Description of the programme

Sixteen hours of extra-curricular financial literacy education was delivered to secondary school students (in their final year before graduation) in classes over a three-month period. The programme covered: basic concepts of economics; economic operators (households, companies, banks); debt, indebtedness and financing; monetary policy and monetary institutions, financial markets; and, finances and ethics.

The educational programme comprised presentation slides delivered by classroom teachers (59 teachers in 59 classrooms). Teachers received detailed (written) guidance along with the presentational material. Students were drawn from three urban environments to reflect the cultural heterogeneity of Italy (Rome, Genoa and Milan) and came from classes studying classical studies, scientific studies and vocational training.

The intended outcome was to improve financial literacy as measured by a questionnaire containing 27 multiple choice questions.

The study

The evaluation used a Randomised Control Trial (RCT) approach and comprised a pre-measurement prior to the financial education course and a post-measurement four months after the end of the course, with class participants. It used a comparison group of secondary school pupils who had not participated in the course. The same questionnaire was administered at each stage, and also to the control group using the same time interval.

The study included a strong statistical evaluation design through difference-in-differences, the use of matched pairs and covariates as control variables. Clustering by school class was also taken into account in the statistical analysis.

Key findings

The evaluation found that positive outcomes were achieved in relation to:

Financial capability (ability):

• There was a significant improvement in students’ financial literacy (1.7 additional correct answers in a pre-and post-education questionnaire comprising 27 questions) within one month of completing the education programme, which could be attributed to the education programme.
• The results of the experiment differed by location - there was no significant, attributable improvement in Rome, while in Milan and Genoa it was very large, where the improvement measured five additional correctly answered questions.
• The impact was greater amongst higher grade students, independently of a range of other factors.

Points to consider

The authors note that the impact of the intervention may have been greater had the course been incorporated into regular teaching structures and subject to examination.

In addition, there were some limitations to the study:

• Assignment to treatment and control groups was not truly random, and it is not clear how segregation of treatment and control groups was assured.
• There was significant item non-response and/or attrition which reduced the initial sample size from 3,820 to 2,124. While a large achieved sample remained, the differential influence of non-random factors on failure to participate fully in the programme or the measurements was not fully accounted for.
• There was also some evidence that treatment and control groups in some instances differ significantly on the outcome measures pre-intervention and this is not fully explained.

Full report

Full research proposal