Abstract
Life course studies have the ambitious goal of relating one portion of the life history to subsequent outcomes, which may themselves be later fragments of the overall history. These studies may also wish to relate certain domains, or dimensions of the life course to others, as they each evolve over time. How can these partial histories be operationalised and incorporated into statistical models? Does the process of doing this require additional assumptions or even limit the types of statements we can make about individuals, their life choices and subsequent prospects? From what social theory can we draw when making these choices? Beginning with an abstract framing of this fundamental problem in social science research, we connect three very different statistical models for life course outcomes to theoretical models based in the social sciences. We show how the choices surrounding historical context have deep implications for interpretation through their connection to theoretical frameworks in life course research. We demonstrate that these models inform individual-specific and population-average interpretations of correlates of change and their corresponding life course pathways. Each approach contributes a unique perspective upon which a more comprehensive narrative may be constructed. We illustrate the models and their interpretation using co-residence information in the transition to adulthood using the Swiss Household Panel.
Similar content being viewed by others
Notes
Under the Rubin Model for Causal Inference (Rubin 2005; Holland 1986), we wish to compare potential outcomes, typically under different treatments. In life course analysis, we connect multiple, evolving processes to potential futures, and the counterfactuals are by definition alternative prior histories.
The underlying transition or output model is usually a multinomial logistic regression; conditioning on covariates with these models often leads to an explosion of parameters that is rarely supported with available data.
It seems reasonable to focus on a single outcome and its relationship to historical information in this theoretical framework.
Financed by the Swiss National Science Foundation. Distributed by FORS, Lausanne, 2021. https://www.swissubase.ch/en/catalogue/studies/6097/17007/description.
We also augmented the nominal states to allow for gaps; using all of the respondents we obtained the same findings, qualitatively, but we report the results from the slightly smaller sample here to correspond most closely to nominal states used in the other models.
These findings are consistent with those of Rossignon (2017, PhD thesis, p. 144). The author posited that these results were related to the fact that children living with a lone parent come from lower socio-economic status with related stressors, demands and constraints, including earlier school-to-work transitions (Kiernan 2006).
Arguably, we should not expect history to be reflected in an HMM, due to the memoryless nature of the (hidden) Markov process. We do identify implicit indications of history embedding, but refer to Han et al. (2020), who describe the limitations to this.
Note that seqsha, an R function implementing the methods of Rossignon, et al. (2018) has recently been added to the TraMineRextras package on CRAN. It was not available when the authors were completing this research.
Based on a re-fitting procedure available in the software package.
An extension of the FMM model to capture this heterogeneity, for example with a gamma-mixture, is possible, but beyond the scope of this article.
References
Abbott, A.: Conceptions of time and events in social science methods. Historical Methods 23(4), 140–150 (1990)
Abbott, A.: From causes to events: notes on narrative positivism. Soc. Methods and Res. 20(4), 428–455 (1992)
Abbott, A.: Time Matters: On theory and method. The University of Chicago Press (2001)
Abbott, A.: Processual Sociology. University of Chicago Press. (2016).
Aisenbrey, S., Fasang, A.E.: New life for old ideas: the “second wave” of sequence analysis bringing the “course” back into the life course. Sociol. Methods Res. 38(3), 420–462 (2010)
Aisenbrey, S., and Fasang, A. E.: Social location matters: Inequality in work and family life courses at the intersection of gender and race (No. SP I 2018–601). WZB Discussion Paper (2018)
Banfield, J.D., Raftery, A.E.: Model-based Gaussian and non-Gaussian clustering. Biometrics 49, 803–821 (1993)
Barban, N., Billari, F.C.: Classifying life course trajectories: a comparison of latent class and sequence analysis. Appl. Stat. 61, 765–784 (2012)
Bartolucci, F., Pandolfi, S., Pennoni, F., LMest,: An R package for latent markov models for longitudinal categorical data. J. Statistical Softw. 81, 1–38 (2017)
Baum, L.E., Petrie, T.: Statistical inference for probabilistic functions of finite state Markov chains. Ann. Math. Stat. 37(6), 1554–1563 (1966). https://doi.org/10.1214/aoms/1177699147
