Information, Beliefs, and Public Preferences for Higher Education Financing and Reform in Spain: A Survey Experiment

Last registered on June 03, 2026
View Trial History

Pre-Trial

Trial Information

General Information

Title
Information, Beliefs, and Public Preferences for Higher Education Financing and Reform in Spain: A Survey Experiment
RCT ID
AEARCTR-0018656
Initial registration date
May 25, 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
June 03, 2026, 8:27 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
Ismael Sanz
Affiliation
Universidad Rey Juan Carlos
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Marta Martínez-Matute
PI Affiliation
Universidad Autónoma de Madrid
Contact Investigator
PI Name
Maria Teresa Ballestar
PI Affiliation
Universidad Rey Juan Carlos
Contact Investigator
PI Name
Jorge Sainz
PI Affiliation
Universidad Rey Juan Carlos
Contact Investigator
PI Name
Antonio Cabrales Goitia
PI Affiliation
Universidad Carlos III de Madrid
Contact Investigator

Additional Trial Information

Status
In development
Start date
2026-06-01
End date
2026-12-31
Keywords
Behavior, Education, Governance, Welfare
Additional Keywords
Higher education, university financing, tuition fees, information provision experiment, beliefs, public preferences, income-contingent loans, accountability, Spain
JEL code(s)
I22, I28, H52, D72, D83
Secondary IDs
PI-ONT-UC3M
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
This study examines how information shapes Spanish citizens' preferences over the financing and reform of higher education. We run a pre-registered survey experiment with at least 5,000 nationally representative Spanish adults.
Respondents are randomly assigned to one of five conditions: a control group that receives a neutral placebo text about general facts of the Spanish university system, and four treatment arms that each receive a distinct block of factual information before answering policy-preference questions. The four treatments are: (1) information on the labour market earnings premium of a university degree relative to higher vocational training; (2) information on the level of public expenditure on universities in Spain and the fiscal trade-offs of increasing it; (3) information on quality heterogeneity across Spanish public universities; and (4) information on alternative financing systems, including income-contingent loans and means-tested fees in use in the United Kingdom, Australia, Chile and New Zealand.
The two pre-registered primary outcomes are support for higher tuition fees in public universities and support for higher public spending on universities even at the cost of taxes or budget cuts elsewhere, both elicited before and after the information treatment. A separate primary outcome for the quality-heterogeneity treatment captures support for budgetary accountability of public universities and the priority destination of additional public spending. Secondary outcomes include the accuracy of prior beliefs about earnings, public costs and university quality, the perceived fairness of alternative financing systems, and willingness to pay.
The analysis applies the analysis-of-covariance specification described in the pre-analysis plan, conditioning on pre-treatment outcomes and baseline covariates. We pre-specify Bonferroni correction across the four treatment-versus-control comparisons on the primary outcomes, with Romano-Wolf step-down adjusted p-values reported as a complement. Pre-specified heterogeneity analyses examine differences in treatment effects by prior belief accuracy, educational attainment, household income, gender and relation to the university system.
This study replicates and extends Lergetporer and Woessmann (2023, Journal of Public Economics) to the Spanish context, and complements it with three additional informational treatments not previously tested.
External Link(s)

Registration Citation

Citation
Ballestar, Maria Teresa et al. 2026. "Information, Beliefs, and Public Preferences for Higher Education Financing and Reform in Spain: A Survey Experiment ." AEA RCT Registry. June 03. https://doi.org/10.1257/rct.18656-1.0
Sponsors & Partners

Sponsors

Experimental Details

Interventions

Intervention(s)
The intervention is an informational message delivered within an online survey. Participants are randomly assigned to one of five conditions. The control group receives a neutral placebo text describing general facts about the Spanish university system (number of universities, degree structure under the Bologna Process, enrolment figures, Erasmus+ participation), without information relevant to financing preferences or labour-market returns. The four treatment arms each receive a different factual information block of approximately 100 to 150 words, accompanied by source citations: (T1) the labour-market earnings premium of a university degree relative to higher vocational training (FP de Grado Superior); (T2) the level of Spanish public expenditure on universities relative to the OECD average, together with the explicit fiscal trade-offs that an increase in spending would entail; (T3) quality heterogeneity across Spanish public universities in international rankings, graduate employment rates, salaries, and research output; (T4) three alternative financing systems used internationally (means-tested fees, merit-based scholarships, and income-contingent loans), with country examples.

Intervention Start Date
2026-06-01
Intervention End Date
2026-06-30

Primary Outcomes

Primary Outcomes (end points)
Two pre-registered primary outcomes are measured post-treatment on a five-point Likert scale: (Y1) support for raising tuition fees in Spanish public universities; (Y2) support for higher public spending on universities even at the cost of higher taxes or cuts in other public spending categories. For the quality-heterogeneity treatment (T3), two additional pre-registered primary outcomes are: support for budgetary accountability of public universities (binarised from a four-option item), and the priority destination of additional public spending on universities (categorical, with pre-specified binary contrasts).
Primary Outcomes (explanation)
The two main primary outcomes (Y1, Y2) are analysed in two complementary ways: as a binary indicator equal to 1 if the respondent answers "very much in favour" or "in favour" and 0 otherwise; and as a five-point ordered outcome. The accountability outcome is binarised as 1 if the respondent supports universities operating with autonomy combined with budgetary consequences for poor performance, and 0 otherwise. The priority outcome is analysed using a multinomial framework with pre-specified binary contrasts between quality-oriented options, fee-reduction-oriented options, and need-based scholarship options.

