Assignment Task
Questions
- Open-source projects are software artifacts, which are developed and maintained by software developers and volunteers. Generally, the source code of these projects is available online and end-users can freely use them under the constraints defined on project license type. GitHub is the largest open-source repository with millions of projects. You have been provided with a panel dataset of projects hosted on GitHub. It contains two years (8 quarters) of data for each project. You can investigate how projects change over time by analyzing this dataset.
- Visualize the growth plots for any random 12 projects. Does Watchers increase over time? Describe the change in Watchers over time based on growth plots.
- Generate the unconditional means and unconditional growth models and write down the corresponding equations. Discuss the fixed effects of the mode to predict the rate of the change in a number of watchers over time.
- Whether the intercept (initial number of watchers) and slope (change over time) affected by owner type? Interpret the estimates of fixed effects, variance components and Pseudo R2 statistics.
- How do issue status and owner types influence the number of watchers? Examine the effect on initial status and rate of change.
- Discuss the best model to explain the change in number of watchers over time.
- Visualize the growth plots for first ten students. Does math achievement (test scores) increase over
- time? If yes, at what rate?
- Using visualization technique, compare the trend of changes in test results across two categories of effective variable over time. Discuss your finding about the observed pattern.
- What is your “intercept only” model? Interpret the fixed effect; variance components (within-person variance and between-person variance); and ICC (Intra-class Correlation Coefficient)?
- Discuss the model to predict student’s math scores from the intercept and time. Interpret the rate of change?
- Whether the intercept (initial math score) and slope (change over time) are affected by effective teachers? Interpret the estimates of fixed effects; variance components and Pseudo R2 statistics?
- How do socioeconomic status (characteristics – level 2) and effective teachers (characteristics – level 2) influence student’s math scores?
- Consider the following case of level-2 sub-models: Initial status is affected by both predictors- socioeconomic status and effective teacher, and slope (rate of change) is affected only by the student’s socioeconomic status. Interpret the estimates of fixed effects; variance components.
- Is there a change in math achievement over time? What model is the most appropriate to use at the end?