Abstract
A combination of WeChat applet and web-based support system is designed and utilized in order to facilitate a massive questionnaire survey for Climate Index of small and medium-sized enterprises. This survey system could assist both in investigation stage and analysis stage. The WeChat mini program enables investigators to retrieve interviewed enterprise information and upload questionnaire answers automatically in a single online applet. The web-based support system provides reinforcement in extreme data exclusion, index calculation, statistical classification as well as result visualization. The whole system has been utilized in Climate Index of SMEs survey 2020 and verified its efficiency and reliability.
Introduction
Since 2014, the Enterprises Ecological Research Center (EERC) has tracked the "Climate Index" of Small and Medium Enterprises (SMEs) in China’s Jiangsu province to monitor economic health. Traditionally, scaling this survey to over 3,000 companies was a logistical challenge. The workflow relied on paper questionnaires and face-to-face interviews, leading to several critical bottlenecks: manual entry errors, inability to verify company details in real-time, and a significant time lag between data collection and index publication.
To modernize this process, researchers developed a two-pronged IT infrastructure designed to digitalize the entire survey lifecycle, replacing manual paper-and-Excel workflows with an automated, reliable system.
System Architecture and Methodology
The proposed system utilizes a dual-platform approach to separate data collection from heavy-duty analysis.
- The WeChat Applet (Front-end Collection): Built on a Client/Server (C/S) architecture using JavaScript and an MVC framework, this tool is designed for investigators in the field. It integrates third-party APIs to retrieve official enterprise information instantly, ensuring that data like revenue or assets matches authorized records at the point of entry.
- The Web-based Support System (Back-end Analysis): This system uses a Browser/Server (B/S) architecture with Vue.js and ECharts. It handles automated index calculation and data preprocessing.
A key innovation is the automated preprocessing mechanism. The system identifies "unreliable" questionnaires—such as those with contradictory answers or extreme outliers—and excludes them before calculation. This shifts the workload from manual Excel filtering to automated database queries.
Implementation and Data Collection
The WeChat applet serves as the primary interface for data entry. By implementing "automatic discrimination" within the applet, the system prevents investigators from submitting incomplete or inconsistent forms. The data is "locked" only after passing basic logic checks, which drastically improves the cleanliness of the raw data.
While the survey relies on subjective executive opinions (scored 1–5), the system manages statistical deviation by assigning neutral values to vacant answers and using accumulated scores to filter out extreme sentiment outliers. This allows for near-real-time index updates; as soon as a batch of questionnaires is uploaded, the index can be recalculated and visualized.
Results and Impact
The system was benchmarked using the 2020 Climate Index Survey, involving over 2,000 valid questionnaires. The results demonstrated significant improvements over traditional methods:
- Reliability: The 2020 survey achieved a Cronbach’s alpha of 0.937, the highest internal consistency in the survey's history.
- Efficiency: The period required for raw information inspection and index calculation was reduced from weeks to just a few hours.
- Visualization: The integration of ECharts allowed for immediate visualization of economic trends. For instance, radar maps clearly depicted the 2020 economic shock caused by COVID-19, showing resilience in policy and legal environments despite declines in production and finance.
Conclusion
The implementation of a WeChat-based survey system has revolutionized the collection of the SME Climate Index. By validating data at the source and automating complex calculations, the system ensures both high data reliability and rapid turnaround. Future work will focus on reducing platform dependency and integrating objective financial data to further validate subjective executive assessments.
Summary updated