SECONDARY DATA SETS

How can I identify secondary data sets for my research? Unlike with obtaining primary data sets, to identify secondary data sets, you do not carry out any intensive actual research from raw materials; instead, you make use of the data from the research which already has been done from the original source. You analyze the information in the primary data sources to come up with new secondary data- facts and statistics- which you carry forward to your new research. There are numerous sources of secondary data that can provide relevant and essential information and statistics which will help you come up with raw secondary data set for your research. They include; internet, databases, numerous scholarly journals, official recorded statistics, review articles, universities, reference books, government documents, libraries, trade journals, tapes and several technical reports (Vartanian, 2010).

All secondary sources are very important when conducting a particular research. Most of recorded tapes, books, magazines, government official records give clear and quality facts and data which can be used when carrying out research. Since most of data contained in the secondary data sets has been analyzed previously, there are less errors and minimal chances of biased facts and figures or false information. The regular surveys such as official registers or the government censuses contain detailed and quality information which can be used in research. However, inappropriateness of data facts contained in some of secondary data sets is the main weakness when conducting a research. Though secondary data set can provide vast information, sometimes it may contain irrelevant, untrue and biased data.

Therefore, a researcher needs to be very careful with the use of available secondary data sets. For instance, in a particular article, which was published in JAMA, Tan- with other authors- aims at comparing probably the outcomes in treated patients with nephrectomy when attacked by early stages of the fatal disease known as cancer (Tan, et. al, 2011). During their research about cancer, they decided to use SEER data set to obtain or get know who had stage T1, a cancer of the kidney. Later on, they linked such patients to Medicare claims data which enabled them to determine the patients who got healed from radical or partial nephrectomy. Appropriate use of secondary is very crucial. However, inappropriate use secondary data can lead to errors when conducting a research.

Moreover, there certain data mining techniques which can be very misleading, for instance, depending on a single data source to acquire a particular information or facts. An effective research need to mine data from various available secondary data sources such as scholarly article, websites, records and books. The hypotheses to be used during the research need to be clear. Large data sets contain complex surveys and data that tend to give complicated estimates and multiple data which may be actually confusing. For example, some of the secondary data sets found in health care institutions have no discharge weights, thought many patients are clustered by either physicians or else hospital. Actually, clustering is vital especially when obtaining accurate estimations in the records kept.