Page 29 - Never Too Far Away? The Roles of Social Network Sites in Sojourners’ Adjustment

interactions and perceived social support, as well as perceived social support and depression. In Chapter 3, I focus on the reciprocal causal associations between social interactions and homesickness, as well as homesickness and sociocultural adjustment.
Long-term vs. Short-term Effects
One crucial factor to consider in testing the proposed general framework with the hypothesized reciprocal effects is the time lag and the question whether effects occur over a long period of time or within a short span of time. As of yet, there are neither established nor compelling theoretical bases for the timing of effects of social interactions and outcomes. In Chapters 2 and 3, I make use of a longitudinal design with an interval of three months in-between measurements, consistent with a previous longitudinal study on student sojourners (Hechanova- Alampay, Beehr, Christiansen, & Van Horn, 2002). According to Hechanova- Alampay et al. (2002), student sojourners might experience varying levels of stress over the period of 6 months. They found that the highest level of stress was 3 months into the academic year, and then 6 months after the start of the academic year. The differences in stress levels coincided with important class requirements. For example, the preparations for the final exams are usually scheduled 3 months into a term.
In this dissertation, I investigate both the patterns of associations between social interactions, subjective outcomes, and adjustment of international students both within a short period of time (short-term), as well as over time (long-term). Long-term associations are based on the effects predicted over a three-month time interval. Meanwhile, short-term associations, are based on the effects predicted at the same time point or wave. This is not to say that the effects are instantaneous, but rather that the causal lag for the effects is short relative to the time elapsed (i.e., three-month interval) between waves of measurement (Finkel, 1995). The longitudinal design employed in Chapters 2 and 3 of this dissertation is helpful in examining the temporal order and the immediacy of effects by testing both long- term and short-term reciprocal causal associations.
Introduction 27