Secondary Outcomes

Secondary Outcomes (end points)
Pre-registered secondary outcomes include: (S1) accuracy of prior beliefs about the earnings premium of a university degree, and post-treatment numerical earnings beliefs (T1 and Control only); (S2) accuracy of prior beliefs about the share and per-student level of public financing of universities, used as a covariate and moderator; (S3) perceived fairness of five alternative financing systems, and identification of the system perceived as most socially just (mechanism for T4); (S4) for treatment arms other than T3, the accountability and priority-of-spending items also reported as secondary outcomes; (S5) willingness to pay for a year of public university education, as an ordered categorical measure complementing Y1. Exploratory outcomes include opinions on teaching and research quality, perceived university contribution to societal goals on a 0 to 10 scale, comparison with the EU average, and self-reported change in earnings perception.
Secondary Outcomes (explanation)
S1 is operationalised as a binary indicator equal to 1 if the respondent selects the correct category in the pre-treatment item on the earnings gap between university and FP Superior graduates. The post-treatment numerical estimate of monthly earnings for university graduates is asked only of respondents in T1 and Control, enabling estimation of belief updating in T1 relative to the control distribution. S3 is captured by a 5x5 Likert matrix and an additional single-choice item. Willingness to pay is captured by a six-category ordered item.

Experimental Design

Experimental Design
Online survey experiment fielded to a nationally representative sample of Spanish adults (18 years or older). The survey consists of four sequential blocks administered in a single session. Block 1 collects sociodemographic information, the respondent's relation to the university system, self-rated knowledge, and frequency of following news on higher education. Block 2 elicits prior beliefs about labour-market earnings premiums, public financing levels, and quality heterogeneity, as well as pre-treatment versions of the policy preference outcomes. Block 3 contains the randomised information treatment (one of five conditions) and an arm-specific manipulation check for the four treatment arms. Block 4 contains the post-treatment outcomes, the secondary outcomes, an attention check, self-reported survey duration, and an item on perceived demand effects. There is no follow-up wave. The fieldwork uses a representative online panel of Spanish adults provided by a commercial survey supplier. Randomisation is implemented at the individual level by the survey platform prior to Block 2. The analysis follows an analysis-of-covariance specification with the pre-treatment value of the outcome and pre-specified baseline covariates as right-hand side variables, with heteroskedasticity-robust standard errors. Multiple hypothesis testing is addressed via Bonferroni correction for four pre-specified treatment-versus-control comparisons on the primary outcomes, complemented by Romano-Wolf step-down adjusted p-values.
Experimental Design Details
Not available
Randomization Method
Randomisation performed by the QuestionPro survey platform at the individual respondent level, immediately before the pre-treatment belief block, using QuestionPro's built-in random assignment functionality with a documented seed.
Randomization Unit
Individual respondent. No clustering.
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
Not applicable (individual-level randomisation, no clustering). For form-filling purposes: 5,000 individuals.
Sample size: planned number of observations
5,000 individual respondents (Spanish adults, 18 years or older).
Sample size (or number of clusters) by treatment arms
Control: 1,000 (20.0%); T1 (labour-market returns): 1,250 (25.0%); T2 (financing and fiscal trade-offs): 915 (18.3%); T3 (university quality heterogeneity): 915 (18.3%); T4 (alternative financing systems): 915 (18.3%). Total: 5,000.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Under the pre-registered analysis-of-covariance specification, conditioning on the pre-treatment outcome, with a within-individual pre-post correlation assumption of 0.6, two-sided tests at 80% power, and Bonferroni correction for four pre-specified treatment-versus-control comparisons (adjusted alpha = 0.0125): minimum detectable effect is approximately 5.6 percentage points for T1 versus Control on Y1, 6.1 percentage points for T2 versus Control on Y2, and 6.0 percentage points for T4 versus Control on Y1. For T3 versus Control on the binarised accountability outcome, the minimum detectable effect is approximately 6.1 percentage points; this is computed without conditioning on a pre-treatment value since the T3 primary outcome has no pre-treatment counterpart in the instrument (see PAP §6.5). Effects are expressed as differences in the proportion of respondents who report being in favour of the policy (binary outcome derived from the five-point Likert primary outcome). Sample standard deviation is approximately 0.49 for binary outcomes with proportions near 0.5.
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
Committee on Ethics in Research, Universidad Carlos III de Madrid
IRB Approval Date
2026-05-19
IRB Approval Number
CEI-26-10-E-ONT-UC3M
Analysis Plan

Analysis Plan Documents

PAP_University_Financing_RCT_25_de_mayo_v2.docx

MD5: d2ab24c5f03bbb0d6b1f9c299aa50aa5

SHA1: acdbb0fc01c992506fbab7c9cda7081bbfea0b4c

Uploaded At: May 25, 2026