Archive for December, 2011

Collecting Secondary Data – research techniques article 3

Thursday, December 8th, 2011


In this third article on research techniques, we are going to explore secondary research methodologies.

The stages of research

 Research data comes from two core types of collection methodologies:-

  1. Secondary sources
  2. Primary research activities

 The building blocks


 Research methods will typically use a mix of the above elements. It is usual to follow the following order:-

a)    Secondary research phase

b)    Qualitative phase

c)    Quantitative phase

d)    Sometimes a further qualitative phase may take place

Observation can be used in both primary phases.

It is good practice totriangulateresearch methods. So by using 3 different methodologies to explore a given objective. This avoids atypical findings and ensures that a representative set of findings has been secured, as far as is possible.

 What is secondary data?


 This is existing data that was originally collected for another purpose. Also known as desk research because it always used to be done sitting at a desk! 

Secondary data should ALWAYS be gathered FIRST.

 It is often described as being less expensive than primary. This can be the case but not always. The activities need to be planned and closely managed to ensure that they do not absorb significant amounts of time and therefore cost.

 Much online research is undertaken by secondary data gathering using search engines, directories, forums and so on.

 Secondary data provides a basis for primary research and can, sometimes, meet current research needs without the need to continue to a primary stage.

 So what is primary data then?


This is data gathered expressly for the research purpose. We usually collect this type of data SECOND  – AFTER secondary sources have been exhausted.

The key elements here are the aspects of observing what happens and questioning why, how, who, when, which and so on.

The methods available will be explored in more detail in a future techniques article.

Data forms

Data comes in two forms. Qualitative and quantitative.

Qualitative data is gathered by using …”A body of research techniques which seeks insights through loosely structured, mainly verbal data rather than measurements. Analysis is interpretive, subjective, impressionistic and diagnostic.” *

Source: The Market Research Society

It is important to remember that this type of data is non-quantifiable. It is about what people ‘feel’, their perceptions, views, attitudes and so on. Things that are hard to quantify but just as important as numbers in many ways. The techniques to gather this data tend to be largely unstructured enabling respondents to talk about things in the way that they wish to.

Quantitative data is gathered by using “… a structured approach with a sample of the population to produce quantifiable insights into behaviour, motivations and attitudes”

Source: Alan Wilson – Marketing Research – An Integrated Approach


So this is about numbers – Percentages, rankings, scores, values, shares, numbers of those that buy, don’t buy, might buy, prices and so on.

The research process



The above visual shows the steps that need to take placed in a typical research process. It is a useful checklist to remind us of the importance of clearly identifying what it is that we want the research to deliver from the outset.

The nature of secondary data

There are 2 distinct sources of secondary data – internal and external.

Always seek internal sources first. Most go straight to Google without considering whether data might exist within the organisation. This can sometimes be in the ‘heads’ of personnel. Competitor information can sometimes be unearthed in sales persons’ cars, for example!


  • Sales figures
  • Operational data – stock levels, etc
  • Customer satisfaction results
  • Advertising spend
  • Customer complaints records
  • Effectiveness data from promotional campaigns
  • Marketing research reports from past studies



External sources are numerous. Consumer Generated Media (CGM), especially, has grown in importance as a data source.The key is to avoid spending too much time following ‘blind alleys’. This is where the time and cost can escalate sharply.

  • Directories
  • Country information
  • Published marketing research reports
  • News sources
  • CGM (Newsgroups, blogs, groups)
  • Internet – single search engines, and multiple search engines


The desk research or secondary research market is complex  – see below for some of the sources:-


Source: N Bradley

Some useful sites include:-

  •  Google Scholar

  •  World Factbook

  •  Published Market Reports

  •  Trade associations

Googlepublic is a particularly useful source for country based data.

Planning secondary or desk research


As mentioned above, planning desk research is all important.

Decide what you will explore and how long you will spend doing it. Do not exceed these boundaries before you review progress!

Do one thing at a time. Gather data sources and bookmark them. Do not be tempted to read each as you find the source. You will be diverted from your plan and end up exploring multiple potential ‘blind allies’! Think how easy it is to get diverted when searching for a holiday on the internet, for example.

Once you have gathered a number of data items, stop and skim read them. Then review and assess which are valuable and worthy of further exploration and which aren’t?

