Replication of Software Engineering Experiments - Empirical Software Engineering and Verification - page 86

A.15 Monroe [74]

Simultaneous Replication. Does the same researcher in the same study investigate
consumer reactions to more than one product, or to more than one advertise-
ment?

Sequential Replication. Does the researcher or another researcher repeat the
study using the same or different stimuli at another point in time?

Nonindependent Replication. The replication is conducted by the same researcher.

Independent Replication. The replication is conducted by different researcher.

Assumed Replication. For example, a researcher using both males and females
simultaneously in a study and finding no gender covariate effect assumes repli-
cation across gender.

Demonstrated Replication. What is preferable is separate gender conditions
wherein the effect has or has not been obtained separately for males an females,
that is, demonstrated.

Strict Replication. The replication is a faithful reproduction of the original study.

Partial Replication. The replication is a faithful reproduction of some aspects of
the original study.

Conceptual Replication. The replication uses a similar conceptual structure but
incorporates changes in procedures and independent variables.

A.16 Radder [75]

Reproducibility of the material realization of an experiment. In this type of repro-
duction, the replicator correctly performs all the experimental actions following
instructions given by the experimenter who ran the previous experiment. This
reproduction is based on a division of labour, where other previously instructed
people can run the replication without being acquainted with the theory under-
lying the experiment. As in this reproduction it is possible to follow the same
procedure to verify the outcome without detailed knowledge of the theory, there
may be differences in the theoretical interpretations of the experiment.

Reproducibility of an experiment under a fixed theoretical interpretation. This
reproduction implies that the conditions of the previous experiment can be in-
tentionally altered in the replications, provided that the variations are irrelevant
to the theoretical interpretation of the experiment.

Reproducibility of the results of an experiment. This type of reproduction refers
to when it is possible to achieve the same result as a previous experiment using
different methods. This category excludes a reproduction of the same material
operationalization.