1. Classification of Statistical Data
Lodging-related statistical data can be classified into complete enumeration data and sample survey data, depending on the scope of aggregation.
In the case of sample surveys, they generally reflect the situation of the sample relatively accurately, but often lack representativeness for the entire group. Nevertheless, due to their efficiency in terms of time and cost, they are widely used.
Complete enumeration data, on the other hand, overcomes the limitations of sample surveys in terms of representativeness, but due to the massive cost and time required, the coverage and frequency of aggregation are inevitably limited.
Lodging-related data can also be classified based on the collection method into survey-based data and record-based data.
Survey-based data, where respondents choose whether to participate, allows for broad data collection and usage, but often includes incorrect or intentionally misleading responses, leading to inevitable data contamination.
In contrast, record-based data is less prone to such contamination, but the availability and usability of the data can be limited depending on the type of records.
2. Case Studies: Public Statistics
3. Lobin's Data Processing Method
Lobin Co. does not generate statistics itself, but rather collects, reconciles, and compiles existing statistical data. The process Lobin Co. uses to collect, reconcile, and compile this data can be summarized in three major steps:
- Define the sequence and correlations for each data item. The sequencing and correlation algorithms are based on formulas accepted in the global lodging industry.
- Collect relevant statistical data for each item. Depending on the item, data is collected at the market level or the individual establishment level, and this data is fed into the sequence and correlation algorithms.
- Reconcile and compile the data by repeatedly cross-validating results from adjacent sequences. This cross-validation loop stops once the difference between iterations falls within a target margin of error.
Through this process, Lobin Co. captures the unique attributes of various statistical datasets while removing individual sources of contamination, thereby ensuring the practical usability of the data and provising a complete visibility into the lodging market of Korea.
