But can this relationship be measured?
Such relationships can be studied at various levels of size or aggregation – typically at
- the community level (comparing municipalities),
- the regional level (comparing counties or districts) – or
- the national level (comparing nations)
We would, in such a case, try to relate the level of library services or output (A) to the level of literacy (B)
- in a set of communities
- in a set of regions
- in a set of nations
If communities (or regions, or nations) with a high level of library services also have a high level of literacy, we can hope that libraries cause (or influence) literacy. But a positive correlation is not a proof of causality. The relationship might go the other way.
It might be the case that a high level of literacy leads to a high demand for library services – and that governments react positively to that demand. Then we would say that “literacy cause libraries” rather than the opposite. To distinguish between these two situations, it would be useful to look at changes over time. The two main possibilities are
- an increase in library services is followed by an increase in literacy (A causes B)
- an increase in literacy is followed by an increase in library services (B causes A)
I simplify, since more complex mechanisms might be at work. But we already have enough background to turn to more practical matters.
To go further we must find relevant indicators for A and B – and look at data over time (time series). Indicators measuring A belong to the field of library statistics. Indicators measuring B belong to the wider field of educational or social statistics.
Literacy Assessment and Monitoring Programme
In 2003, the UNESCO Institute for Statistics (UIS) initiated the development of its Literacy Assessment and Monitoring Programme (LAMP) as a major effort to improve the body of available statistics on literacy. After five years of developmental work shared with experts, specialised institutions and a group of countries representing different regions of the world, the UIS is now ready to share the first results of this endeavour: a validated approach to measuring literacy skills in developing countries.
The UIS report is called:
I quote from the Executive Summary
In 2003, the UNESCO Institute for Statistics (UIS) started the Literacy Assessment and Monitoring Programme (LAMP) in partnership with several countries and organisations in order to develop a new methodology for measuring literacy and numeracy skills among youth and adults to improve the available body of statistical evidence. Its design was inspired by surveys conducted mostly in countries of the Organisation for Economic Co-operation and Development (OECD): the International Adult Literacy Survey (IALS) and the Adult Literacy and Life Skills Survey (ALL), which are thus far the most significant efforts in cross-national measurement of literacy and numeracy. Addressing their strengths and weaknesses helped shape LAMP in its early years.
LAMP aims to provide policymakers with robust information on population profiles in terms of literacy and numeracy, thus helping inform public debates while influencing the design of literacy and adult education programmes to expand the opportunities of individuals, families, communities and countries. …
LAMP intends to equip countries with the methodological tools needed for direct assessment of literacy and numeracy skills and to strengthen national capacities. By combining these two elements, it aims to ensure that countries move towards a sustained production of robust literacy data. This would become part of the key set of international statistics that the UIS would monitor and continuously improve.
The project takes, in my judgment, a practical and realistic approach to the production of literacy statistics. It is very aware of the complexities involved in measuring literacy skills. LAMP is based on household surveys rather than on census data. It has developed a set of tested instruments for data collection – and provides a clear guide for countries that want to implement this approach (p. 8-9).