Volume 35 - Article 38 | Pages 1135–1148  

Multiple imputation for demographic hazard models with left-censored predictor variables: Application to employment duration and fertility in the EU-SILC

By Michael S. Rendall, Angela Greulich

Abstract

Objective: A common problem when using panel data is that individuals’ histories are incompletely known at the first wave. We demonstrate the use of multiple imputation as a method to handle this partial information, and thereby increase statistical power without compromising the model specification.

Methods: Using EU-SILC panel data to investigate full-time employment as a predictor of partnered women’s risk of first birth in Poland, we first multiply imputed employment status two years earlier to cases for which employment status is observed only in the most recent year. We then derived regression estimates from the full, multiply imputed sample, and compared the coefficient and standard error estimates to those from complete-case estimation with employment status observed both one and two years earlier.

Results: Relative to not being full-time employed, having been full-time employed for two or more years was a positive and statistically significant predictor of childbearing in the multiply imputed sample, but was not significant when using complete-case estimation. The variance about the ‘two or more years’ coefficient was one third lower in the multiply imputed sample than in the complete-case sample.

Contribution: By using MI for left-censored observations, researchers using panel data may specify a model that includes characteristics of state or event histories without discarding observations for which that information is only partially available. Using conventional methods, either the analysis model must be simplified to ignore potentially important information about the state or event history (risking biased estimation), or cases with partial information must be dropped from the analytical sample (resulting in inefficient estimation).

Author's Affiliation

Other articles by the same author/authors in Demographic Research

Measuring the educational gradient of period fertility in 28 European countries: A new approach based on parity-specific fertility estimates
Volume 49 - Article 34

Education, labour, and the demographic consequences of birth postponement in Europe
Volume 36 - Article 23

The quality of periodic fertility measures in EU-SILC
Volume 36 - Article 17

Most recent similar articles in Demographic Research

The short- and long-term determinants of fertility in Uruguay
Volume 51 - Article 10    | Keywords: fertility, panel data, stages of female reproductive life, time series, Uruguay

Predictive utility of key family planning indicators on dynamic contraceptive outcomes: Results from longitudinal surveys in Burkina Faso, Kenya, Uganda, and Côte d'Ivoire
Volume 50 - Article 45    | Keywords: contraception, contraceptive adoption, contraceptive discontinuation, contraceptive use, family planning, longitudinal data, methods, panel data, Performance and Monitoring for Action (PMA) surveys, sub-Saharan Africa

Gone and forgotten? Predictors of birth history omissions in India
Volume 50 - Article 32    | Keywords: fertility history, interviewer effects, interviewer observations, measurement error, missing data, panel data, survey methodology

The big decline: Lowest-low fertility in Uruguay (2016–2021)
Volume 50 - Article 16    | Keywords: adolescent fertility, birth order, fertility, Latin America, ultra-low fertility, Uruguay

Cohort fertility of immigrants to Israel from the former Soviet Union
Volume 50 - Article 13    | Keywords: age at first birth, assimilation, cohort analysis, fertility, immigration, parity, religiosity