Authour: Thuy Linh Do
Edited by Christine Keene
Secondary Research is a common research method; it involves using information that others have gathered through primary research.
- The information already exists and is readily available -> quick & low cost
- Helps guide the focus of any subsequent primary research being conducted
- Internal secondary data uses categories and breakdowns that reflect a corporation’s preferred way of structuring the world
- Secondary research may be the only available source of specific pieces of information (i.e. government data)
- The information lacks specificity or does not exactly address question of concern
- Some external secondary data may be of suspect quality or outdated
- Internal secondary data such as sales reports and customer databases may only describe existing customers
- Information is less likely to exist, particularly in developing countries, due to the lack of primary research conducted in unpopular markets or strict media control from the governments
This technique is performed in order to:
- Assess easy, low-cost and quick knowledge;
- Clarify the research question;
- Help align the focus of primary research in a larger scale and can also help to identify the answer; and
- Rule out potentially irrelevant project proposals (ex. The proposed work may have already been carried out).
This technique is also known as Desk Research.
There are two types of Secondary Research hence two types of data collected from this technique:
- Internal Secondary Data consists of information gathered within researcher’s firm (i.e. customers databases and reports from past primary research)
- External Secondary Data consists of information gathered outside of researcher’s firm (i.e. government statistics and information from media sources)
Using the Technique
Secondary Research can happen at any stage of the creative process. Each Secondary Research process involves 4 steps that can be repeated as necessary:
- Identifying the subject domain and where to acquire the information;
- Gathering existing data;
- Comparing data from different sources, if necessary and if feasible; and
- Analyzing the data
1. IDENTIFYING WHAT & WHERE
Before starting any Secondary Research, it is helpful to define the research topic/domain. Next, the researcher would prepare a list of questions to be solved by the end of the process. This step helps narrow down the topic and also allows researcher to have an active role in conducting the research. After identifying the research domain, the researcher would look at various sources of information and decide where to get necessary data.
Good sources of information include:
- Internal data such as databases, sale reports, past primary researches;
- Government statistics and information from government agencies such as Canada Business Service Centre (http://www.canadabusiness.ca), Statistics Canada (http://www.statcan.gc.ca/);
- Information resources companies (ex. Passport GMID or Datamonitor360); and
- Different media such as articles from respected magazines and newspaper, reports from university research centers or non-profit agency.
2. GATHERING EXISTING DATA
At this step, researcher looks at the topic and breaks it down in to keywords and their synonyms. For example, when looking at the topic: “What are the trends in woman clothing market?” the keywords would be “clothing”, “women” and “trend”. Accordingly, their synonyms would be “apparel”, “female” and “fashion”. Using these words to search can save researcher a lot of time in finding valuable data and also warrant no important information to be missed out.
3. NORMALIZING DATA IF NEEDED
Sometimes researchers would want to normalize the data to make it easier to analyze later.
Example for this step comes from a research project of area household income data in the US. The collected information came from 3 different sources: US Census Bureau Data (1997 data), a telephone survey of area residents (2000 data) and a published article (2007 data).
Raw information table
Secondary data is one type of quantitative data that has already been collected by someone else for a different purpose to yours. For example, this could mean using:
- data collected by a hotel on its customers through its guest history system.
- data supplied by a marketing organization.
- annual school testing reports.
- government health statistics.
Secondary data can be used in different ways:
- You can simply report the data in its original format. If so, then it is most likely that the place for this data will be in your main introduction or literature review as support or evidence for your argument.
- You can do something with the data. If you use it (analyze it or re-interpret it) for a different purpose to the original then the most likely place would be in the ‘Analysis of findings’ section of your dissertation.
Example: A good example of this usage was the work on suicide carried out by Durkheim. He took the official suicide statistics of different countries (recorded by coroners or their equivalent) and analyzed them to see if he could identify variables that would mean that some people are more likely to commit suicide than others. He found, for example, that Catholics were less likely to commit suicide than Protestants. In this way, he took data that had been collected for quite a different purpose and used it in his own study – but he had to do a lot of comparisons and statistical correlations himself in order to analyze the data. (See Haralambos, 1995, for details of Durkheim’s work).
Most research requires the collection of primary data (data that you collect at first hand), and this is what students concentrate on. Unfortunately, many research reports do not include secondary data in their findings section although it is perfectly acceptable to do so, providing you have analyzed it. It is always a good idea to use data collected by someone else if it exists – it may be on a much larger scale than you could hope to collect and could contribute to your findings considerably.
As secondary data has been collected for a different purpose to yours, you should treat it with care. The basic questions you should ask are:
- Where has the data come from?
- Does it cover the correct geographical location?
- Is it current (not too out of date)?
- If you are going to combine with other data are the data the same (for example, units, time, etc.)?
- If you are going to compare with other data are you comparing like with like?
Thus you should make a detailed examination of the following:
- Title (for example, the time period that the data refers to and the geographical coverage).
- Units of the data.
- Source (some secondary data is already secondary data).
- Column and row headings, if presented in tabular form.
- Definitions and abbreviations, for example, what does SIC stand for? For example, how is ‘small’ defined in the phrase ‘small hotel’? Is ‘small’ based on the number of rooms, value of sales, number of employees, profit, turnover, square meters of space, etc., and do different sources use the word ‘small’ in different ways? Even if the same unit of measurement is used, there still could be problems. For example, in Norway, firms with 200-499 employees are defined as ‘medium’, whereas in the USA firms with less than 500 employees are defined as ‘small’.
There are many sources of data and most people tend to underestimate the number of sources and the amount of data within each of these sources.
Sources can be classified as:
- paper-based sources– books, journals, periodicals, abstracts, indexes, directories, research reports, conference papers, market reports, annual reports, internal records of organizations, newspapers and magazines
- electronic sources– CD-ROMs, on-line databases, Internet, videos and broadcasts.
Return to Data Instrumentation