For the mechanic, the “hammer and spanner” are essential tools. They enable the mechanic to complete his tasks. Today, software tools are key for us – students – to meet our academic requirements. They also empower us beyond our student life too. They help us achieve proficiency in any given task.

Open source vs Proprietary Software

R, Python and Jamovi come under open source software. Users can customize, contribute, and access the code and logic behind the software. Unlike proprietary products such as SPSS, SAS STATA, and Eview, open source is cost-free and the access is for the lifetime of the software, especially the desktop version. Proprietary software can be accessed free of cost only for a trial period ranging between 14 to 365 days. R and Python are ‘double-barrel guns’ which democratize the data analysis field by providing everything cost free. Open source software has disrupted existing players in the market (Matt Asay 2016). The critical advantage of open source is that they enable various data analyses under one umbrella.

The proprietary products business model demands that the customer procure each separate analysis module; the customer has to shell out money every time. To amplify this, cross-section data analysis, time-series data analysis, textual analytics, and image analytics are all available under the umbrella of R and Python. Still, in proprietary modes, different products have to be purchased. My suggestion for the cross-section data analysis starter is to begin with ‘Jamovi’ developed by the Amsterdam university. They have taken the pain of writing code and providing the gain of getting all output merely as ‘drag and drop’ (Muenchen. A 2018). However, for a marathon in the data analysis field, it is better to learn to use R and Python.

Data Visualization

Many top executives are encouraged to use “Tableau” software because of its user-friendly and quick time to turn into a dashboard (Hale 2019). A focused 10-hours is sufficient to create a decent dashboard in Tableau. The same job with MS-Excel may take at least 25 hours. R & Python may take more than 50 hours with the support of R Shiny and Python Dash, respectively. Besides Tableau, Power BI and Zoho Analytics are doing an incredible job in data visualization.

Econometrics: Time series and panel-data analysis

In public policy and economics, most of the data is time-based. Econometrics is the obvious choice. R and Python can do an excellent job. Still, nothing can match “STATA” from the proprietary stable. Many international organizations like WTO use STATA extensively , especially for time series based analysis (World Bank 2020)

Qualitative data analysis

Qualitative data such as text, images, and audio transcribes through R and Python can do well. However, the user needs command over niche areas like natural language processes and AI-based algorithms. To avoid this, users can embrace ‘Nvivo’, which is quite user-friendly, but at a considerable cost for individual users. The kind of insight provided by Nvivo from various sources such as websites, Twitter, and various pdf documents is unimaginable (Welsh 2002).

Following a brief exploration of all software that can support public policymakers to leverage their data analysis skills, I vote for R, if it is academic or R&D projects. For AI-based real-time projects Python has an edge over all other software including R (R and Python, 2020). Now the choice is ours. In the era of big data, accessing and using data-driven policy provides a competitive advantage (Mählmann et al. 2017). Policymakers, both from corporate and public institutions, realize that data-driven decision-making (DDD) significantly affects the organization’s performance (Brynjolfsson and McElheran 2016). Hence, to better exploit the data economy, let us embrace new tools a.k.a. software to support our decision-making.

References:

  1. Brynjolfsson, Erik, and Kristina McElheran. 2016. “The Rapid Adoption of Data-Driven Decision-Making.” The American Economic Review 106 (5): 133–39.
    https://www.jstor.org/stable/43861002.
  2. DuBois, Jen. 2020. “The Role of Data Analysts in 2020 and Beyond.” QuantHub (blog). 2020.
    https://quanthub.com/data-analysts/.
  3. Hale, Jeff. 2019. “Tableau Basics in Six Minutes.” Medium. December 13, 2019.
    https://towardsdatascience.com/tableau-basics-in-six-minutes-35b50ec9ae9b.
  4. Matt Asay. 2016. “Exponential Growth of R’s Open Source Community Threatens Commercial Competitors.” TechRepublic. 2016.
    https://www.techrepublic.com/article/exponential-growth-of-rs-open-source-community-threatens-commercial-competitors/.
  5. Muenchen. A, Robert. 2018. “Review of the Jamovi GUI for R | R4stats.Com.” 2018.
    http://r4stats.com/articles/software-reviews/jamovi/.

Leave a Reply

Your email address will not be published.