Page 50 - IJPS-2-2
P. 50

International Journal of Population Studies


       RESEARCH ARTICLE


       Latent class models for cross-national compar-

       isons: the association between individual and

       national-level fertility and partnership charac-

       teristics




                         *
       Mark Lyons-Amos
       London School of Hygiene and Tropical Medicine, Keppel Street, London, UK


       Abstract: Multilevel modelling techniques such as random models or fixed effect are increasingly used in
       social sciences and demography to both account for clustering within higher level aggregations and eva-
       luate the interaction between individual and contextual information. While this is justifiable in some stu-
       dies, the extension of multilevel models to national level analysis — and particularly cross-national com-
       parative analysis — is problematic and can hamper the understanding of the interplay between individual
       and country level characteristics. This paper proposes an alternative approach, which allocates countries
       to classes based on economic, labour market and policy characteristics. Classes influence the profiles of
       three key demographic behaviours at a sub-national level: marriage, cohabitation and first birth timing.
       Woman level data are drawn from a subset of the Harmonized Histories dataset, and national level infor-
       mation from the GGP contextual database. In this example, three country classes are extracted reflecting
       two Western patterns and an Eastern pattern, divided approximately along the Hajnal line. While West-
       ern countries tend to exhibit higher levels of family allowances albeit accounting for a lower share of
       spending which is associated with lower marriage and later fertility, Eastern countries generally show a
       higher share of spending but at lower absolute levels with lower cohabitation rates and early fertility.
       Keywords: multilevel models, latent class analysis, cross-national, marriage, cohabitation, fertility


       *Correspondence to: Mark Lyons-Amos, London School of Hygiene and Tropical Medicine, Keppel Street, London, UK;
       Email: mark.lyons-amos@lshtm.ac.uk

       Received: May 10, 2016; Accepted: June 28, 2016; Published Online: July 5, 2016
       Citation: Lyons-Amos M. (2016). Latent class models for cross-national comparisons: the association between indi-
       vidual and  national-level fertility and  partnership  characteristics.  International Journal of  Population Studies,
       vol.2(2): 44–57. http://dx.doi.org/10.18063/IJPS.2016.02.002.

       1. Introduction                                                                        Copyright: © 2016 Mark Ly-
                                                                                              ons-Amos. This is an Open Ac-
                                                                                              cess article distributed under
       Multilevel  modelling is  increasingly popular  within both general social  science  and demography.   the terms of the Creative Com-
       The importance of the interaction between macro and micro level is recognised within demographic   mons  Attribution-NonC-  om-
       literature (e.g., Neels, Theunynck, and Wood, 2013; Billingsley and Farrini, 2014; Perelli-Harris and   mercial  4.0  International  Li-
                                                                                              cense (http://creativecomm-
       Sánchez-Gassen, 2012). Multilevel models seem a natural  method to be applied for this research   ons.org/licenses/by-nc/4.0/),
       prima facie, due to their ability to partition variation in demographic outcomes between individual   permitting all non-commer-cial
       and contextual influences, and explicitly model the relationship between individual and macro level   use, distribution, and reprodu-
       (e.g., national policies) characteristics.                                             ction  in  any  medium,  provi-
                                                                                              ded the  original  work is pro-
         Within the longitudinal context, multilevel models have been frequently applied to examining the   perly cited.

       44                     International Journal of Population Studies | 2016, Volume 2, Issue 2
   45   46   47   48   49   50   51   52   53   54   55