Becker, S., Sosa, D.: An experiment using a month-by-month calendar in a family planning survey in costa rica. Stud. Fam. Plann. 23(6), 386–391 (1992). https://doi.org/10.2307/1966896
Belli, R.F.: The structure of autobiographical memory and the event history calendar: potential improvements in the quality of retrospective reports in surveys. Memory 6(4), 383–406 (1998). https://doi.org/10.1080/741942610
Bolano, D., and Studer, M.: The link between previous life trajectories and a later life outcome: A feature selection approach. LIVES Working paper, 82, 1–38 (2020)
Brückner, H., Mayer, K.U.: De-standardization of the life course : what it might mean? And if it means anything, whether it actually took place? Adv. Life Course Res. 9, 27–53 (2005)
Clogg, C. C.: Latent class models. In Arminger, G., Clogg, C. C., and Sobel, M. E. (Eds.). Handbook of statistical modeling for the social and behavioral sciences (pp. 311–359). Springer, Boston, MA (1995)
De Coninck, F., and Godard, F.: L’approche biographique à l’épreuve de l’interprétation: Les formes temporelles de la causalité. Revue Française de Sociologie, pp. 23–53 (1990)
Di Prete, T., Eirich, G.R.: Cumulative advantages as a mechanism for inequality. A review of theoretical and empirical developments. Ann Rev. Soc. 32, 271–297 (2006)
Elder, G. H., Johnson, M. K., and Crosnoe, R.: The emergence and development of life course theory. In J. T. Mortimer and M. J. Shanahan (Eds.), Handbook of the life course (pp. 3–19). Kluwer (2003)
Elder, G.H., Jr.: The life course as developmental theory. Child Dev. 69(1), 1–12 (1998)
Elzinga, C.H.: Complexity of categorical time series. Soc. Methods Res. 38(3), 463–481 (2010). https://doi.org/10.1177/0049124109357535
Fraley, C., Raftery, A.E.: How many clusters? Which clustering method? Answers via model-based cluster analysis. Comput. J. 41, 578–588 (1998)
Fuller, S.: Do pathways matter? Linking early immigrant employment sequences and later economic outcomes: evidence from Canada. Int. Migr. Rev. 49(2), 355–405 (2015)
Gabadinho, A., Ritschard, G., Muller, N.S., Studer, M.: Analyzing and visualizing state sequences in R with TraMineR. J. Stat. Softw. 40(4), 1–37 (2011)
Goodman, L.A.: Exploratory latent structure analysis using both identifiable and unidentifiable models. Biometrika 61(2), 215–231 (1974)
Grün, B., Leisch, F.: FlexMix version 2: finite mixtures with concomitant variables and varying and constant parameters. J. Stat. Softw. 28(4), 1–35 (2008). https://doi.org/10.18637/jss.v028.i04
Han, Y., Liefbroer, A.C., Elzinga, C.H.: Comparing methods of classifying life courses: sequence analysis and latent class analysis. Longitudinal and Life Course Stud. 8(4), 319–341 (2017)
Han, S.Y., Liefbroer, A.C., Elzinga, C.H.: Mechanisms of family formation: an application of Hidden Markov Models to a life course process. Advances in Life Course Research, p. 43 (2020)
Helske, S., Helske, J.: Mixture hidden markov models for sequence data. The seqHMM package in R. J. Statistical Softw. 88(3), 1–32 (2019)
Helske, S., Helske, J., and Chihaya, G. K.: From sequences to variables – Rethinking the relationship between sequences and outcomes. INVEST Working Paper 36/2021. https://doi.org/10.31235/osf.io/srxag (2021). Accessed 3 Jan 2022
Hekman, S. J.: Weber, the ideal type, and contemporary social theory. University of Notre Dame Press (1983)
Holland, P.W.: Statistics and causal inference. J. Am. Stat. Assoc. 81(396), 945–960 (1986)
Imai, K., Kim, I.S., Wang, E.S.: Matching methods for causal inference with time-series cross-sectional data. Am. J. Political Sci. (2021). https://doi.org/10.1111/ajps.12685
Kasy, M.: A nonparametric test for path dependence in discrete panel data. Econ. Lett. 113(2), 172–175 (2011)
Kohli, M.: The world we forgot: a historical review of the life course. In V.W. Marshall (Ed.), Later Life. The social psychology of aging (pp. 271–303), London: Sage. (1986)
Lazarsfeld, P.F., Henry, N.W.: Latent structure analysis. Houghton Mifflin Co., Boston (1968)
Levy, R.: Analysis of life courses-A theoretical sketch. In R. Levy, and E. Widmer (Ed.) Gendered Life Courses Between Individualization and Standardization. A European Approach Applied to Switzerland, 13–36 (2013)
McKinney, J.C.: Typification, typologies, and sociological theory. Soc. Forces 48(1), 1–12 (1969)
Morselli, D., Le Goff, J.-M., Gauthier, J.-A.: Self-administered event history calendars: a possibility for surveys? Contemp. Soc. Sci. 14(3–4), 423–446 (2019)
Palardy, G.J., Vermunt, J.K.: Multilevel growth mixture models for classifying groups. J. Educ. Behav. Statistics 35, 532–565 (2010)
Pelletier, D., Bignami, S., Simard-Gendron, A.: Measuring life course complexity with dynamic sequence analysis. Soc. Indic. Res. 152(1), 1127–1151 (2020). https://doi.org/10.1007/s11205-020-02464-y
Piccarreta, R, and Studer, M.: Holistic analysis of the life course: Methodological challenges and new perspectives, Advances in Life Course Research, p. 41 (2019)
