Unit 4 sec 2.1 primary and secondary data

Unit 4 sec 2.1 primary and secondary data

17 November 2015

23:22

You might consider collecting some data yourself. This might be particularly appropriate to question of interest is one very specific to you and your surroundings or on which you can collect relevant data quite easily. Data that you collect yourself a called primary data. For more substantial research questions, this tends to be a reasonable broke only in yourself refers to a research team in the University, company or other research unit.

Types of statistical question, before rushing into collecting data about any question, it is always a good idea to ask. Has anyone collected data on this before? The answer is often yes. Secondary data are data that already exists and can be used to adopt for your purpose.

Secondary data are plentiful and often readily available through the Internet, published literature and other sources. There are a number of consistently reliable sources such as UK government sister statistics which are generally professional way collected and presented and free from bias.

 However, other sites are set up to by organisation that may want to sell you some product or promote the particular set of ideas. In some cases, the data that they present may be subject to bias or distortion, and search site are best avoided as sources of reliable secondary data.

 Data are usually percent as “datasets”. A dataset is collecting data usually presented in the tubular form, or as a single row, or something as a single column.

An importing convention when presenting any dataset, where the primary or secondary, is to provide an accurate reference to the data source (so that the reader can get the details if they wish).

Backache in pregnancy

secondary dataset collected at the London Hospital (new Royal London Hospital). It was designed to help answer questions concerning backache in pregnant women, including: how common it is and how severe? What factors affected? What factors alleviate it?

In order to make the datasets manageable for your work in this unit, the number of respondents has been reduced from 180 women to 33, and the number of items of information reduced from 33 to 13.

 This dataset will be used to illustrate most of the issues concerned with handling data in this section, and that the end, the queue of the data values from the original source have been changed. Several of the corns in this table have been entered into the module software resource dataplotter but for reasons that will be explained shortly, pure the data values from this table have been changed.

Take a quick look at these data. This 1^st thing to notice is that each row corresponds to results from one patient, and the column -expects the first to a specific item measured.