Network analysis has become increasingly popular among psychologists. Recognizing that psychological research requires substantive theory has prompted a critical development from descriptively mapping network structures towards testing specific hypotheses. While confirmatory approaches are becoming more accessible to applied researchers, the broader set of available statistical tools for this purpose has not yet been fully utilized by network psychometrics. The present work aims to reduce this gap in two steps: First, we review existing approaches for confirmatory network testing, spanning from classical SEM goodness-of-fit indices to covariance- and resampling-based tests. Next, in a simulation study, we evaluate how effective they are in determining whether a theoretical network aligns with empirical data given varying degrees of network (dis)similarity, number of variables (p), sample size (N), and network estimation methods. Performance is assessed regarding Type I error rates and power, thus providing insights into their applicability and robustness across various data scenarios.