It contains survey data collected in Kaski and Chitwan districts of Nepal between 2014 and 2016, related to education and skills, migration, caste relations, and cultural consumption. Survey data entails 837 variables and 1203 respondents. Field research were carried out for a year and half in these villages and in their migration satellites in the Tarai (the Gangetic strip of south Nepal abutting India), in urban centres, and international migrants also were included. Participant observation and interviews were combined with detailed surveys of both households and individuals in order to reveal changing attitudes to education, employment, and migration. The two next-biggest local ethnic groups, the Chhetris and Gurungs, who rank in between Bahuns and Dalits in the traditional caste hierarchy, were also included in the quantitative part of the study in order to bring out contrasts and comparisons. By producing an empirically sound, ethnographically sensitive, and quantitatively sophisticated study of the social history and migration of these two key Nepali groups, one of which is the most significant disadvantaged caste bloc, the research aimed having considerable potential policy impact in Nepal. The timing of research, coming as it did during the ongoing peace process and while disadvantage and exclusion are still very much part of the political debate, was appropriate and indeed advantageous.Nepal, like India, has traditionally been a caste society, with Bahuns (Brahmans) at the top, Chhetris (Kshatriyas) second, and Dalits (ex-Untouchables) at the bottom. Groups that used to be known as tribes and are now called Janajatis (the groups most commonly recruited to the Gurkha regiments) were slotted into the middle of the hierarchy. Between 1854 and 1951 this caste hierarchy was enforced in an authoritarian way by the state, and until 1963 regulated by law. In India, Dalits have, since 1947, if not before, benefited from positive discrimination in government employment and gradually in education. In Nepal there were till recently no such provisions. Comparing different groups in the country, Nepali Dalits today have the lowest life expectancy, the highest rates of illiteracy, the worst job prospects, the lowest incomes and wealth, and the worst rates of achievement in education. Of all groups of any size they are most disadvantaged and the most discriminated against. Bahuns, by contrast, do extremely well in education, have higher levels of educational attainment, and obtain more elite and professional jobs than any other group. They also provide the bulk of the political elite. Neither Dalits nor Bahuns have been studied as much they should have, given their importance in Nepali society, and this study aims to fill this gap. The political situation in Nepal is in flux. The Constituent Assembly, elected in April 2008 on the most inclusive franchise ever used in Nepal (surpassing even India's measures to ensure representation for marginal groups), failed ignominiously to produce a constitution, even after four years and four extensions of time, in May 2012. The Supreme Court refused to prolong the Assembly, leaving Nepal with a caretaker Prime Minister, no parliament, and an uncertain future. The key issue, over which the constitution-writing faltered, was that of ethnicity. In this context, it was essential to understand from the bottom up, the new process of ethnic identity formation among Bahuns and Dalits - a reaction to the much longer-standing and politically more assertive ethnicity formation among ex-tribal Janajati groups. This project aimed to examine in detail exactly how the patterns of disadvantage and exclusion, on the one hand, and achievement and success, on the other, are produced and reproduced. In doing so it focused on six neighbouring villages in west central Nepal where the two largest population groups are Bahuns and Dalits.

The qualitative data was collected through semi-structured interviews with Dalit activists, scholars and politicians using purposive sampling. The quantitative data was collected through a survey of individuals. It was built upon an initial census-type household exploration representing different caste groups, including Bahuns and Dalits, from a set of six neighbouring villages. First, households were selected using stratified random sample (caste and class) taking 50% of the original households, and individuals 13 years or older were administered the survey questionnaire. Some temporarily migrated individuals, as well as some permanent migrants, were also interviewed at their migration destinations in Nepal. Altogether 1,203 respondents were covered in the face to face survey.