Dataset and Analysis Procedure for Job-Work Fit as a Determinant of the Acceptance of Large-Scale Agile Methodology

Published: 06-02-2020| Version 2 | DOI: 10.17632/227b5x4vg6.2
Dinesh Batra


The data analysis requires the creation of a structural model, which can be analyzed using SmartPLS 3 as follows: Subjective Norm -> Job_Work Fit, External Support -> Job_Work Fit, Subjective Norm -> Methodology Acceptance, External Support -> Methodology Acceptance, Job_Work Fit -> Methodology Acceptance, Mandatoriness -> Methodology Acceptance, Exp Agile (single item measure) -> Methodology Acceptance, Org Size (single item measure) -> Methodology Acceptance. Job_Work Fit is a second-order reflective construct composed of Perceived Usefulness and Compatibility, and thus, you should have Job_Work Fit -> Perceived Usefulness and Job_Work Fit -> Compatibility. The indicators of Job_Work fit will be the union of those of Perceived Usefulness and Compatibility. Methodology Acceptance (Agile_Val_P). Agile values and principles (Sys_Think). Systems thinking (Lean). Lean product development (Transparency). Transparency (Cont_Improve). Continual improvement (Decent_Decision). Decentralized decision making The constructs subjective norm, perceived usefulness, mandatoriness, external support, and compatibility were measured using the following Likert scale: Strongly disagree (1) Disagree (2) Somewhat disagree (3) Neither agree nor disagree (4) Somewhat agree (5) Agree (6) Strongly agree (7) Subjective Norm To what extent have the following influenced your acceptance of the large-scale agile methodology (LSAM) <method name from a previous choice>? (Peer). Peer groups (Supervisor). Supervisors (Trade_Pub). Trade Publications (Prof_Org). Professional Organizations Perceived Usefulness As compared with alternative methodologies, I perceive that the adopted LSAM <method name from a previous choice> provides a distinct improvement in (Soft_Qual). Software quality (Deliv_Time). Delivery Time (Lower_Cust). Lowering costs (Cust_Satisf). Customer satisfaction (Team_Satisf). Team satisfaction Mandatoriness Regarding the adopted large-scale agile methodology (LSAM) <method name from a previous choice> mentioned by you: (Compulsory). Using the LSAM is compulsory in my job (Required). Although I might prefer another methodology, I am required to use the LSAM (Explicit_Pol). The company has an explicit policy for using the LSAM (Strong_Pref). The company has a strong preference for standardizing development using the LSAM External Support For helping me to better understand and use the adopted LSAM <method name from a previous choice>, the following resources are widely available: (Spec_Inst). Specialized instruction (Coaching). Coaching resource (Net_Support). Network of support (Help_Res). Help and resources Compatibility I feel that using the adopted LSAM <method name from a previous choice> is more compatible than other methodologies for (Way_Dev). The way I develop systems (Handle_Large). Handling large projects (Recon_Stake). Reconciling different views of stakeholders (Resp_Change). Responding to changes in requirements


Steps to reproduce

You will need to make two runs; the second involves the use of latent values of Perceived Usefulness and Compatibility as indicators of Job_Work Fit. See Hair, J. F., Hult, G. T. M., Ringle, C., & Sarstedt, M. (2016). A primer on partial least squares structural equation modeling (PLS-SEM). Thousand Oaks, CA: Sage Publications for more details. YouTube has excellent videos on how to conduct analysis when second-order constructs are involved. 1. In the first step, create a structural model in SmartPLS 3 as indicated above. Import the CSV data (you may want to remove headings that are not required to run the data). Ignore the Perceived Usefulness and Compatibility latent values (the last two columns in the datasheet) at this stage. If the first step is done correctly, your latent values should match the ones shown in the datasheet.You can obtain the latent values after running the PLS algorithm. For obtaining p-values, you should use bootstrapping. 2. In the second step, recreate the model without the Perceived Usefulness and Compatibility constructs; instead, use the latent values of these lower constructs as indicators of Job_Work Fit. Run the PLS algorithm and bootstrapping. Experience and Organization size are single-item measures. The logarithm of Organization Size is considered because the distribution is highly skewed. Other demographic variables are not considered in the analysis.