Data illustrate correlations between results from individual participants in cross-over experiment usb2 conducted by Scanniello et al: [1] G. Scanniello, A. Marcus, and D. Pascale, 'Link analysis algorithms for static concept location: an empirical assessment', Empirical Software Engineering, vol. 20, no. 6, pp. 1666–1720, 2015. The goal of the experiment is to assess whether a new technique (implemented as an Eclipse plug-in) for static concept location (proposed by the authors) supports users in identifying the places in the code where changes are to be made.
KitchenhamEtAl.CorrelationsAmongParticipants.Scanniello15EMSE`KitchenhamEtAl.CorrelationsAmongParticipants.Scanniello15EMSE`: a data frame with 48 rows and 10 variables:
<fct>|ExperimentID: A unique identifier for each experiment in the data set.
<fct>|Participant ID: An identifier for each participant, unique for a specific experiment.
<fct>|Experimental Sequence Group: A (CL-NOCL,Jedit-Atunes), B (NOCL-CL,Atunes-Jedit), C(NOCL-CL,Jedit-Atunes), D(CL-NOCL,Atunes-Jedit)
<fct>|Software systems used in the experiment: Jedit and Atunes
<fct>|Experimental Treatment: Use of Concept Location plug-in (CL) vs no Concept Location plug-in (NOCL)
<fct>|Time period of the cross-over experiment: 1 or 2
<int>|Dependent variable: 0, 1, 2, 3, 4. The participants are asked to indicate a single change method for each of 4 bug reports. A change method is correctly identified if that method is in the change set of the bug report.
<dbl>|Dependent variable: The total time [min] to accomplish concept location tasks, i.e.,to identify (four) bugs given their reports
<dbl>|Dependent variable: The participants’ efficiency in the execution of concept location tasks. It is computed dividing correctness by time.
<fct>|Crossover category: For 4 group crossover designs, the crossover category specifies the matching pairs of sequence groups, CO1 and CO2.
If you use this data set please cite: [1] G. Scanniello, A. Marcus, and D. Pascale, 'Link analysis algorithms for static concept location: an empirical assessment', Empirical Software Engineering, vol. 20, no. 6, pp. 1666–1720, 2015. [2] Barbara Kitchenham, Lech Madeyski, Giuseppe Scanniello and Carmine Gravino, 'The importance of the Correlation between Results from Individual Participants in Crossover Experiments' (to be submitted as of 2020).
KitchenhamEtAl.CorrelationsAmongParticipants.Scanniello15EMSE
Run the code above in your browser using DataLab