Pina Sánchez, J., Koskinen, J., Plewis, I.: Measurement error in retrospective work histories. Surv Res Methods 8(1), 43–55 (2014)
Pohle, J., Langrock, R., van Beest, F.M., Schmidt, N.M.: Selecting the number of states in hidden markov models: pragmatic solutions illustrated using animal movement. J. Agric. Biol. Environ. Stat. 22(3), 270–293 (2017)
Pollien, A.: Faire une pause ou bifurquer? Essai de typologie des trajectoires de formation. Educ. Et Sociétés 2, 123–143 (2010)
Poulain, M., Riandey, B., Firdion, J.M.: Data from a life history survey and from Belgian population register: a comparison. Population: An English Select. 4, 77–96 (1992)
Pyy-Martikainen, M., Rendtel, U.: Measurement errors in retrospective reports of event histories. A validation study with Finnish register data. Surv. Res. Methods 3(3), 139–155 (2009)
Rabiner, L.: A tutorial on hidden markov models and selected applications in speech recognition. Proc. IEEE 77(2), 257–286 (1989). https://doi.org/10.1109/5.18626
Ritschard, G.: Measuring the nature of individual sequences. Soc. Methods and Res. (2021). https://doi.org/10.1177/00491241211036156
Rossignon, F.: Transition to Adulthood for Vulnerable Populations in Switzerland : How Past Trajectories Matter. PhD Thesis (2017)
Rossignon F., Studer M., Gauthier J.A., Goff J.M.L.: Sequence History Analysis (SHA): Estimating the Effect of Past Trajectories on an Upcoming Event. In: Ritschard G., Studer M. (Eds.) Sequence Analysis and Related Approaches. Life Course Research and Social Policies, vol 10. Springer, Cham. (2018). https://doi.org/10.1007/978-3-319-95420-2_6
Rubin, D.B.: Causal inference using potential outcomes. J. Am. Stat. Assoc. 100(469), 322–331 (2005). https://doi.org/10.1198/016214504000001880
Schafer, J.L.: Analysis of Incomplete Multivariate Data (1st ed.). Chapman and Hall/CRC (1997). https://doi.org/10.1201/9780367803025
Scherer, S.: Early career patterns. A comparison of great Britain and west Germany. Eur. Soc. Rev. 17, 119–144 (2001)
Schwarz, G.: Estimating the dimension of a model. Ann. Stat. 6, 461–464 (1978)
Scott, M.A., Mohan, K., Gauthier, J.-A.: Model-based clustering and analysis of life history data. J. r. Stat. Soc. a. Stat. Soc. 183(3), 1231–1251 (2020)
Studer, M., Ritschard, G.: What matters in differences between life trajectories : a comparative review of sequence dissimilarity measures. J. r. Stat. Soc. a. Stat. Soc. 179(2), 481–511 (2016)
Tillmann, R., Voorpostel, M., Antal, E., Dasoki, N., Klaas, H., Kuhn, U., Lebert, F., Monsch, G.-A., Ryser, V.-A.: The Swiss Household Panel (SHP). J. Econom. Statistics (2021). https://doi.org/10.1515/jbnst-2021-0039
Tillmann, R., Vorpoostel, M., and Farago, P.: Introduction, in Tillmann, R., Vorpoostel, M., and Farago, P. (Eds.), Social Dynamics in Swiss Society Empirical Studies Based on the Swiss Household Panel (pp. IX-XIII). Cham: Springer, col Life course Research and social policies (2018). https://doi.org/10.1007/978-3-319-89557-4
Weber, M.: Economy and Society: An Outline of Interpretive Sociology. Bedminster Press, New York (1978)
Wu, L.L.: Some comments on “Sequence analysis and optimal matching methods in sociology: review and prospect.” Soc. Methods and Res. 29(1), 41–64 (2000)
Zucchini, W., MacDonald, I.L., and Langrock, R.: Hidden Markov Models for Time Series: An Introduction Using R (2nd ed.). Chapman and Hall/CRC (2016)
Acknowledgements
The authors wish to thank LIVES, and especially the Visiting Scholars Invitation Programme (LIVES VIP). The core ideas embedded in this research stem from collaborations that began with and are a result of participation in the Programme, February, 2020, at the University of Lausanne (UNIL). The authors would like to especially thank LIVES colleagues Rossignon and Studer, for providing code associated with their work developing Sequence History Analysis. This publication benefited from the support of the Swiss National Centre of Competence in Research LIVES – Overcoming vulnerability: Life course perspectives (NCCR 51NF40–160590).
Funding
The authors have no relevant financial or non-financial interests to disclose.
Author information
Authors and Affiliations
Contributions
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by all authors. The first draft of the manuscript was primarily written by Marc Scott, but all authors contributed substantially to all sections, and then commented/edited/ammended all versions of the manuscript. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Conflict of interest
This publication benefited from the support of the Swiss National Centre of Competence in Research LIVES—Overcoming vulnerability: Life course perspectives (NCCR 51NF40–160590).
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Scott, M.A., Le Goff, JM. & Gauthier, JA. History matters: the statistical modelling of the life course. Qual Quant 58, 445–469 (2024). https://doi.org/10.1007/s11135-023-01648-1
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11135-023-01648-1