Examinamos a realização do aborto segundo cor ou raça das mulheres no Brasil usando dados de três edições da PNA, 2016, 2019 e 2021. Comparamos diferenciais raciais aplicando testes de diferenças de médias a dados de levantamentos separados, levantamentos combinados sem reponderação, e levantamentos combinados e reponderados. Na comparação usamos ainda modelos logísticos para a chance de realização de aborto. Os resultados indicam que há diferencial racial consistente nas três edições da PNA, com os percentuais de aborto entre as mulheres negras sendo mais elevados do que entre as mulheres brancas. Nos levantamentos combinados e reponderados, entre as mulheres negras de todas as idades a probabilidade de ter feito um aborto é de 11,03% enquanto entre as mulheres brancas é de 7,55%. Isso significa uma diferença de 3,5 pontos percentuais, que se traduz em uma probabilidade 46% maior para as negras, valores estatisticamente significantes para um intervalo de 5%. Regressões logísticas produzem resultados similares, com probabilidade média predita das negras de 12,61% e das brancas de 8,90%, também significantes. As diferenças raciais persistem estatisticamente significantes para várias combinações de edições da PNA. Porém, nos levantamentos isolados os testes de diferenças de média são estatisticamente significantes apenas em 2016. Devido ao tamanho reduzido das amostras não é possível dizer muito sobre as diferenças entre mulheres amarelas e indígenas.
Palavras-chave:
Aborto. Cor ou raça. Pesquisa Nacional de Aborto. Brasil
Abstract:
We examined the incidence of abortion by color or race of women in Brazil using datathree editions of the PNA, 2016, 2019 and 2021. We compared racial differentials by applying tests of differences in means to data in separate surveys, combined surveys without reweighting, and combined and reweighted surveys. In the comparison, we also used logistic models for estimating the chance of having an abortion. The results indicate that there is a consistent racial difference in the three editions of the PNA, with the percentage of abortion among black women being higher than among white women. In the combined and reweighted surveys, among black women of all ages, the probability of having had an abortion is 11.03%, while among white women it is 7.55%. This is a difference of 3.5 percentage points, which translates into a 46% greater probability for black women. All values statistically significant for a 5% interval. Logistic regressions produce similar results, a mean predicted probability of blacks of 12.61% and of whites of 8.90%, also significant. Racial differences remain statistically significant for various combinations of PNA editions. However, in the isolated surveys, the tests of mean differences are statistically significant only in 2016. Due to the reduced sample size, it is not possible to say much about the differences between Asian and indigenous women.
Keywords:
Abortion. Color or race. National Abortion Survey. Brazil
We examined the incidence of abortion by color or race of women in Brazil using datathree editions of the PNA, 2016, 2019 and 2021. We compared racial differentials by applying tests of differences in means to data in separate surveys, combined surveys without reweighting, and combined and reweighted surveys. In the comparison, we also used logistic models for estimating the chance of having an abortion. The results indicate that there is a consistent racial difference in the three editions of the PNA, with the percentage of abortion among black women being higher than among white women. In the combined and reweighted surveys, among black women of all ages, the probability of having had an abortion is 11.03%, while among white women it is 7.55%. This is a difference of 3.5 percentage points, which translates into a 46% greater probability for black women. All values statistically significant for a 5% interval. Logistic regressions produce similar results, a mean predicted probability of blacks of 12.61% and of whites of 8.90%, also significant. Racial differences remain statistically significant for various combinations of PNA editions. However, in the isolated surveys, the tests of mean differences are statistically significant only in 2016. Due to the reduced sample size, it is not possible to say much about the differences between Asian and indigenous women.
Palavras-chave (keywords):
Abortion. Color or race. National Abortion Survey. Brazil
Diniz, D., Medeiros, M., Souza, P.H.G.F, Goés, E.. Aborto e raça no Brasil, 2016 a 2021. Cien Saude Colet [periódico na internet] (2023/set). [Citado em 23/12/2024].
Está disponível em: http://cienciaesaudecoletiva.com.br/artigos/aborto-e-raca-no-brasil-2016-a-2021/18886?id=18886
