A variety of measures are used to capture different aspects of inequality. Measures of well-being can be classed as objective or subjective and the Framework includes both objective and subjective inequality measures.

Objective measures capture measurable activities or events, such as rates of recorded violent crime or employment rates by disability status. Some objective measures capture the different outcomes between groups for a particular indicator. Others can be broken down by income levels or family background to measure socio-economic gradients, such as highest educational attainment or child mortality rates. Some measures, such as the ones for income inequality or inequality in life expectancy, directly capture dispersion across society, while others focus on concentration at the top end (the top 1% income share).

Subjective measures are those that are collected from individual people, which generally involves asking them questions and assessing their opinions. These measures can capture differences in life satisfaction or individuals assessments of their own general health. Other subjective measures include percentage of people who report feeling lonely and the share of people who express that they feel unsafe alone in local area after dark. These measures can also be used to analyse differences between groups of people as well as to assess overall dispersion.

There are a number of ways in which we can measure inequalities between individuals or groups of individuals. Measures of dispersion which assess differences between people across the whole population, for example income inequality measured by the Gini coefficient, are sometime referred to as vertical inequality measures. These measures capture differences across the whole distribution, from the most advantaged to the least advantaged. Disaggregating an outcome measure (for example, health status) on the basis of income level can also give you a snapshot of vertical inequality.

Measuring inequalities between different groups, defined in terms of personal or household characteristics such as gender, disability status, work status, ethnicity, or region of residence, provide a snapshot of horizontal inequalities. Horizontal inequalities can be observed in many forms – from indigenous women facing a higher risk of dying in childbirth, to health status varying across different age groups or lower employment rates among people with disabilities. Measuring horizontal inequalities will be key to your analysis but you are less likely to find published data in this form for most of the measures included in the framework and so this analysis is likely to require access to micro-data.

Inequalities experienced by people living in different geographical locations are forms of horizontal inequalities, but are also sometimes described as spatial inequalities. A common spatial dimension of inequality in developing countries is the inequality between urban and rural areas. Measuring differences in educational, health and nutritional outcomes, and other dimensions of wellbeing such as access to potable water, sanitation and electricity, between people living in urban and rural areas, or different regions, can be an important part of an inequality analysis. These geographical inequalities are often given a strong focus as they can be important for directing public investment patterns.

You may also be interested in looking at the interaction of more than one or two personal characteristics which you can measure through estimating intersecting inequalities. For example, you might be interested in contrasting the general health of women living in a rural area with those living in an urban area (the intersection between gender and urban/rural living), or you may be interested in measuring the loneliness of older men with low levels of household income compared with those living in a higher income household (the intersection between gender, age and household income). While these contrasts will provide you with more detailed information on inequality, it is important to be aware that data requirements will be greater and you could end up with a large number of outcome measures to analyse, so being selective will be key to good analysis of intersecting inequalities.

To measure the inequalities covered by the Framework, you will need to access quantitative data, usually based on statistics from surveys or administrative sources which may be shown numerically in tables, graphs and maps. Quantitative data is often presented in the form of summary statistics such as, headcounts, percentages, ratios, and other measures that can be used to assess inequality. An example of quantitative data is The World Bank’s Gini index which is derived from primary household survey data obtained from government statistical agencies.

A good place to start looking for quantitative data is with your National Statistics centre or institute website. There you will be able to find information from national household surveys and administrative data.

There are many different household surveys with some focused on a single issue, such as employment, health, income or wealth, and some more general in nature covering many topics.

An important aspect to bear in mind is that sometimes official data sources are not the most reliable. When it comes to certain measures such as deaths in police detention or prisons, and crime statistics, there is a risk that official data can be biased. Where you suspect that this is the case, it is a good idea to try and verify official statistics from other sources such as bespoke surveys or documented allegations e.g. allegations of disappearances, or police violence, and other forms of quantitative evidence gathered by specialist NGOs, the media and other sources. You will also have to assess how reliable these sources are and make sure that you highlight any concerns with the quality of the data when you present you findings.

You may also find little official data for some areas of well-being covered by the Framework. For example, information on bullying and violence in schools, or exploitation in workplaces may not be collected and reported in official statistics. Again research or surveys by specialist NGOs, or for academic research or research institutes might shed light on incidences of mistreatment and different people’s experiences.

As well as looking at quantitative measures for income, wealth, educational attainment and so on, in some cases qualitative data can help to measure inequality.

Qualitative research is the term used for collecting and analysing the views and opinions of people. This can be done through collecting individual human stories which can illustrate inequalities how people live. It can be done in relation to many of the objective or subjective information collected by larger. We can ask people how poor health affects the quality of their lives or their experience of discrimination to add personal testimony and richer detail.

Qualitative research includes observation, interviews, surveys, and the analysis of voice, speeches, and conversation (discourse analysis). Case studies can provide valuable qualitative information. For example, you can use a case study to explore aspects of social mobility: The contrasting life stories of individuals from different backgrounds can provide powerful narratives about the impact of inequalities in family background and how opportunity is related to a person’s starting point in life. If quantitative information on social mobility is not available, this type of qualitative data can help to fill this gap.

Another relatively new method is the use of Inequality Diaries where participants record their daily lived experiences. Participants may find writing or recording their own accounts easier than being interviewed. The process can allow participants - particularly those who are marginalised – a ‘voice’ and agency in their situation. Their insights can be extremely powerful for advocacy and campaigning.

For instance, if you were looking at education and aggregate school graduation rates you could use disaggregation to see how many girls graduated compared to boys, or contrast the graduation rates between people from different religious or ethnic backgrounds, or contrast the results for people from rural areas as opposed to urban areas, or uncover any disparities in the graduation rates for people with disabilities, or people from different family backgrounds.

Here are some disaggregation variables that you could use:

1. Gender
2. Age groups
3. Education level
4. Income groups
5. Caste/Social class
6. Race/Ethnicity
7. Urban-rural location
8. Geographical region
9. Disability status
10. Citizenship and immigration status
11. Religion