AN IMPROVED INFORMATION MODEL FOR KNOWLEDGE-BASED WORK : How data is treated in process and what is quality of results?
Nieminen, Kristiina (2018)
Nieminen, Kristiina
Metropolia Ammattikorkeakoulu
2018
All rights reserved
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2018052510280
https://urn.fi/URN:NBN:fi:amk-2018052510280
Tiivistelmä
Since the 18th century, statistics of the Finnish society have been compiled in Scandinavia. In the beginning, compiled statistics covered information about the population but over the decades new subject areas such as economy, living conditions, environment and business sectors have been included into statistics production.
Statistics are compiled to support decision-making, and therefore they must meet the set quality criteria. In order to ensure this obligation, statistical producers need descriptive information, the so-called metadata, of statistical data and the compilation process. The aim of defining and using metadata is to enable the users and statisticians to identify and to understand the key qualitative factors related to data and results.
The purpose of this Master’s thesis was to create an information model that may be used to store detailed information about the process and the data. The first task was to analyze the current practices and information needs, the production guidelines of general statistics and the available information models. Then the analysis results were used in the design of an improved information model. The new information model complements the existing common metadata system in Statistics Finland, and creates a completely new information content for statistical process related information.
The research method was an inductive, data-based analysis that was based on a survey questionnaire about the current status and information needs. The respondents were randomly selected from Statistics Finland's statistical units. Based on the responses, a list of the current production system’s weaknesses was created. In addition to this, a literature review covering international development work, research reports and generic information models was compiled. The theoretical framework of this thesis was founded on ethical principles, legislation, working instructions composed by international actors and EU quality requirements.
An information model design was accomplished from the point of view of process related information. The designed final information model is a process-based model, i.e. it brings together the key elements of statistics production: the process and its phases and steps, statistical data, statistical methods, rules and actors.
As a results the designed information model is simple and general in nature, so it may be applied in other national statistical institutions as well. The structure of the final information model is hierarchical, whereby dependencies between the different elements are identified, described and clearly understood.
Statistics are compiled to support decision-making, and therefore they must meet the set quality criteria. In order to ensure this obligation, statistical producers need descriptive information, the so-called metadata, of statistical data and the compilation process. The aim of defining and using metadata is to enable the users and statisticians to identify and to understand the key qualitative factors related to data and results.
The purpose of this Master’s thesis was to create an information model that may be used to store detailed information about the process and the data. The first task was to analyze the current practices and information needs, the production guidelines of general statistics and the available information models. Then the analysis results were used in the design of an improved information model. The new information model complements the existing common metadata system in Statistics Finland, and creates a completely new information content for statistical process related information.
The research method was an inductive, data-based analysis that was based on a survey questionnaire about the current status and information needs. The respondents were randomly selected from Statistics Finland's statistical units. Based on the responses, a list of the current production system’s weaknesses was created. In addition to this, a literature review covering international development work, research reports and generic information models was compiled. The theoretical framework of this thesis was founded on ethical principles, legislation, working instructions composed by international actors and EU quality requirements.
An information model design was accomplished from the point of view of process related information. The designed final information model is a process-based model, i.e. it brings together the key elements of statistics production: the process and its phases and steps, statistical data, statistical methods, rules and actors.
As a results the designed information model is simple and general in nature, so it may be applied in other national statistical institutions as well. The structure of the final information model is hierarchical, whereby dependencies between the different elements are identified, described and clearly understood.