Question1: Importance of Data Quality
Data is essentially the basis of every scientific research. Therefore, the collection of good and high quality data actually plays a very huge and important role in the supply of objective information and possible solutions to the problems that are being studied. This is usually occurs across all fields of science. Decisions based on low quality data are usually very risk and may give rise to disastrous in occurrences. This may consequently result in the distortion of the particular study situation making all the subsequent analyses, studies and decisions to rest on shaky ground. It is therefore very important to ensure that collected is of high quality. Data quality ensures that the data is fit to serve the purpose for which it was intended to serve in the first place when it was being collected. It also ensures client satisfaction. Moreover data quality helps in building the credibility of the data collecting firm or institution.
Question 2: Parameters of data quality
Validity
This is used to establish whether the data is actually a measure of what the experiment is supposed to measure. Validity shows a defined and clear relationship between the project’s goals, project activities and what is being measured.
Reliability
This establishes whether the data collected can actually be managed using similar processes at all times. It establishes whether the data collection processes are documented and whether these processes can be replicated at another time to obtain similar results.
Integrity
This measures the ability of the data to be free from any form of manipulation, whether for political or for personal reasons. Data with integrity is one that does not exhibit any form of manipulation.
Precision
This measures the degree to which different data measurements for the same study agree with each other. Essentially, it measures how close different measurements come close to each other. It is sometimes referred to as repeatability or reproducibility of data.
Timeliness
This measures the level to which data is still important relevant or useful at the time when it is being reported.
Question 3: Satisfaction data quality parameters
There were several things that we did to ensure that data was of high quality.
In terms of validity, we ensured that the measurement units that we used were actually concordant with those of previous researchers who had conducted studies on the water quality aspect. This essentially added more strength to the validity of our research.
Question 4: Difference between quality assurance and quality control
Quality assurance is a process in scientific research that is oriented towards the prevention of defects occurrence in the processes used to make products or conduct research. It is proactive in nature. The goal of this process is actually to test processes and improve development to minimize or eliminate the occurrence of defects.
Quality control on the other hand is a process aimed at identifying defects in the final product. It is reactive in nature. The main goals of this process are to actually identify defects after the development of a product.
Question 5: Satisfying quality assurance
The main aspect that actually helped us to satisfy the quality assurance aspect in the data experiment was the design itself. For example, we used several parameters to ensure that the data collected was indeed truthful and free form defects. We also ensured that the tools used to collect and record each parameter were working effectively. The third quality assurance activity we did was to develop standards for our experiment. These standards were developed in line with previous research on the subject and this way, any defects in the study process could be identified.
Question 6: Data Quality Objectives Summary Report
Data quality objectives are the quantitative and the qualitative statements that cover or talk about the uncertainty levels that are acceptable by a given decision maker who has sanctioned a research and these are actually based on environmental dat. Data quality objectives provide a solid statistical framework for managing and planning operations of environmental data that are consistent with the needs of users.
The United States Environmental Protection Agency (EPA) has established a quality system that it uses to manage the overall quality of the collection, recording, generation and use of environmental data. The main goal of this system is to make sure that the environmental data that is collected every day for various uses is actually of sufficient quality and quantity so that it can support its initial intended use. This quality system has several lifecycle components
Systematic planning is actually a process that is largely based on widely accepted and used scientific methods in the collection of environmental data. It includes concepts like approach objectivity and results acceptability. Essentially, the process uses an approach that is to a large part common sense based to make sure that documentation levels and planning efforts are consistent with the information’s intended use and also with the resources available. It uses well established and scientific elements that accomplish various goals like logical development of project, efficient resources use, project transparency and proper documentation of results.
The performance criteria represents a set or series of specifications needed to formulate an effective information or data collection effort design , that when implemented will ultimately generate new data that is of sufficient quantity and quality and that will fully address the foals of the project.
Acceptance criteria is the set of specifications that are formulated with the intention of evaluating the adequacy or reliability of a particular information source as acceptable for supporting the intended use of the project.
The systematic planning process has various elements and these include: organization, project goals, schedule, data needs, criteria, data collection, quality assurance and analysis. Al these elements are coordinated in a harmonious way to ensure that the whole systematic planning process indeed yields positive results.
One of the most integral components of the EPA mission is the collection and dissemination of appropriate quality data or information. The Information Quality Guidelines (IGQs) provide a description of the agency’s policies about the level of quality of data disseminated by the agency. The guidelines apply to data or information that is generated by the agency or for the agency. It also applies to information and data adopted by the EPA or used by it to develop decisions or regulations.
There are actually several types of systematic planning. These include: the observational method (usually used by engineering professions), Triad approach, that combines systematic planning with advanced technological aspects and finally the data quality objectives process that is actually the most commonly used systematic planning application.
The data quality objectives (DQO) process is mainly used to establish the acceptance and performance criteria. These two criteria then provide a basis for the data collection plan design. The collection design established is one that provides data of appropriate quantity and quality to satisfy the project goals. The DQO process has seven primary steps and these are:
- Statement of the problem- Here, the problem that essentially necessitates the study is defined. The planning team is identified and an examination of the schedule and the budget is done.
- Identification of project or study goals-The method by which environmental data or information will be used in meeting project goals and objectives is stated. Study questions are identified and alternative outcomes are defined
- Identification of information inputs-In this step, the information and data that is needed to answer or solve study questions is identified.
- Definition of study boundaries-The target population and interest characteristics are specified. The temporal and spatial limits together with the inference scale are also defined.
- Development of the analytic approach- Here, the interest parameters are defined followed by the specification of the inference type to be used. The logic for drawing various conclusions from the research findings is also developed.
- Specification of acceptance and performance criteria- This is the step where the probability limits and boundaries for false acceptance and false rejection decision errors are specified. New data performance criteria and existing data acceptable criteria are developed.
- Development of the data obtainment plan- this is the final step where the most resource effective analysis and sampling plan and that is consistent with the performance criteria is selected.
References
United States & United States. (2000). EPA: 600 9 / Office of Research and Development, US Environmental Protection Agency. Washington, DC: US Gov.Print.Off.
Environmental Protection Agency (n.d.). EPA Quality System: Systematic Planning FAQs. Retrieved from http://www.epa.gov/QUALITY/faq10.html
Madans, J. H. (2011). Question evaluation methods: Contributing to the science of data quality. Hoboken, N.J: John Wiley & Sons.