The shadow education market in Portugal: A Discrete Choice Experiment

How to cite

Peralta, Susana; Freitas, Pedro; Carvalho, Bruno P.. 2026. "The shadow education market in Portugal: A Discrete Choice Experiment." AEA RCT Registry. AEARCTR-0018418. https://doi.org/10.1257/rct.18418-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Susana Peralta* , Pedro Freitas , Bruno P. Carvalho

* Primary investigator

Registry ID
AEARCTR-0018418
DOI
https://doi.org/10.1257/rct.18418-1.0
First registered
Last updated
Status
In development
Trial dates
2026-04-21 – 2027-04-30
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/18418

Abstract

This study uses a discrete choice experiment (DCE) to investigate parental preferences in the market for private tutoring for students in upper secondary education (10th to 12th grades). Parents or legal guardians are presented with a series of hypothetical tutoring options that vary across key attributes. Each respondent completes eight choice tasks drawn from a larger experimental design of sixteen choice sets.

The DCE is implemented through an online questionnaire administered to a representative sample of the population, stratified by NUTS II regions and maternal educational. The study aims to estimate preferences for tutoring characteristics and to explore heterogeneity across households.

The results will provide insights into the determinants of demand for private tutoring and inform understanding of inequalities in access to supplementary education.

Intervention

No treatment is provided. Pure DCE.

Experimental design

The study uses a discrete choice experiment (DCE) to measure preferences for private tutoring among parents or legal guardians of students in compulsory education. In each choice task, respondents are asked to choose between hypothetical tutoring alternatives defined by a fixed number of attributes, or to select a no-tutoring option.

The experiment includes 5 attributes describing tutoring services: price, tutor experience, format of tutoring, whether the tutor is the student’s subject teacher, and proximity to home (travel time).

The experiment consists of an online survey incorporating a discrete choice experiment (DCE) designed to elicit parental preferences for private tutoring services for students in compulsory education. Respondents (parents or legal guardians) are presented with a series of hypothetical choice tasks in which they select their preferred option among alternative tutoring profiles.

Each tutoring profile is defined by a set of attributes. The experimental design comprises 16 choice sets, from which each respondent is randomly assigned 8 choice tasks to complete.

The survey is administered online to a representative sample of the population, stratified by NUTS II regions and maternal education. In addition to the choice tasks, the questionnaire collects information on household socioeconomic characteristics, current and past use of tutoring, student academic outcomes, parental aspirations and expectations, and reasons for (non-)participation in private tutoring, among other socioeconomic and educational variables.

The intervention does not manipulate real market conditions or provide actual tutoring services; it is limited to stated preference elicitation through hypothetical scenarios.

Design details

Randomization method
Randomization via the Qualtrics platform.
Randomization unit
Individual. Each respondent is randomly allocated to a subset of 8 of the 16 choice sets. When the respondent declares more than one compulsory school student in the household, the survey selects one student at random from the subset of students that attends upper secondary schooling (10th to 12th grade) and asks the subject to decide on the hypothetical private tutoring scenarios referring to the sampled student. If no student in the household attend upper secondary schooling, the survey is terminated.
Clusters
No clustering
Observations
1000 respondents
Arms
16 000
Primary outcomes
Willingness to pay for different tutoring characteristics.
Secondary outcomes
Private tutoring market size

Cited by

No citation data available for this registration yet.

Keywords