A Meta-analysis of the Relation Between Spatial Anxiety and Spatial Skills

• In Event: Poster Session 03
  In Poster Session: PS 03 Topic - Cognitive Processes

Abstract

Research has shown spatial skills to be key predictors of achievement and attainment in science, technology, engineering and math (STEM) disciplines, despite being skills acquired through everyday life and not specifically taught in schools (e.g., Uttal et al., 2013; Shea, et al., 2001). Spatial skills have been defined as a diverse group of skills broadly related to reasoning about properties of space such as distance and direction (Vasilyeva & Lourenco, 2010). Recently, there has been more research investigating the link between spatial skills and spatial anxiety, which is a fear or apprehension felt when engaged in spatial thinking (Geer, 2019; Lyons et al., 2019; Malanchini et al., 2017). Despite this, there has yet to be a meta-analysis summarizing the results. Thus, the goal of this pre-registered meta-analysis is to provide an estimate of the size of the relation between spatial anxiety and spatial skills while considering several moderators (age/grade, sex, spatial skills measure/type, spatial anxiety measure/subtype, country, publication type/year, and study quality). Based on our knowledge of the existing research, we anticipate the results to demonstrate a significant, small-to-moderate negative association between spatial anxiety-spatial skills such that greater anxiety is associated with worse spatial skill (Lawton, 1994; Lyons, et al. 2018).
Before conducting our search, we preregistered this meta-analysis on Prospero. We searched five online databases (PsycINFO, ERIC, PubMed, Google Scholar, Web of Science) with the search terms: (spatial OR visuospatial or “mental rotation”) AND (“spatial anxiety”). From this search, we identified 1736 articles, but 131 were removed for being duplicates. We screened the remaining 1605 articles using titles and abstracts, sorting them into ‘include’, ‘exclude’, or ‘maybe’ categories. In Round 2, we assessed eligibility for inclusion for the 143 studies that were coded as ‘include’ or ‘maybe’ using full-text articles and sorted them into ‘include’ or ‘exclude’ categories. We also emailed the CogDevSoc listserv to request unpublished data, accounting for grey literature. With the inclusion of grey literature, 95 studies went on to full coding, which is currently underway. See Figure 1 for flow chart for current progress.
After coding, we will determine the overall effect size of the relation between spatial anxiety and spatial skills using the metafor package in R to conduct multilevel meta-analyses to model the nested structure of the data (some samples have multiple correlations reported; Maas & Hox, 2004). We will then examine the impact of our moderators using interactions (i.e., age). Based on current progress, analyses will be completed by December 2020.
The results will provide a more generalizable view of the relation between spatial anxiety and spatial skills. The moderator analysis examining the impact of age on this relation will help us understand the development of the relation between these constructs and can help inform future research on intervention and when it is best to intervene. The other moderation analyses will allow us to examine the specificity of the relations between the different subtypes of spatial anxiety and spatial skills, giving us more insight into the nature of these relations.

Authors

• Elyssa A Geer, University of Oregon
  Presenting Author

• Colleen M Ganley, Florida State Univeristy
  Non-Presenting Author

• Connie Barroso, Texas A&M University - College Station
  Non-Presenting Author

• Rachel A. Conlon, Florida State University
  Non-Presenting Author

• Jamie M. Dasher, Florida State University
  Non-Presenting Author