Back to History Current Version

Skills Mismatch: Sources and Consequences

Last registered on July 10, 2024
View Trial History

Pre-Trial

Trial Information

General Information

Title
Skills Mismatch: Sources and Consequences
RCT ID
AEARCTR-0013932
Initial registration date
July 03, 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
July 08, 2024, 1:13 PM EDT

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

Last updated
July 10, 2024, 1:39 AM EDT

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

Locations

Country
Thailand
Region

Primary Investigator

Name
Suphanit Piyapromdee
Affiliation
University College London
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Suphanit Piyapromdee
PI Affiliation
University College London
Contact Investigator
PI Name
Weerachart Kilenthong
PI Affiliation
Research Institute for Policy Evaluation and Design, University of the Thai Chamber of Commerce
Contact Investigator
PI Name
Ilse Lindenlaub
PI Affiliation
Yale University
Contact Investigator

Additional Trial Information

Status
In development
Start date
2024-07-15
End date
2029-08-31
Keywords
Education, Labor
Additional Keywords
Mismatch, Abilities, Growth Mindset, Occupational Choices
JEL code(s)
J24, J62, J64
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
The project will study sources and consequences of skills mismatch, a phenomenon where workers’ skills differ from job skill requirements. Skills mismatch can have a negative effect on workers’ income and career progression in both short and long terms, particularly for young workers.

Past studies have found that skills mismatch may be caused by (i) workers not fully knowing their own skill levels; (ii) lacking of labour market information, for example, different occupations may require different skill levels; (iii) lacking of confidence and adherence to social norms or traditional gender stereotypes. For example, female workers may perceive that a programmer is not a suitable job for women.

Importantly, in-depth information has never been collected to analyse the causes of skills mismatch in Thailand. Therefore, the research team is interested in studying the role of these three factors on occupational choices among graduates in Thailand. The research team expects that this study will shed light on policy solutions to the skills mismatch issue, and that the findings will be useful in designing policies to raise workers' incomes, promote labour market efficiency and reduce social inequality.

The project will use a randomised controlled trial (RCT) approach. The RCT involves giving participants who are final-year college students in Thailand different treatments. The three main treatments are as follows:

1. Providing individual ability: the intervention will evaluate skill levels using tests of quantitative, language (thai) and noncognitive skills. The test results will be presented to the participants in this treatment group. Each participant receives this result in private and individually.

2. Providing information on labour market: the intervention will use currently available data of labor market demand and present the results to the participants. Each participant will receive personalised information about the returns and skill requirements of occupations they are interested in as well as broad information of the current labor market environment.

3. Promoting Growth Mindset: the intervention will produce video media that provide training of a growth mindset in a job search process context. All participants in this treatment will receive the same video.

Treatments are randomized at the individual level. This experiment has 8 experimental groups: 7 treatment groups (G1-G7) and 1 control group (G8).

To measure the treatment effects, we will collect information using an online survey, before and after the interventions, on labour market outcomes (earnings, occupations, job search behaviour, job satisfaction), subjective beliefs (regarding labour markets and own skills), individual characteristics (gender, family background and skills). These are standard variables commonly used in labour economics research.

There are three sets of survey. The baseline survey will measure the variables mentioned above before the interventions. We will then follow these individuals’ outcomes using two sets of endline surveys afterwards. The first endline survey will measure how beliefs change over time each year. The second endline survey will biennially ask about the actual labour market outcomes.

Using the collected data, the project will analyse sources, mechanisms, and consequences of skills mismatch in an empirical framework. Results will be aggregated such that individuals cannot be identified.
External Link(s)

Registration Citation

Citation
Piyapromdee, Suphanit et al. 2024. "Skills Mismatch: Sources and Consequences." AEA RCT Registry. July 10. https://doi.org/10.1257/rct.13932-1.1
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)
The project will use a randomised controlled trial (RCT) approach. The RCT involves giving participants who are final-year college students in Thailand different treatments. The three main treatments are as follows:

1. Providing individual ability: the intervention will evaluate skill levels using tests of quantitative, language (thai) and noncognitive skills. The test results will be presented to the participants in this treatment group. Each participant receives this result in private and individually.

