Do State and Trait Affect Measures Retain Their Measurement Properties during a Disaster? An Investigation of Measurement Invariance during the COVID-19 Pandemic.

Psychologists have begun to study the impact of the COVID-19 pandemic on emotionality, though such investigations assume that the measurement properties of affect scales have not changed as a function of the pandemic. To our knowledge, no prior study has tested measurement invariance (MI) of an affect scale during a disaster, and very few studies have explored MI of scales administered through ecological momentary assessment (EMA). The current study tested invariance of trait and state affect measures across homogenous groups of 18-year-olds assessed prior to and during the first acute phase of the COVID-19 pandemic in Long Island, New York. Trait affect was measured with a single administration of the Schedule for Nonadaptive and Adaptive Personality-Youth Version. State affect was assessed with items developed for the EMA portion of this study, which were administered 5 times daily for 14 consecutive days using a smartphone application. A baseline correlated 2-factor (positive and negative affect) model was fit for both measures. Invariances tests established up to strict MI across pre-/COVID-19 groups for both affect measures. These findings suggest that valid comparisons of observed score means and variances can be made between groups assessed before and during the COVID-19 pandemic.

Testing Invariance of Measures of Internalizing Symptoms Before and After a Major Life Stressor: The Impact of COVID-19 in an Adolescent and Young Adult Sample.

There are reports of increases in levels of internalizing psychopathology during the COVID-19 pandemic. However, these studies presume that measurement properties of these constructs remained unchanged from before the pandemic. In this study, we examined longitudinal measurement invariance of assessments of depression, anxiety, and intolerance of uncertainty (IU) in adolescents and young adults from ongoing longitudinal studies. We found consistent support for configural and metric invariance across all constructs, but scalar invariance was unsupported for depression and IU. Thus, it is necessary to interpret pandemic-associated mean-level changes in depression and IU cautiously. In contrast, mean-level comparisons of panic, generalized, and social anxiety symptoms were not compromised. These findings are limited to the specific measures examined and the developmental period of the sample. We acknowledge that there is tremendous distress accompanying disruptions due to the COVID-19 outbreak. However, for some instruments, comparisons of symptom levels before and during the pandemic may be limited.

Trajectories of depression, anxiety and pandemic experiences; A longitudinal study of youth in New York during the Spring-Summer of 2020.

Initial reports suggest that mental health problems were elevated early in the COVID-19 pandemic. However, few studies have followed-up participants as the pandemic evolved and examined both between and within person predictors of symptom trajectories. In the current study, adolescents and young adults (N=532) in New York were surveyed monthly between March 27th and July 14th, 2020, a period spanning the first peak and subsequent decline in COVID-19 infection rates in the region. Surveys assessed symptoms of depression and anxiety using the Child Depression Inventory and the Screen for Child Anxiety Related Disorders, as well as experiences related to the pandemic. Multilevel growth modeling indicated that symptoms of depression and anxiety peaked around late April/early May and then decreased through May-July. Some pandemic experiences followed a similar quadratic trajectory, while others decreased linearly across the study. Specific relationships emerged between some types of pandemic experiences and depression and anxiety symptoms. While symptoms of depression and anxiety in youth may have been elevated early in the pandemic, these findings suggest they subsided across Spring-Summer of 2020, with higher levels of both corresponding to a period of peak infection rates and decreases paralleling the decline in pandemic experiences and COVID-19 infection rates.