In this page:
The main purpose of the pilot survey was to test both the questionnaire (modules specifically focused on discrimination experiences and the question focused on respondents’ self-identification in terms of ethnic origin) and the sample design, to ensure that the diversity of the resident population was captured (mainly the minority ethnic groups).
The policy framework of this project is the national parliament’s recommendation to the government for implementing a national strategy against racism (Resolution of the Assembly of the Republic No. 16/2021) and for adopting measures to combat racism, within which a recommendation for collecting data on ethnic discrimination was made (Resolution of the Assembly of the Republic No. 11/2021).
Besides the policy framework mentioned above, various data users have asked Statistics Portugal to collect data on the ethnic origins of the population, and on experiences of discrimination and inequality in several domains that might be related to the individuals’ migratory background and ethnic origins.
In Portugal, the debate about the need to measure ethnic and racial self-identification began a few years ago. There had been a widespread debate about the possibility of including a question on ethnic and racial self-identification in the 2021 Census. A working group was set up, which was composed of academic researchers and members of several stakeholders (the High Commission for Migration, the Commission for Equality and Against Racial Discrimination and various representatives of ethnic minorities) who worked together to provide recommendations around this issue, and to propose a question and a classification to be included in taking the Census. After several tests in different focus groups, and long discussion on the subject, a final question and a classification were proposed. At the end, Statistics Portugal decided not to include the question in the 2021 Census, but committed itself to developing a survey on this issue, allowing for better characterisation of the population and more complete results, by including several other questions that it would not be possible to include in the Census.
The design of the pilot questionnaire was prepared internally by a team from different Statistics Portugal units. Researchers and several anti-racism organisations were consulted, and the discussion was extended to the High Commission for Migration.
The major challenge concerned the development of a classification of the ethnic origin of the population without having harmonised and common guidelines applied widely in other European countries.
The question for ethnic and racial self-identification was developed from a first typology proposed by the Census working group. The question proposed to test the pilot questionnaire resulted from a simplification of the initial typology, which originally included two levels: the first was a more general classification, in which phenotype was included; and the second included geographical origins.
The second level of this classification was measured in the pilot by collecting data on the country of birth of respondents, their parents and their grandparents. The question in the pilot included only the major categories. Using information from the pilot, it is possible to combine migratory background (up to the third generation) and ethnic self-identification.
For more information see the screenshot of the survey question on self-identification.
The self-classification question includes the categories ‘Prefer not to answer’ and ‘Don’t know’, as well as the option of ‘Doesn’t identify with any of these groups’ (leading the respondents to an open question on how they describe and identify themselves and their origins). There is a question about religion, as it is an important element of cultural background.
To include other important categories that were not considered in the previous classification of ethnic and racial self-identification, respondents were also asked to describe their origins in their own words. One of the main objectives of this survey was to create a comprehensible and not too extensive classification (with a limited number of categories) of ethnic and racial background of the resident population to use in the future main questionnaire. Another important issue asked about in the pilot survey was experiences and perceptions of discrimination. These questions were about individuals’ experiences of discrimination, and their perceptions of the characteristics and grounds of discrimination (with an extensive list of discrimination factors) that could be the basis for their experiences. Perceptions of discrimination in Portuguese society were also observed, as were the factors behind it, by using the same list of discrimination factors.
Respondents were also asked specifically about their understanding of the (non-)existence of discrimination against ethnic minorities and immigrants, and about their perceptions of discrimination in relation to the COVID-19 pandemic. The results of the pilot will be analysed to improve the final questionnaire and methodology, and the sample design.
Regarding the sample design, since the pilot questionnaire aimed to test the wording of questions, answering options, structure, etc. and the different interview modes (computer-assisted personal, telephone or web interviewing), the option was to apply the questionnaire only to a NUTS 2 region (Área Metropolitana de Lisboa). It was assumed that the sampling would contain some level of bias: dwellings were chosen among those where there was a higher concentration of individuals who were born abroad (this information was obtained from various household surveys).
Moreover, the data collection that Statistics Portugal has been doing helped to select smaller areas within the NUTS 2 region, mostly parishes and districts where there are more individuals from different backgrounds and ethnic origins. Given the objective of the survey, it was important to test the questionnaire with individuals from different backgrounds.
Owing to the personal and sensitive nature of the questions that were asked, and the entry into force of the General Data Protection Regulation, the questionnaire was developed in cooperation with Statistics Portugal’s Data Protection Officer.
The content of the questionnaire raised several concerns in terms of its complexity, its sensitive context and its comprehensiveness, which carried the risk of getting negative reactions from individuals and hampering the development of the overall project.
In this context, some adjustments needed to be made. (1) The survey was made not mandatory (in contrast to all household surveys). (2) The names of the respondents and those living in the household were left optional. (3) Interviewers were obliged to submit interviews to the central box immediately after finishing them so that personal information from interviewees could be erased from their work computers as soon as possible (in cases of computer-assisted personal or telephone interviewing modes). (4) Names, addresses and other characteristics of the household members were not included in the final database for internal analysis purposes. (5) The flow of all data within Statistics Portugal computer systems was assessed so that no security concerns would be raised.
The pilot survey data collection took place between October 2021 and January 2022. The first results are being analysed with the purpose of forming the main survey. In addition to this, meetings were held with the staff involved in data collection (interviewers and supervisors) to collect their feedback on the fieldwork, the difficulties and the perceptions of respondents about the questionnaire.
The main survey preparation started in April 2022, and the fieldwork is expected to run from the end of 2022 to March 2023. The release of the main results is expected to take place by the end of July 2023.
The data collection ended recently, and now the statistical analysis is taking place. However, some important points can already be highlighted:
António Ribeiro: antonio.ribeiro@ine.pt
