About WebMILS

WebMILS is a way of organizing and presenting all available information that is relevant to the core labor standards. It enables assessors -- all interested parties throughout the world -- to have access to what information exists and to use that information as best suits their needs.

The database structure reflects four sets of indicators that CMILS developed for the core labor standards and acceptable conditions of work. These indicators are organized by legal framework, government performance, and overall outcomes. The committee also included associated factors that might either provide contextual information or serve as useful signals that there may be problems with compliance. The database also contains measures of human capital investment recognizing that the linkages between human capital and labor standards compliance are critical. For each indicator, there are from 4 to 21 items.

Assessing compliance along the dimensions proposed by the committee requires consideration of both qualitative reports and quantitative data. But the quality of both types of information is uneven and research methods for measuring compliance with international labor standards and acceptable conditions of work are still in the developmental stages.

Given the emphasis on legal frameworks and government performance, qualitative sources are essential. Such sources provide on-the-ground reporting on working conditions and workers' rights. Ideally, such sources would be developed independently of the stakeholders directly involved -- governments, employers, and workers -- and would be prepared by experts in the field. But, as examined below, qualitative sources of information reflect the perspective of the reporters and may be biased in one direction or another. Some of these reports are complaint-driven, which can introduce selection bias, while others may be written by a committee, or skewed or muted for political reasons. They often lack consistency over time because many of these reports are ad hoc rather than regular and systematic. In developing countries with large informal sectors, both qualitative and quantitative sources are likely to be unrepresentative of conditions overall within an economy.

The order in which sources appear is roughly the order in which assessors retrieve evidence from the database to perform assessments, and the order in which subsequent efforts to fill out the database be conducted, based on criteria of independence and expertise, and ready availability of information. In populating the database, the committee focused first on sources that were international, covered a majority of countries in the world, were secondary, were easily accessible electronically, and which would be relatively easy to sustain. Such sources were both quantitative and qualitative in nature. Although no source is ideal, the International Labor Organization is the obvious starting point for information on member countries' laws and practices with respect to labor standards. For countries or standards not covered by ILO sources, other official sources, such as the United Nations, require similar country reporting under other relevant international conventions, such as the Convention on the Elimination of Discrimination Against Women, that are also sometimes accompanied by third-party review. The World Bank is also beginning to pay attention to labor standards issues in its country reports, but not as systematically as the other international sources.

After international sources, other official sources at the national level can provide supplemental information. Similarly, nongovernmental organization reports are often quite detailed and can quite be useful, particularly those that cover large numbers of countries. But they will also often be subject to selection bias and, whatever the quality of their information, will be viewed with suspicion by the parties they criticize. When attempting to assess compliance with regard to any given country, NGO and other unofficial sources are most valuable, therefore, to provide complementary or contrasting perspectives for matching against other sources.

Labor market, economic and other quantitative data are also essential for assessing compliance, particularly for the indicators in the overall outcomes. Ideally, quantitative data should meet a number of criteria:

• The data were collected in an actual census or survey and are not estimates that extrapolate from data collected in earlier surveys.

• Survey recipients in the survey sample are drawn from an actual census, which permits issues of representation, selection bias, and nonresponse to be assessed and reported, which is important in judging the validity and reliability of data that emerge from the survey.

• The data are national in coverage. In some countries, labor data may be collected only for urban areas or only for limited industries. As a consequence, the data are inherently biased because not all parts of the labor force are represented.

• Survey questions are consistent over time, in order to allow assessments of trends. Any necessary changes in data definitions, survey questions, or collection methods should be fully documented.

In general, national-level data included in the database are applicable to that country using country-specific definitions. The data are often not comparable across countries, and database users are cautioned against using them in this manner. The data may or may not be comparable within countries over time. In addition, assessors using the database should be sensitive to issues related to questionnaire design, sample design and sampling error, data collection and nonsampling errors (e.g., nonresponse), and data preparation problems (e.g., the possibility of data entry errors).

The database is designed to include all available information, from all possible sources. The committee did not select data for inclusion or exclusion nor did it establish criteria for inclusion or exclusion. Rather, the database is designed to include all available data, along with the comments on anything that is known about the data or its source (such as frequency or range of coverage). Thus, the database structure provides a way for all assessors to have access to all available data and information about those data.Also, there are other sources of information that have yet to be tapped, such as information collected by national statistical agencies that is not compiled by international agencies, such as the ILO. In spite of inherent weaknesses in data generated from national statistical agencies, they may be the only source of national-level data on labor practices and policies, especially for data on government performance in endorsing and enforcing labor legislation.

Additionally, local nongovernmental organizations (NGOs) may produce information on violations of a specific labor standard within a single country that has not yet been identified or entered into the database. The committee's database uses information from a number of nongovernmental organizations (NGO) on various aspects of core labor standards. NGO reports vary considerably in the validity and reliability of data reported. In addition, many NGO reports are restricted in terms of time frame, geography, and industrial focus -- with some reports focusing on one manufacturing facility in one city for one time period. Many NGOs also have strong political orientations that may introduce a selection bias in the data that are published. However, NGO reports may provide early warning of an emerging compliance problem or may provide more detailed information than other sources. As a result of the potential for selection or other bias, NGO reports should be used primarily to supplement official sources and should be cross-checked with other sources.