Supriya Ghosh (Editor)

Global Slavery Index

Updated on
Edit
Like
Comment
Share on FacebookTweet on TwitterShare on LinkedInShare on Reddit
Global Slavery Index

The Global Slavery Index is an annual study of world-wide slavery conditions by country published by the Walk Free Foundation. In 2016, the study estimated a total of 45.8 million people to be in some form of modern slavery in 167 countries.

Contents

The report includes three data points for each country: national estimates of the prevalence of modern slavery, vulnerability measures and an assessment of the strength of government responses. The Index pioneered the use of random-sampled, nationally-representative surveys to estimate prevalence. This included commissioning seven such surveys in 2014 and a further 19 surveys through Gallup World Poll in 2015.

The 2014 Index includes country studies with policy recommendations for many countries including: Taiwan, Australia, Pakistan, India, Brazil, the United Kingdom, the United States, Qatar, and others.

Calculation

The 2014 Global Slavery Index includes data on three key variables: the prevalence of modern slavery in each country, vulnerability, and government responses to modern slavery. The methodology is written up in detail in a methodology paper.

The first of these factors, the prevalence estimates, were derived using a statistical process known as extrapolation. In 2015, Joudo Larsen, Bales and Datta published an article that describes the extrapolation process, and provides a test of it. Writing in Significance, the magazine of the UK Royal Statistics Society, Joudo Larson et al. note that the estimates in the 2014 Global Slavery Index involved several steps:

  • Clustering: countries were grouped into seven clusters, based on their degree of similarity for factors relevant to vulnerability of enslavement. The data on vulnerability included variables on five dimensions: state stability, social discrimination, presence or absence of human rights protections, economic and social development, and presence of governmental slavery policy. The groups were chosen by a K-means clustering algorithm.
  • Extrapolation: survey data on the prevalence of slavery (expressed as a percentage of the population) that was available for each cluster was then used to estimate the prevalence for other countries within that same cluster. A slightly different method was followed for 'cluster 1', which involved mostly European and highly developed countries. This is discussed further below and in the article by Bales and Datta.
  • National-level adjustments were done by hand on a country-by-country basis, reflecting considerations such as geography.

    • This resulted in a prevalence estimate for each country, calculated as a proportion of the total population that was enslaved within that country. For all 167 countries this produced a total global estimate in 2014 of 35.8 million enslaved people.

    Writing in 2015, Larsen et al. note that newer 2015 survey estimates now allow a comparative test of some of the 2014 extrapolated estimates. This is important for checking the validity of the earlier results and the continued application and use of this methodology. The comparisons suggest that extrapolation, while not perfect, is a useful and valid method. All but one of the extrapolated estimates of the prevalence of slavery fell within one percentage point of the estimates derived from random sample surveys.

    In their article "Slavery in Europe: Part 1, Estimating the Dark Figure", Monti Datta and Kevin Bales demonstrate the statistical techniques applied to the estimation of the dark figure of the prevalence of modern slavery across Europe. These same techniques are applied to a dataset that includes random survey sampling (where possible), secondary source estimates, a process of extrapolation, and a country-level adjustment to determine the prevalence of people in modern slavery around the world in the Global Slavery Index.

    Controversy

    According to researchers Andrew Guth, Robyn Anderson, Kasey Kinnard and Hang Tran, they feel the Index's methods reveal weaknesses and they raise questions about its replicability and validity. They claim that the publicity given to the Index is leading to the use of its data in popular culture, reputable magazines, by news outlets, by academic journals and top policy makers.

    Some countries, for which no data were available, were given the same rate as countries that were judged to be similar. For example, prevalence rates for Britain were applied to Ireland and Iceland, and those for America to several western European nations, including Germany. This form of extrapolation has attracted some criticism.

    Global Slavery Index 2013

    The following is a ranking of the top 30 countries/territories in order of the lowest prevalence of modern slavery. The full rankings can be found in the Walk Free Foundation's website.

    1.  Iceland,  Ireland,  United Kingdom
    4.  New Zealand
    5.  Austria,  Belgium,  Denmark,  Finland,  Greece,  Luxembourg,  Norway,  Sweden,   Switzerland
    14.  Cuba
    15.  Portugal,  Spain
    17.  Costa Rica
    18.  Panama
    19.  Canada
    20.  Mauritius
    22.  Singapore
    23.  Hong Kong
    24.  France,  Netherlands
    25.  Australia
    26.  South Korea
    27.  Germany
    28.  Barbados
    29.  United States
    30.  Trinidad and Tobago
    31.  Italy
    32.  Latvia
    33.  Japan

    References

    Global Slavery Index Wikipedia
    (Text) CC BY-SA