A Theory of Scrum Team Effectiveness: Dataset and Supplementary Materials

This bundle contains supplementary materials for an upcoming academic publication A Theory of Scrum Team Effectiveness, by Christiaan Verwijs and Daniel Russo. Included in the bundle are the dataset, SPSS syntaxes, and model definitions (AMOS). This replication package is made available by C. Verwijs under a "Creative Commons Attribution Non-Commercial Share-Alike 4.0 International"-license (CC-BY-NC-SA 4.0).

About the dataset

The dataset (SPSS) contains anonymized response data from 4.940 respondents from 1.978 Scrum Teams that participated from the https://scrumteamsurvey.org. Data was gathered between June 3, 2020, and October 13, 2021. We cleaned the individual response data from careless responses and removed all data that could potentially identify teams, individuals, or their parent organizations. Because we wanted to analyze our measures at the team level, we calculated a team level mean for each item in the survey. Such aggregation is only justified when at least 10% of the variance exists at the team level (Hair, 2019), which was the case (ICC = 51%). Because the percentage of missing data was modest, and to prevent list-wise deletion of cases and lose information, we performed EM maximum likelihood imputation in SPSS.

The dataset contains question labels and answer option definitions. To conform to the privacy statement of scrumteamsurvey.org, the bundle does not include individual response data from before the team-level aggregation.

About the model definitions

The bundle includes definitions for Structural Equation Models (SEM) for AMOS. We added the four iterations of the measurement model, four models used to perform a test for common method bias, the final path model, and the model used for mediation testing. Mediation testing was performed with the procedure outlined by Podsakoff (2003). Mediation testing was performed with the "Indirect Effects" plugin for AMOS by James Gaskin.

About the SPSS syntaxes

The bundle includes the syntaxes we used to prepare the dataset from the raw import, as well as the syntax we used to generate descriptives. This is mostly there for other researchers to verify our procedure.