Biological invasions are an important aspect of anthropogenic global change and considered to be one of the major threats to biodiversity worldwide. Invasive species can change native species abundance, community structure, and ecosystem processes in invaded communities. However, due to context-dependencies it is often difficult to generalize the impact of invasive species. The outcome of an invasion is for example influenced by species traits, the invaded ecosystem, and invasion stage. Developmental stages and site-specific conditions can modify the performance of the invader and its interaction with other organisms. Another challenge is to disentangle cause and effect of an invasion. An alien species can cause changes in a previously intact ecosystem, thus be the “driver” of the changes. Alternatively, it can be a “passenger” which is facilitated by previous ecosystem changes. Context-dependencies as well as causality of invasions are important issues to understand and evaluate invasions, and to develop more targeted management plans. Thus, I investigated context-dependencies and causality of the impact of invasions using the model plant species Impatiens glandulifera. It heavily invaded several habitat types in Central Europe, but its impact on native plant communities is rated ambiguously, pointing towards context-dependencies. In an experimental study under controlled conditions I assessed the competitive and allelopathic effects of I. glandulifera on native co-occurring plant species. I found that seedlings and juvenile plants were negatively affected by a combination of allelopathy and competition. Native species differed in their susceptibility to I. glandulifera, and juveniles were more affected than seedlings. 2-metoxy-1,4-naphtoquinone (2-MNQ), the supposed main allelochemical, led to minor reductions in plant growth, suggesting that it may not be the only allelopathic substance of I. glandulifera. With two field studies I tested whether this species-specific response of native plants to allelopathy and competition of I. glandulifera leads to changed community patterns. I performed an observational vegetation survey within heterogeneous riparian meadows and alder forests in Germany. The vegetation was recorded in summer and spring because of seasonal species turnover and thus potentially different impact of I. glandulifera. The abundance of I. glandulifera as well as its impact on native vegetation depended on the environmental conditions at a particular patch. Plant species α-diversity was found to be not affected, but native plant cover was reduced specifically to species and season. To see whether the impact of I. glandulifera is causal and thus I. glandulifera a driver of ecosystem changes, I conducted a field experiment within the same study sites. Invaded and uninvaded plots were compared with plots from which I. glandulifera was removed and plots where I. glandulifera was planted. A negative impact of planting I. glandulifera and a concurrent positive effect of removal on native vegetation biomass and Urtica dioica performance indicated a causal but low effect of I. glandulifera. Species α-diversity was again not affected. I suggest that I. glandulifera is a “back-seat driver” of changes, which is facilitated by previous ecosystem changes but is also a driver of further changes. In summary, I found that the impact of I. glandulifera depended on the native target species, their developmental stage, on the habitat, and season. Dominant plant species, especially U. dioica were most affected, species α-diversity was not affected, species composition only under specific conditions. I conclude that the impact of I. glandulifera on native vegetation is only minor to moderate. However, from literature it is known that I. glandulifera has a negative impact on several additional ecosystem properties such as mycorrhiza, soil properties, and herbivores. The results of my studies are also relevant to develop management strategies for I. glandulifera. Generally, eradication measures are not feasible at large scales because of the wide distribution of I. glandulifera. They should primarily be applied to sites which are valuable in terms of nature conservation and to sites, where the strongest impact of I. glandulifera is to be expected. This is the case in habitats with high light conditions in combination with high soil moisture. Attention should also be paid to habitats with distinct spring communities, because they were also affected by I. glandulifera. The understanding that I. glandulifera is not a clear driver of ecosystem changes, but has some characteristics of a back-seat driver, indicates that removal is not sufficient for management of an I. glandulifera population. Additionally, habitat-restoration is required to counter those changes that benefited the invasion of I. glandulifera.