Skip to main content icon/video/no-internet
icon backReturn to Entries

Replication

Replication is conducting a study in another case (or population) to assess whether a research finding from previous studies can be confirmed. The aim of replication is to assess the generalizability of a theoretical claim and establish that the research finding that is (or is not) confirmed in the replication study is derived from the previous studies.

Sometimes the term replication is also used to refer to a study conducted in the same case (or population) to assess whether the finding from the previous study can be reproduced. To avoid confusion, the latter method of assessing the reliability of a study might better be called duplication. Duplication is not discussed in this entry.

Conceptual Overview and Discussion

The Concept of Replication

David Hume formulated the general problem of induction, stating that the fact that a claim has passed one test does not provide any evidence that it will pass other tests. This implies that one cannot know for sure that a statement about a class of entities is true if that statement has not been tested in every instance of that class. Using a well-known example of such a claim, “All swans are white,” Hume's problem implies that we know this for sure only after we have observed the color of every single swan. Because this is usually not feasible, Hume concluded that induction is not possible. Karl Popper's conclusion was that such confirmatory certainty should not be sought and that instead one should attempt to achieve certainty about the incorrectness of the statement. A single counterexample (a non-white swan) would be sufficient for achieving this. Replication is the search for confirmations (Hume) or disconfirmations (Popper) of a claim about a domain of, in principle, an indefinite number of entities to which the claim applies.

Replication is also the core procedure by which credible theoretical knowledge is generated. It is not a one-off activity but a strategy of subsequent tests in which the likelihood of the correctness of a claim for a domain increases with each failure to find a disconfirmation. The more different an entity (or a population of such entities) is from the ones in which earlier tests were conducted, the larger the contribution of that test to the confidence in the correctness of the claim. Replication is a general procedure that applies to all sciences and all research strategies. Its relevance and applicability are not confined to only some types of tests, such as case studies or experiments.

The Yin-Eisenhardt Approach to Replication

Robert Yin stated that replication is the attempt to find support for a proposition in a number of single case studies, arranged effectively within a multiple-case design. He emphasized the analogy between such a series of single case studies and multiple experiments but did not mention the general principle of replication that applies to all other research strategies (including the survey) as well. Yin implicitly assumes that replication applies only to propositions about characteristics of single cases (see the discussion of types of proposition in the Theory-Testing With Cases entry, this volume). Kathleen Eisenhardt, who proposed Yin's replication logic as the core procedure of her approach to theory-building, made this assumption explicit by stating that, in her approach to theory-building, each hypothesis is examined for each case, not for the aggregate cases (see the Theory-Testing With Cases entry, this volume).

...

  • Loading...
locked icon

Sign in to access this content

Get a 30 day FREE TRIAL

  • Watch videos from a variety of sources bringing classroom topics to life
  • Read modern, diverse business cases
  • Explore hundreds of books and reference titles

Read next

More like this

Sage Recommends

We found other relevant content for you on other Sage platforms.

close

Have you created a personal profile? Login or create a profile so that you can save clips, playlists and searches

close
close
close Signed In to profile You are signed in as:
close
Logged in Institution Institution