Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Using Causal Inference to Explore Government Policy Impact on Computer Usage

Mingjia Zhu, Lechuan Wang, Julien Sebot, Bijan Arbab, Babak Salimi, Alexander Cloninger

TL;DR

The paper addresses how COVID-19 workplace closing policies causally affected personal computer usage by merging the Oxford COVID-19 policy tracker with Intel DCA telemetry. It employs Difference-in-Differences, Synthetic Control, and offline breakpoint detection to quantify usage and power changes and to assess heterogeneity across device types and continents. Key findings show that policy activation increased usage by roughly 1.2–2.3 hours, with the largest effects on 2-in-1 devices and high-end CPUs, while deactivation effects are weaker and less consistent; change-point analysis links shifts in intensity and user types to lockdown events. The work demonstrates how combining policy trackers with large-scale device telemetry enables robust causal evaluation of policy effects on everyday technology use, with practical implications for hardware provisioning and policy assessment during disruptive events.

Abstract

We explore the causal relationship between COVID-19 lockdown policies and changes in personal computer usage. In particular, we examine how lockdown policies affected average daily computer usage, as well as how it affected usage patterns of different groups of users. This is done through a merging of the Oxford Policy public data set, which describes the timeline of implementation of COVID policies across the world, and a collection of Intel's Data Collection and Analytics (DCA) telemetry data, which includes millions of computer usage records and updates daily. Through difference-in-difference, synthetic control, and change-point detection algorithms, we identify causal links between the increase in intensity (watts) and time (hours) of computer usage and the implementation of work from home policy. We also show an interesting trend in the individual's computer usage affected by the policy. We also conclude that computer usage behaviors are much less predictable during reduction in COVID lockdown policies than during increases in COVID lockdown policies.

Using Causal Inference to Explore Government Policy Impact on Computer Usage

TL;DR

The paper addresses how COVID-19 workplace closing policies causally affected personal computer usage by merging the Oxford COVID-19 policy tracker with Intel DCA telemetry. It employs Difference-in-Differences, Synthetic Control, and offline breakpoint detection to quantify usage and power changes and to assess heterogeneity across device types and continents. Key findings show that policy activation increased usage by roughly 1.2–2.3 hours, with the largest effects on 2-in-1 devices and high-end CPUs, while deactivation effects are weaker and less consistent; change-point analysis links shifts in intensity and user types to lockdown events. The work demonstrates how combining policy trackers with large-scale device telemetry enables robust causal evaluation of policy effects on everyday technology use, with practical implications for hardware provisioning and policy assessment during disruptive events.

Abstract

We explore the causal relationship between COVID-19 lockdown policies and changes in personal computer usage. In particular, we examine how lockdown policies affected average daily computer usage, as well as how it affected usage patterns of different groups of users. This is done through a merging of the Oxford Policy public data set, which describes the timeline of implementation of COVID policies across the world, and a collection of Intel's Data Collection and Analytics (DCA) telemetry data, which includes millions of computer usage records and updates daily. Through difference-in-difference, synthetic control, and change-point detection algorithms, we identify causal links between the increase in intensity (watts) and time (hours) of computer usage and the implementation of work from home policy. We also show an interesting trend in the individual's computer usage affected by the policy. We also conclude that computer usage behaviors are much less predictable during reduction in COVID lockdown policies than during increases in COVID lockdown policies.

Paper Structure

This paper contains 22 sections, 4 equations, 9 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Method
  3. Datasets
  4. Oxford Covid-19 Government Response Tracker
  5. Intel DCA Dataset
  6. Methods
  7. Difference-in-differences
  8. Synthetic Control Method
  9. Offline Breakpoint Detection Methods
  10. Results
  11. Activation of workplace closing policy on computer usage time
  12. Difference in differences, without confounders
  13. Continent information affects the baseline computer usage
  14. Effects of the Workplace Closing Policy on Different CPUs and Computer Types
  15. Final Results with Vpro Percentage as a Confounder
  16. ...and 7 more sections

Figures (9)

  • Figure 1: Panel OLS analysis illustrating the effect of workplace closing policy on average computer usage. The estimated policy effect is positive, indicating an increase in computer usage post-policy implementation. The near-zero p-value suggests statistical significance.
  • Figure 2: Estimation of policy effect weight (2.3092), p-value, and confidence interval.
  • Figure 3: Comparison of average computer usage across continents and the impact of workplace closing policy.
  • Figure 4: Comparison of Policy Effects Across Various Computer Types
  • Figure 5: Comparison of Deactivation Effects Across Various Computer Types
  • ...and 4 more figures