contact Investigator
PI Name
Simon Janssen
PI Affiliation
Institute for Employment Research (IAB)
Contact Investigator
PI Name
Pascal Heß
PI Affiliation
Institute for Employment Research (IAB)
Contact Investigator

Additional Trial Information

Status
In development
Start date
2024-02-26
End date
2025-02-28
Keywords
Behavior, Education, Labor
Additional Keywords
JEL code(s)
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
As recent waves of technological change have considerably altered the demand for skills in the labor market, many scholars, policy makers, and practitioners emphasize the growing importance of on-the-job training. However, the training participation of individual workers hardly increased throughout the recent decade, and training rates are extremely heterogenous across workers and firms. Particularly, workers who commonly have high returns to training, such as low educated workers and workers in small firms, train on average less than high educated workers in larger firms.
A main reason might be that the heterogeneity in expected training costs and benefits might explain the heterogeneity in the training participation of individuals. Particularly, certain workers such as, for example, low educated and older workers who do not train might overestimate their costs or underestimate their benefits from training.
However, although expected training costs and benefits may have a stronger influence workers’ decision to participate in training than their realized training costs and benefits, virtually no study has explored workers’ expectations of the costs and benefits of training.
In this study, we analyze the causal effect of expected training costs and benefits on workers’ training activities. Therefore, we, first, elicit workers’ willingness to train and their expected training costs and benefits in a high frequency panel survey. Second, we implement two information treatments to shift workers’ expectations about their training cost and benefits.
We use this experimental data to study impact of expected training costs and benefits on their willingness to train and their actual training participation. We then link this survey data to detailed register data that allow us to follow the labor market careers of workers in the long run.
External Link(s)

Registration Citation

Citation
Anger, Silke et al. 2024. "Perceived Costs and Benefits of Participating in Further Training." AEA RCT Registry. March 06. https://doi.org/10.1257/rct.12479-1.0
Sponsors & Partners
Experimental Details

Interventions

Intervention(s)
Within a high-frequency survey (OPAL), administered by the German Institute for Employment Research (IAB), we elicit workers’ expectations of the costs and benefits of on-the-job training and their willingness to look for and take part in on-the-job training in the future. We also measure realized costs and benefits for those respondents who recently did some training (in the six months before the survey). We embed two information experiments in the survey, one providing information about the benefits of training and one providing information about the costs of training. After the main survey, participants will be recontacted for two short and one longer follow-up surveys (about 1-2 months, 6 months, and 9 months, respectively, after the intervention).
Intervention Start Date
2024-02-26
Intervention End Date
2024-09-30

Primary Outcomes

Primary Outcomes (end points)
Expectations of costs and benefits of on-the job training, intentions to look for training, intention to take part in training, actual participation in training.
Primary Outcomes (explanation)
Right after the experiment, in the same survey, we measure beliefs about the costs and benefits of on-the-job training in the following way:
1. For the costs, we ask participants to rate how costly a potential training would be in terms of time, financial cost, and personal effort. Participants can answer these questions using a likert scale from 1 to 7 where 1 is not costly at all and 7 is extremely costly.
2. For the benefits, we ask participants how likely it is, in percentage terms, that the potential training will increase their job satisfaction, increase the security of their current job, improve their job opportunities with other employers, or increase their wage.

We measure intentions to train both directly after the treatment in the same survey, and in three follow-up surveys. We record actual participation in training in the three follow-up surveys by asking respondents for any training they did since we last interviewed them.

In addition, in the same survey before the experiment, we measure baseline (pre-treatment) beliefs about the benefits and costs of training and past training experience. We use these questions to study how different people perceive the costs and benefits of on-the-job training at the baseline. We also plan to study how the treatment effects vary depending on these baseline beliefs on the costs and benefits of training,

1. We split the sample into respondents who have participated in on-the-job training during the past six months and respondents who have not participated in on-the-job training during the past six month.
2. We ask the non-participants to imagine a hypothetical on-the-job training course and give them questions on their expected costs and benefits that this training course would have for them. We focus on three types of costs: time, monetary, and psychological costs. For each cost type, we ask respondents to rate their expected costs for specific training aspects on a Likert scale from 1 to 7. For instance, we ask respondents about how time consuming they believe various phases of the training would be, from finding the suitable training course to preparing for and taking the exam. To elicit the expected benefits, we ask respondents how likely it is that the training will improve their condition along multiple dimensions, including wage, job satisfaction, job security, and job prospects with other employers.
3. Instead of asking about expected costs and benefits, we ask respondents who took part in a training course in the last six months about their realized training costs and benefits. We inquire about which benefits they gained from training and which benefits they believe will gain in the next five years, following a structure similar to the questions about the hypothetical training scenario.

We also plan to explore the heterogeneity in the treatment effects by respondents’ socio-economic characteristics (such as gender, education, age), occupation, location (local labor market) and past training experiences.

Secondary Outcomes

Secondary Outcomes (end points)
Employment, wages and career trajectories, work satisfaction and life satisfaction.
Secondary Outcomes (explanation)
We can observe respondents’ short-term labor market outcomes in later waves of the same high-frequency survey (OPAL), 6-9 months after the treatment. We can also link survey responses to administrative data – for those respondents who give their consent – where we can measure respondents’ labor market outcomes over a longer horizon. Thanks to all these data will be able to study whether the participation in training eventually induced by the treatments had any effects on respondents’ labor market outcomes – such as employment, wages, and career trajectories – over different horizons.

Furthermore, in one of the follow-up surveys we will measure respondents’ life and job satisfaction and study whether training participation had any impacts on those.

Experimental Design

Experimental Design
We embed two information experiments in the online survey.
We split respondents in 3 randomly selected and mutually exclusive groups: Treatment 1, Treatment 2 and Control.

Participants in the Treatment 1 group receive information about the benefits of on-the-job training in terms of wages, job security, and life/job satisfaction. In more detail, we use the German National Education Panel Survey (NEPS) to calculate average differences between wages, employment, and job/life satisfaction of workers who trained in the past and workers who did not train in the past. Afterwards, we show these statistics to the treated individuals in our benefit treatment.
Participants in the Treatment 2 group receive information about average training costs. In more detail, the treatment includes statistics based on a previous survey from the Bibb (German Federal Institute for Vocational Education and Training) about the share of workers who were able to train during their working time, the share of workers who did not find it strenuous to find suitable training courses, and the share of workers who did not experience financial restrictions in response to their training participation.
In both treatments the information is presented using a combination of text, graphs, and pictures. The control group receives neither information on training benefits nor information about training costs.
Participants in the control group does not receive any information.
Experimental Design Details
Not available
Randomization Method
As they take the survey, respondents are classified in three groups according to their highest level of education – no professional degree, vocational degree, college degree. Within each group they are randomly assigned to one of the treatment groups or to the control group by the survey software (Ingress).
Randomization Unit
Individual respondent.
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
1 country (Germany).
Sample size: planned number of observations
8,000 wage and salary workers not in education or in vocational education and training (VET), not in retirement (civil servants and self-employed not included).
Sample size (or number of clusters) by treatment arms
We are going to split respondents equally among the control group and the two treatment groups. Hence, we expect to have about 2,600 respondents in each treatment arm (control, benefit treatment, cost treatment).
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
IAB Ethics Commission
IRB Approval Date
2024-02-21
IRB Approval Number
24_002
Analysis Plan