Dissertation Title: Motivations in Open-Source Development: A Quantitative Study on Drupal in Academia Mentee: Shallon Elizabeth Brown, Ph.D. Committee Members: Dr. Apiwan Born Dr. Lawrence Ness Dr. Bernard Sharum, Chair Originally Presented: Monday, 5 Sept 2016 11AM Central Time 1 Introduction Open source development and the impact on the software industry The existence of open source communities and developer networks Quantitative, correlational study focuses on open source software development with specific emphasis on Drupal development for academic purposes 2 Statement of the Problem
Previous studies explored the connection Drupal had to academia and various uses within institutions Drupal continues to be one of the most widely used open source CMS tools for universities, particularly for libraries (Valenza, Boyer, & Curtis, 2014) We did not know specifics on what motivational factors may encourage Drupal open source community developers and how those factors may differ across other OSS enterprises (Darlack, 2013) Lack of information on specifically what motivates software developers who are contributing OSS for academic use (Darlack, 2013) 3 Methodology Research Design This study used a quantitative research method.
Extended Klandermans model (EKM) was the theory used to facilitate the analysis of how software engineers described their motivations for contributing to the open source community, as originally described by Hertel, Niedner, and Herrman (2003). Extends the traditional Klandermans theory of motivation analysis, which is rooted in sociological theory that describes participation in collective action as the product of balancing the expected cost and benefits of participation (Colazo & Fang, 2009). This cross-sectional study was a non-experimental design Study used an interval variable that represented the number of hours the participant had been cumulatively involved in Drupal open source community academic projects. Used forward stepwise multiple linear regression, was an exploratory correlational model, and used a simple random sampling method (Pernet, Wilcox, & Rousselet, 2013). 4 Methodology Data Collection The inclusion criteria were that participants must be over the age of 18 and have participated in at least one Drupal project or module for academic purposes.
Exclusion criteria included anyone under the age of 18 and participants who had not participated in a Drupal project, group, or module development for academic purposes. Only one participant chose not to accept the terms within the informed consent form, and was disqualified from participation. The target population was OSS software developers, with a sample frame specifically of Drupal developers involved in academic related Drupal projects. Minimal acceptable computed sample size - 126 Survio Cint Link was used for the consumer panel Data Collection Sample Response Results N % Originally recruited 446 100 Responses to recruitment request 330 74
Responses agreeing to informed consent 329 73.8 Total qualified participants after screening 144 32.3 Responses with incomplete/missing data 13 2.9 Total usable data sample 131 29.4 5
Methodology - Instrumentation Investigated four primary constructs of the EKM using a questionnaire as the measurement tool (collective motives, norm-oriented motives, reputation motives, and intrinsic motives), while noting the demographics reported for gender, age, experience, and intent to contribute in the future Instrument for this study was derived from the one developed and utilized by Nov, Arazy, and Anderson (2014) to test (confirm or refute) that EKM as a valid theory for assessing time invested in OSS based on the number of hours logged as a correlational model specifically for online open source citizen science projects. Original instrument was used in its entirety, and was previously validated by the authors Nov, Arazy, and Anderson (2014) 6 Data Analysis Normality testing showed fairly normal distribution after the hours variable (HOURS) was transformed (SQRT_HOURS) Both skewness (.714) and kurtosis (-.326) were observed to be within the acceptable range of -2.00 and 2.00, respectively Further proof of normality, homoscedasticity, and linearity were confirmed by viewing the scatterplot matrix of all IVs (COLALL, INTALL, NORALL, and REPALL) against the DV HOURS. Generally, matrices resemble an elliptical and do not closely hug the edges of the graphs. The Cronbachs alpha value of the =.93 was an acceptable value, given that any value exceeding a = .70 was acceptable, as noted by Field (2009).
