Presentation is loading. Please wait.

Presentation is loading. Please wait.

Data compilation and pre-validation

Similar presentations


Presentation on theme: "Data compilation and pre-validation"— Presentation transcript:

1 Data compilation and pre-validation
Mihaela bratu Razvan pavel National institute of statistics romania

2 Summary Data compilation Data validation EDIT Validation Tools

3 Data compilation City statistics
ONA’s Databases NIS STO Data Collection Data estimations City statistics Data taken from different data sources NIS + ONA’s (Other National Authorities) Data estimation NIS + STO (Statistical Territorial Offices)

4 Data compilation Category A Category B Category C
An analysis of the requested variables is performed Each variable is classified into 3 categories For the available data or estimated data we also gather information's regarding the data sources and whether the definition is according to the Eurostat manual data available (NIS, ONA’s and/or STO) Category A data can be estimated Category B data not available Category C

5 Data compilation Verification file for each variable E-damis data file
The data is available in different formats, depending on the unit that produces the data The data format available at NIS is different from the data format requested by Eurostat The data compiled in different formats is imported into the internal Urban Audit Database The imported data is checked for errors and inconsistencies The data from the Urban Audit Database is exported into the Eurostat requested model The programmes generate 3 files for each domain File no.1 – worksheets for each variable – excel file format File no.2 – one worksheet for all variables on one domain – excel file format File no.3 – one worksheet for all variables on one domain – csv file format File no.1 Verification file for each variable File no.2 E-damis data file File no.3 EDIT data file

6 Data pre-validation Pre-validation process is done in 2 stages
The data is checked for errors and inconsistencies using internal validation software EDIT Validation Tool

7 Hierarchical validation
EDIT Validation Tool Record validation Vertical validation Hierarchical validation EDIT = editing system developed by Eurostat EDIT = allows users to import data, perform a set of predefined operations on the imported datasets and export data resulted from these processing operations. EDIT = Validations EDIT = Dataset Operations Access to EDIT via: EDITT/Information/index.html

8 EDIT Validation Tool EDIT = Web-based User Interface

9 EDIT Validation Tool EDIT = Web-based User Interface

10 EDIT Validation Tool EDIT = Web-based User Interface

11 EDIT Validation Tool EDIT = Web-based User Interface

12 EDIT Validation Tool EDIT = Web-based User Interface

13 EDIT Validation Tool EDIT = Web-based User Interface

14 EDIT Validation Tool EDIT = Web-based User Interface
How does an error file looks ? .CSV file

15 EDIT Validation Tool EDIT = Web-based User Interface
IS IT DIFFICULT TO USE ? TO USE = NO, IT IS NOT DIFFICULT TO USE! TO IDENTIFY THE ERRORS = THE SHORT ANSWER: “IT DEPENDS”!

16 Thank you!


Download ppt "Data compilation and pre-validation"

Similar presentations


Ads by Google