Abstract
During the progressing renaissance of psychological research, various suggestions have been proposed concerning how to improve the current situation. Among them, a prominent place belongs to replication studies.
However the authors aiming to critically evaluate conducted replications or to conduct a replication study themselves needs to explicitly reflect two crucial issues. The first regards conditions that need to be met in order to consider the second study as a replication of the first.
This concerns the definition of replication and broader categorization of replication types from a methodological point of view, reflecting various conceptual and terminological issues. The second regards conditions that need to be met in order to say that the results of the replication study replicated/did not replicate the original study.
This involves the way in which the study is statistically analysed with respect to its main goal. Therefore, the aim of the present article is to discuss and to summarize the present state of the literature regarding these two issues and to synthesize it into two functional classifications: A, various types of replications from a methodological point of view; B, various possibilities of data analysis from a statistical point of view.
It is argued that instead of non-critically preferring one specific approach, the researcher should reflect the pros and cons of various approaches and the broader context of the goal of the replication study.