How is respect for workers’ rights actually being measured in workplaces around the world? What exactly are the facts to collect; the questions to ask; the indicators to look to?
Different participants in this field are using very different sets of yardsticks, which vary widely in the type, scope, level of detail, and precision of the measurements that they can produce. Not surprisingly, the different yardsticks also vary widely in their ability to pinpoint issues and problems.
This database offers as complete a picture as possible of current practice. It brings together a full range of the yardsticks now being used to assess the treatment of workers against international labor rights standards, in each major subject area (involuntary labor; wages; freedom of association; etc.). These measurement tools, broken down for analysis into their individual components, show up as questions that factory monitors are instructed to ask; in statistics and field observations used in published studies; as the bottom line of corporate compliance reports; and elsewhere. See Data and sources.
A database makes it possible to see at a glance the full current range of measurements now being used to assess workers rights conditions, for any particular topic in workers rights and for major issues within each topic. Key characteristics of each measurement (qualitative or quantitative; type of data produced; etc.) are identified and can be analyzed . See Methodology and caveats . Strengths and weaknesses of any group of measurements can then be evaluated and compared.
The purpose of presenting this database is to make it easier to identify current best practices in measuring labor standards compliance, pinpoint where the greatest needs for improvement are, and stimulate thinking about better measurements using current experience as a foundation.
Reliable measurements are particularly important today, as consumer demand continues to rise for information about the working conditions behind popular products and services. Major brand-name companies are trying to reassure the public about workers rights in their global operations. Labor rights advocates are trying to hold those companies accountable. Customers, investors and others want information they can rely on. All sides are therefore having to grapple with the practical difficulties of measuring actual working conditions in faraway workplaces.
This database contains approximately 2,500 units of measurement, extracted from approximately 100 public and confidential source documents including guidelines and questionnaires used by monitoring organizations, multinational corporations, and trade associations; public reports by governmental bodies and non-government organizations; public corporate reports; and others (see list of sources). Measurements are identified in as disaggregated a form as possible; i.e., individual units of measurement rather than summaries or syntheses. Each unit of measurement is classified by source, topic, and other key features, to facilitate analysis (see Methodology and caveats ).
The existence of a measurement unit does not mean that information is necessarily available about what value would be recorded for that measurement if it were actually applied in any particular place. As far as public information is concerned, the actual results from applying any particular measurement are few and far between. Where actual data are available from the same source document that produced the measurement unit, the database includes those data and classifies them for analysis.
The set of source documents used for this database does not exhaust every existing document that could be considered relevant to the measurement of labor standards compliance. For example, researchers evaluated approximately 200 corporate annual reports and corporate social reports that include statements about a corporation’s compliance with labor standards, compiled by staff of the International Labor Organisation (ILO), but this database includes only those that provided at least a modicum of specific measurement (approximately 64 of the 200 reports). However, based on our research, we are reasonably confident that this database reflects the full range of measurement units, of every type and level of detail, that are currently in use in this field, gleaned from a large and widely representative set of sources.
The compilation of sources, the extraction of individual units of measurements from each source, and the classification of each unit of measurement for entry into the database were conducted by researchers for the Lawyers Committee for Human Rights. The classifications applied to each individual unit were intended to show key features that would be useful in analyzing strengths and weaknesses of particular measurements, or groups of measurements. as reliable indicators of labor rights conditions at the factory level.
Some of the classifications are straightforward to apply, but others are not. As discussed below, researchers were required to exercise judgment in classifying many of the measurement units, in part because of ambiguity in the sources themselves and in part because some classification distinctions were inherently difficult to make at the borderline. The research team worked together to apply consistent judgments at the margin, but at least some classification decisions for some individual units of measurement will be debatable. The relatively large total number of measurements, overall and within most individual categories and subcategories, should help to minimize analytic distortion from any individual classification decision with which the database user does not agree.
Each measurement unit in the database is classified by:
The database can be searched by any of these characteristics, or by any combination of them. See How to use this database. Measurements that duplicate others in the database are also identified, so they can be included in or excluded from any specific search.
The major topics used in the database are generally recognized categories of workers rights. However, there is no generally accepted set of subtopics within each major topic. This project devised its own, based on the primary areas of concern addressed by the measurement units within the major topic. The emphasis was on subtopics that would be pragmatically useful for measurement concerns within the particular topic. For example, the Child Labor topic has a subtopic on verification of age records for young employees; the Freedom of Association topic has a subtopic on the ability of workers to communicate among themselves; etc.
