Measurement in Intensive Longitudinal Data

ABSTRACT

Technological advances have increased the prevalence of intensive longitudinal data as well as statistical techniques appropriate for these data, such as dynamic structural equation modeling (DSEM). Intensive longitudinal designs often investigate constructs related to affect or mood and do so with multiple item scales. However, applications of intensive longitudinal methods often rely on simple sums or averages of the administered items rather than considering a proper measurement model. This paper demonstrates how to incorporate measurement models into DSEM to (1) provide more rigorous measurement of constructs used in intensive longitudinal studies and (2) assess whether scales are invariant across time and across people, which is not possible when item responses are summed or averaged. We provide an example from an ecological momentary assessment study on self-regulation in adults with binge eating disorder and walkthrough how to fit the model in Mplus and how to interpret the results.

KEYWORDS:

• Measurement invariance
• EMA
• dynamic structural equation modelling
• time-series analysis
• cross-classified factor analysis

Disclosures

This work was supported by the National Institutes of Health (NIH) Science of Behavior Change Common Fund Program through an award administered by the National Institute for Drug Abuse (NIDA) (UH2/UH3DA041713).

Notes

¹ https://www.clinicaltrials.gov/ct2/show/NCT03774433?term=marsch&draw=2&rank=3.

² The Open Science Framework project link is https://osf.io/f83km

³ Consistent with Bayesian estimation, it would be more appropriate to refer to these as the median of the posterior distribution for each parameter. To keep the terminology succinct and to retain focus on aspects of the model rather than the estimation, we refer to them as “estimates” throughout the paper even though they are not technically the same as point estimates that frequentist methods would yield.

⁴ We thank Tihomir Asparouhov for providing us with these details about Mplus mentioned throughout this paragraph.

Additional information

Funding

This work was supported by the National Institutes of Health (NIH) Science of Behavior Change Common Fund Program through an award administered by the National Institute on Drug Abuse (NIDA) (UH2/UH3DA041713) and by an an award administered by the National Institute on Drug abuse [R37DA09757].