(Field, 2009). There were no problems of multicollinearity within the variables for this study; tolerance levels should be above .20, and all variables for this study ranged from .21 and .31 Furthermore, VIF values ranged from 3.26 to 4.73, also noted as acceptable (Field 2009). Based on these results, the assumptions for using forward stepwise multiple linear regression were confirmed for this study 7 Results Forward Stepwise Multiple Linear Regression Analysis Model Summary R Statistics R R Square Change Statistics Adjusted R R Square Change .15
Model Square 1 .39 .15 .14 Predictors: (Constant), Reputation Motives Dependent Variable: Average hours per week contributed F Change 21.21 df1 1 df2 122 Sig. F Change .000 DurbinWatson 1.89 Descriptive Statistics - Demographics
N Minimum Maximum Mean Std. Deviation Variance Age 131 18 77 36.10 10.89 118.51
1.30 Intent to Contribute 131 2 14 10.53 3.17 10.07 Valid N (listwise) 131 8 Results - Outcome Results show that only reputation motive had a statistical correlation to the number of hours contributed to Drupal academic-related projects R squared = .15, Delta R squared = .15
Omnibus research hypothesis and research subquestions 3 are proven in favor of the alternative hypothesis. Null hypothesis accepted for research subquestions 1,2, and 4 9 Results Post Hoc Interested in using the demographic variable intent to contribute in the future in place of hours contributed presently against EKM The results of post-hoc testing show a positive correlation between intent to contribute and collective motives, intrinsic motives, and reputation motives Norm-oriented motives had no correlation to either past or future contribution All participants who contributed over 41 hours a week were male 10 Conclusions Results of this study show that the desire to heighten reputation is correlated positively to increased participation in Drupal academic related projects. As such, gaining an enhanced understanding as to how reputation can be tracked within OSS communities over time is of increased importance, especially for
academic related OSS projects More research into open source developer networks and the existing established mechanisms for tracking the reputation of participants 11 Recommendations Recommended that future research be done with varied instruments that help further explore more specific elements of reputation that contribute to the overall correlation Recommended that future research explore more CMSs and other open source tools in relation to academia utilizing the EKM Recommend to exploring software process control and developer contribution to open source projects that are a collaborative effort between academia and different industry sectors. 12 Questions? 13
The plate is found in children and adolescents; in adults, who have stopped growing, the plate is replaced by an epiphyseal line. This replacement is known as epiphyseal closure. 14 years old +/- 2 years is the age of growth...
Literary Analysis. An analysis explain ... Including the narrator of a story or the speaker of a poem. Setting: when and where the action happens. Point of View ... American flag = freedom. Dove = peace. What is the raven...
Voluntarios de villas capacitados en pocas sesiones aumentan resultados de niños de modo significativo. Cabezas et al. (en progreso): Servicio País en Educación en Chile. Estudiantes universitarios ~10 sesiones de apoyo en lectura a cuarto básico: mejoras significativos en algunos...
Dos and Don'ts for the existing firm . Do clarify member status in the LLP documents. Update and tailor RCs as closely as possible. Do include express duties of good faith, information, cooperation, and control over communications
to generate single or multiple prescriptions, sequences of FMAs applied to a stand over a pre-defined planning horizon. Falcão and Borges (2005) wrote that the key aspects of the architecture of a prescription simulator are database interaction, linkage to growth...
edge (L = number of edges) separating surface (of area A) sharp corner (C = number of corners) (method of derivation: spectral geometry) (Fursaev, hep-th/0602134) for ground state a is a cutoff C - topological term (first pointed out in...
Prehistoric Culture Vocabulary(page 27). cultivate - to prepare and use land for raising crops, marked the beginning of the New Stone Age. domesticate - to tame animals in order to make them useful to people
Modal dasar Perseroan paling sedikit Rp 50.000.000,00 (lima puluh juta rupiah). Modal yang ditempatkan (issued capital atau scribed capital) adalah yang telah diambil dan sebenarnya telah terjual kepada para pendiri maupun pemegang saham perseroan.
Ready to download the document? Go ahead and hit continue!