Sources were highly inconsistent in the precision with which measurements were stated, the level of detail sought, the separation of objective and subjective inquiries, etc. Even sources specifically designed for factory assessment purposes, such as audit instruments and monitoring questionnaires, varied widely, with some asking for conclusory judgments (e.g., "Is physical abuse used on employees?") , others focusing on more directly observable facts (e.g., "Do workers show bruises or other signs that might indicate physical abuse?"), and others on interview questions (“Are you aware of incidents of physical abuse?” ).
Determining whether a measurement is quantitative or qualitative took the greatest degree of subjective judgment, since there is no simple formula that can be automatically applied to measurements occurring in widely varying types and forms. See below.
Quantitative and qualitative measurements
Distinguishing between units of measurement that are quantitative and those that are qualitative is important for analysis, but the distinction is not one that can be determined by a rigid formula. The following guidelines were used.
In general, quantitative measurements are intended to produce objective information in quantified form, such as a number (e.g., “what is the air temperature in the factory?”). Qualitative measurements are intended to produce subjective information reflecting a person’s or group’s perceptions (e.g., “is it too hot to work in the factory?”).
Quantitative measurements should be objective and quantifiable enough so that different observers would be likely to record the same or very similar results when facing the same conditions. This is the standard set out by a coalition of toy manufacturers in its efforts to monitor labor conditions in toy factories in China (see “Monitoring Mattel” at 9 & fn. 32). Measurements not meeting this standard are classified as qualititative.
Measurements in binary form (i.e., with yes-or-no answers) can be quantitative (e.g., “Are exit doors in the factory unlocked?”), and many quantitative measures in this database are in binary form. However, binary measurements present many borderline questions. Examples: “Are work hours too long?” seeks a qualitative answer, in yes/no form. “Do workers work more than 60 hours a week?” seeks a quantitative answer in yes/no form. “Is overtime compulsory?” is a yes/no question that could be considered either quantitative (an objective statement of factory practice) or qualitative (a subjective perception by workers that overtime requests cannot be turned down, even if managers claim otherwise). Context was used to help classify borderline cases.
Many of the sources used for this database cover the same ground. When measurement units are exactly or nearly identical, one is chosen as the most representative, and the remainder are marked as “duplicate.” A search that requests no duplicates (by choosing “Yes” opposite “Leave out near-duplicates?”) will leave out the records marked as duplicates, so that only the single most-representative record appears. All records including all duplicates can be seen by performing the same search without choosing "Yes".
Records marked as duplicates are not necessarily identical to the most-representative record; variants can be classified as duplicates. Borderline cases are not marked as duplicates. NOTE: if a search excludes duplicates, the search results will not necessarily show the full range of sources which include the measurement in question. Search without excluding duplicates in order to see how often a particular measurement occurs, in what forms, and in which sources.
Major topics of workers rights
The major subject-matter areas of workers rights covered in this database are:
You can search this database for the measurement units that fit any of the categories or criteria described above (see Methodology and caveats), or any combination of them. Use the CATEGORY SEARCH section of the Query Page. For example, you can see all of the measurements that address Child Labor; or a subtopic of Child Labor; or only the quantitative measurements for Child Labor; or only those from a specific source; etc. We recommend you begin your search with only one or two choices of category, such as a topic or a topic and subtopic, and then add additional choices to narrow your search further. Any combination will work.
The ADDITIONAL RESTRICTIONS section of the Query Page lets you refine your search more narrowly; for example, to only quantitative measurements that are expressed in percentage terms, or only qualitative measurements that reflect local NGOs’ views, or only measurements for which companies have reported results to the public.
Using the WORD SEARCH section, you can also search by word(s); for example, every measurement unit that includes the word “grievance,” or the words “legal minimum wage.” A word search can be limited to a single topic, or subtopic within a topic; for example, only the measurement units that include the word “grievance” and are classified under Non-discrimination/gender. Word search cannot be combined with any criteria other than Topic and Subtopic.
Results of any search appear in the grid at the bottom of the Query Page, and can be printed. The number of measurement units that result from your query is shown at the top of the grid; for example, “Your query returned 55 records on 3 page(s). This is page 1/3.”
NOTE: if your query returns more than 20 results, only 20 will be shown on the first page. Navigation buttons at the top and bottom of the grid allow you to move from page to page in multi-page results.
We welcome any and all feedback on any aspect of this database, including any part of its content. Please email the database administrator with your comments.