Higher-Order Reasoning and the Development of Social and Redistributive Preferences

The intervention consists of a structured set of curricular materials designed to strengthen key components of analytical thinking in young children. Its overarching aim is to accelerate learning and improve learning efficiency, thereby reinforcing foundational numeracy and literacy skills in the early primary years. In the first year, the program targets abstract reasoning, short-term memory, and analytical categorization. In the second year, it emphasizes inference and analogical reasoning. Trained teachers were instructed to allocate at least four hours per week to the materials during regular class hours. Program implementation concluded in April 2025.

2023-10-16

2025-04-30

social preferences (trust, reciprocity and altruism), redistributive preferences and moral reasoning.

We will collect the following data in May 2026 (the final wave of the original RCT) in the province of Gaziantep. The purpose of this data collection is to evaluate the impact of the program on the development of social and economic preferences.

We will assess program effects in the following domains:


Redistributive Preferences:

Measurement: We will introduce a novel ``tax game'' to measure children's willingness to redistribute their earned payoffs within the classroom. This task captures normative fairness ideals, self-interested fairness, and self-serving bias.
Expectation: We do not take a directional stance on the expected effect. The program may strengthen fairness ideals via improving pro-social attitudes or, alternatively, promote merit-based interpretations of inequality depending on how analytical reasoning, critical thinking and growth mindset shape fairness perceptions.


Moral Reasoning:

Measurement: We will implement a ``morality game'' to assess moral judgment and reasoning, distinguishing between outcome-based and action-based morality.
Expectation: We expect that the program may foster greater sensitivity to intentions (reflecting more sophisticated moral reasoning).

Note: We aim to collect moral reasoning outcomes in Şırnak province as part of the higher-order reasoning module. These outcomes will likely be included in the original RCT paper.

Social Preferences:

Measurement:

Altruism: Measured using a standard dictator game.
Trust and Reciprocity: Measured using a structured trust game in which children make both sender and receiver decisions under the strategy method.

Expectation: We expect that the program may promote prosocial behavior through changes in empathy and the formation of productive social capital.

We will use all measured socioemotional skills and indicators of social capital from the original RCT to explore potential mechanisms underlying the program’s effects. Specifically, we will examine growth mindset, locus of control, cognitive empathy, norm-based reasoning, and social capital (friendship networks). These outcomes were collected in one or more earlier waves and will be collected again in the final wave.

We will use all measured socioemotional skills and indicators of social capital from the original RCT to explore potential mechanisms underlying the program’s effects. Specifically, we will examine growth mindset, locus of control, cognitive empathy, norm-based reasoning, and social capital (friendship networks). These outcomes were collected in one or more earlier waves and will be collected again in the final wave.

The ongoing RCT includes 132 state primary schools, with 48 located in Şırnak province and 84 in Gaziantep. Schools were assigned to treatment with a 50% probability, stratified by district. This resulted in 67 treatment schools and 65 control schools.

For the present study, we will collect social and redistributive outcomes only in Gaziantep in May 2026, and the analysis will therefore focus on the Gaziantep sample. If logistics allow, we will also collect the moral reasoning measures in Şırnak. Due to the closure of one control school, the Gaziantep sample now consists of 83 schools: 43 treatment and 40 control. If current funding constraints are relaxed by the time of the fieldwork (April 2026), we will collect all of the above outcomes in Şırnak as well.

Not available

Done in office by a computer, using Stata command "randtreat"

school

Yes

132 for the whole study (as registered before), 84 in this specific study (only Gaziantep province), currently 83 schools in Gaziantep.

over 10,000 children

40 control schools and 43 treatment schools (previously 41 control schools).