2. Providing information on labour market: the intervention will use currently available data of labor market demand and present the results to the participants. Each participant will receive personalised information about the returns and skill requirements of occupations they are interested in as well as broad information of the current labor market environment.

3. Promoting Growth Mindset: the intervention will produce video media that provide training of a growth mindset in a job search process context. All participants in this treatment will receive the same video.

Treatments are randomized at the individual level. This experiment has 8 experimental groups: 7 treatment groups (G1-G7) and 1 control group (G8).

We target fourth-year (final-year) college students in general majors such as arts and humanities. The age range is based on what the age range of final-year college students in Thailand.
Intervention Start Date
2024-08-01
Intervention End Date
2029-08-31

Primary Outcomes

Primary Outcomes (end points)
Types of jobs applied; Job search behaviours; Perception on job prospects; Employment status; Sequential earnings; Perception on own abilities
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
The target population are fourth-year (final-year) undergraduate students in public and private universities in Bangkok and surrounding areas. The sample includes students in faculties and fields that are not considered specialized in terms of occupations, such as humanities, economics, business administration, science, and engineering. We exclude specialised majors such as medicine, dentistry, and law. Approximately, the number of students taking part in the first survey is 5,500 people.

There are 8 groups of programs that voluntary participants. The assignment of participants into each group is conducted on the basis of randomisation. Each group will take part in each different program’s curriculum and activities.

The randomisation is conducted at the individual level.
Experimental Design Details
Not available
Randomization Method
Randomisation is done by a computer after the invited participants will have agreed (and have seen the information sheet and the consent form) to continue taking part in the baseline survey in the first meeting.
Randomization Unit
The randomisation is conducted at the individual level.
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
There is no cluster. However, we start by randomly selecting the targetting universities/higher institutions from the full population of higher institutions in and around Bangkok Metropolitan. The randomisation is stratified by the type of the institutions (public or private).
Sample size: planned number of observations
The sample size is based on statistical calculation. Because the randomization unit for this experiment is at the individual level, therefore, if we aim to test the hypothesis at the significance level 0.05, with an effect size of approximately 0.20 SD, the size of sample in each experimental group with statistical power is approximately 0.80 is approximately 400 people. That means the overall sample size should be approximately 3,200 people at the end of the project. We anticipate that there will be approximately 10 percent of the sample attrition annually. Therefore, the size of sample students in the first survey should be approximately 5,500 people so that in the final year (the next 5 years) there will still be about 3,200 people in the final round.
Sample size (or number of clusters) by treatment arms
The sample size is based on statistical calculation. Because the randomization unit for this experiment is at the individual level, therefore, if we aim to test the hypothesis at the significance level 0.05, with an effect size of approximately 0.20 SD, the size of sample in each experimental group with statistical power is approximately 0.80 is approximately 400 people. That means the overall sample size should be approximately 3,200 people at the end of the project. We anticipate that there will be approximately 10 percent of the sample attrition annually. Therefore, the size of sample students in the first survey should be approximately 5,500 people so that in the final year (the next 5 years) there will still be about 3,200 people in the final round.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
The sample size is based on statistical calculation. Because the randomization unit for this experiment is at the individual level, therefore, if we aim to test the hypothesis at the significance level 0.05, with an effect size of approximately 0.20 SD, the size of sample in each experimental group with statistical power is approximately 0.80 is approximately 400 people. That means the overall sample size should be approximately 3,200 people at the end of the project. We anticipate that there will be approximately 10 percent of the sample attrition annually. Therefore, the size of sample students in the first survey should be approximately 5,500 people so that in the final year (the next 5 years) there will still be about 3,200 people in the final round.
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
the UCL Humanities, Arts, and Sciences Research Ethics Committee (HAS REC)
IRB Approval Date
2024-06-24
IRB Approval Number
16083.004
IRB Name
Human Research Ethic Committee, University of the Thai Chamber of Commerce
IRB Approval Date
2024-02-19
IRB Approval Number
UTCCEC/Expedited005/2025
Analysis Plan