Abbie Cahoon, University of Ulster
Longitudinal studies are often considered to be a gold standard for research (Vincent et al., 2012). Developmental scientists have argued that the application of longitudinal methods is necessary, as these studies have the capacity to provide invaluable information for understanding developmental change and specifically causal mechanisms (Stevenson, 2016; Magnusson & Stattin, 2006). However, most analytical strategies invoke a variable-oriented approach, as opposed to a person-oriented approach. In variable-oriented approaches, such as regression analysis, factor analysis, and structural equation modelling, the emphasis is on identifying relations between variables, and it is assumed that these relations apply across all people. In contrast, in person-oriented approaches, such as cluster analysis, latent profile analysis and latent transition analysis, the emphasis is on the individual, looking at subtypes of participants that exhibit similar patterns of individual characteristics (Bergman & Magnusson, 1997). This self-reflective article investigates my experience of exploring the use of Mplus (Muthén & Muthén, 2009) for complex longitudinal methods though my PhD research program.
During my PhD I had the pleasure of completing a longitudinal study on how young children develop mathematical skills over time. At first, the largescale work load felt truly impossible. However, I knew the benefits of this type of research would be vast and looking at changes in development though longitudinal methods seemed fascinating. A year later and I’m finished, after working with some wonderful people, I feel a sense of discovery and realisation that the largescale work load was actually possible.
Many mathematical cognition research questions require methods that take a person-centred approach, yet this is rarely achieved. Latent profile analysis and latent class analysis are types of models that are used to trace back the heterogeneity in a group to a number of underlying homogeneous subgroups, at a specific measurement point (Hickendorff et al., 2017). Thus, the profiles that are formed obtain as much similarity within a profile while at the same time as much difference between the profiles as possible (Lanza & Cooper, 2016). A Latent Transition Analysis (LTA) is the longitudinal extension of these models where the transitional component reflects changes in learners’ profiles over time, demonstrating potential non-linear learning pathways (Hickendorff et al., 2017).
In order to make an informed decision on the precise learner profiles and pathways of individuals during the transition between pre-school and school in mathematical skills I decided to use LTA. This method was very insightful for understanding children’s profile characteristics as children vary substantially in their level of number knowledge prior to school-entry (Manolitsis, Georgiou & Tziraki, 2013; Zill & West, 2001). I embarked on a massive learning curve by self-teaching myself this analysis method. Below are some recommended reads for those looking to explore person-oriented approaches, in particular latent transition analysis, to statistical analysis. I found these really helpful:
1) Collins, L. M., & Lanza, S. T. (2010). Latent class analysis with covariates. Latent class and latent transition analysis: With applications in the social, behavioral, and health sciences, 149-177.
2) Hickendorff, M., Edelsbrunner, P. A., McMullen, J., Schneider, M., & Trezise, K. (2017). Informative tools for characterizing individual differences in learning: Latent class, latent profile, and latent transition analysis. Learning and Individual Differences.
3) Kam, C., Morin, A. J., Meyer, J. P., & Topolnytsky, L. (2016). Are commitment profiles stable and predictable? A latent transition analysis. Journal of Management, 42(6), 1462-1490.
Bergman, L. R., & Magnusson, D. (1997). A person-oriented approach in research on developmental psychopathology. Development and psychopathology, 9(2), 291-319.
Duncan, G. J., Dowsett, C. J., Claessens, A., Magnuson, K., Huston, A. C., Klebanov, P., ... & Sexton, H. (2007). School readiness and later achievement. Developmental psychology, 43(6), 1428.
Fryer, L. K. (2017). (Latent) transitions to learning at university: A latent profile transition analysis of first-year Japanese students. Higher Education, 73(3), 519-537.
Hickendorff, M., Edelsbrunner, P. A., McMullen, J., Schneider, M., & Trezise, K. (2017). Informative tools for characterizing individual differences in learning: Latent class, latent profile, and latent transition analysis. Learning and Individual Differences.
Lanza, S. T., & Cooper, B. R. (2016). Latent class analysis for developmental research. Child Development Perspectives, 10(1), 59-64.
Magnusson, D., & Stattin, H. (2006). The person in context: A holistic‐interactionistic approach. Handbook of child psychology.
Manolitsis, G., Georgiou, G. K., & Tziraki, N. (2013). Examining the effects of home literacy and numeracy environment on early reading and math acquisition. Early Childhood Research Quarterly, 28(4), 692-703.
Muthén, B., & Muthén, B. O. (2009). Statistical analysis with latent variables. Hoboken: Wiley.
Stevenson, N. (2016). Reflections upon the experience of longitudinal research into cultural event production in a developing destination. International Journal of Tourism Research, 18(5), 486-493.
Vincent, K. B., Kasperski, S. J., Caldeira, K. M., Garnier-Dykstra, L. M., Pinchevsky, G. M., O’grady, K. E., & Arria, A. M. (2012). Maintaining superior follow-up rates in a longitudinal study: Experiences from the College Life Study. International journal of multiple research approaches, 6(1), 56-72.
Zill, N., & West, J. (2001). Entering Kindergarten: A Portrait of American Children When They Begin School. Findings from the Condition of Education, 2000.