Internet Search Strategies


You must decide:

  • Time to spend
  • Number of sources you will look at
  • Age of sources. How far back will you go?
  • Format of data. What type of data are you seeking?
  • Methodology to use. What approach(es) to data gathering will you use?


Allocate time to:

  • Identify sources
  • Locate sources
  • Secure access to published/ peer approved material


Useful techniques to use

Use Boolean logic to make your search more efficient. This uses AND, OR, +, – and placing search words in speech marks to hone your search approach and save considerable time. The effectiveness of these can vary by search engine but generally the defaults are as follows:-


OR delivers material containing just College, just University and where College and University appear together.


AND delivers ONLY material where poverty AND crime appear.



NOT means that material where a word appears will NOT be included. So in this case only material with cats will appear. Beware however, where cats and dogs appear together the material will not appear in the search.

Internet search strategies

  • Wildcard *
  • Dictionary of synonyms
  • Bookmarks/Favourites
  • Find in page facility (via view or edit)
    • CTRL F
  • Recommendations, links and rings from sites you visit initially

The wildcard (*) is a useful tool to reduce typing long search strings. Add the * to a short set of words and the search engine will find all material starting with these words. Most search engines tend to do this by default now.

 “Ask your question” in inverted commas to deliver the exact words in that order in the search results.

 One-off needs

  • Set up temporary information requests
  • Use human experts
  • Talk to people! Email authors!


Don’t forget the power of talking to people that have some insight into what you are searching for. As indicated, these people may be in your own organisation, a trade body, an author of an original source or a librarian in a library!

  • Plan to stop receiving information.



Due to the inherent problems with secondary data – it was collected for some purpose other than yours – it is important to evaluate its quality and relevance. The internet provides us with large volumes of data but this makes it all the more important to ensure that it is carefully sifted and tested for validity and reliability.

Validity and reliability issues with all data

Validity = Are we measuring what we think we are measuring?

Reliability = If we did it again would we get the same results?


Use this simple test to help you to evaluate the data that you have sourced.


To assess the reliability of information look at the source and the context. This is particularly useful when assessing secondary data.

If the total is six or above then the information is likely to be trustworthy, if it is five or below then use should be made with caution.


Completely reliable (5)

Usually reliable (4)

Fairly reliable (3)

Not usually reliable (2)

Unreliable (1)

Reliability cannot be judged (0)


Confirmed elsewhere (5)

Probably True (4)

Possibly True (3)

Doubtful (2)

Improbable (1)

Truth cannot be determined (0)

Source: Nigel Bradley


Recording and reporting sources

Use Harvard referencing here to ensure that sources are captured and then reported effectively.

It is especially important to record dates and addresses for internet based data as these can frequently be updated, removed, amended and so on due to their nature.

You can find a detailed set of advice given by the CIM to its students and tutors on using Harvard referencing here.

CIM guide to Harvard Referencing

 Limitations of secondary data

  • Availability

Remember that data is not always available on the internet. Not all material gets placed on the net. Think about your requirements and seek others input on how and where data to meet these might be found.

  • Applicability
  • Accuracy
  • Comparability
  • BIAS
  • Age
  • How parameters defined for sampling, for example
  • Sufficiency – it there enough data for your purpose?

As already mentioned, key limitations revolve around the fact that the data was not created for your purpose. More often than not it provides you with a base for primary research rather than being totally sufficient for your needs. ALWAYS look at secondary data with a healthy degree of scepticism. It is valid? Who was the audience for the data in the first place? What are the likely biases implicit for that audience? How was the sample structured? Are the recommendations sound and based on valid research data?

Finally, the data may just not exist so primary data collection is your only route!

Some pros and cons for secondary collection methods


DESK RESEARCH Not everything is available on the internet! There are books and other printed media that are not available via traditional searches. Using a library and especially  the skills and knowledge of a good librarian can significantly short circuit the secondary data gathering process Time. As with all secondary data collection there are plenty of dead ends that will be arrived at with this method. You have to travel (usually) to the source. So significant potential for time and cost.
SECONDARY DATA SEARCHING ONLINE Inexpensive if planned and managed well. Vast quantities of data available. Not all desired data is online! Need to go back to source where at all possible. Need to combine data to verify and validate it.
BLOGS/ CGM (CONSUMER GENERATED MEDIA) A good source of data potentially. A subset of secondary data collection. Same issues as other secondary sources. Who says? Why are they saying it? Be sceptical. Ask validity and reliability questions of any data you derive from these sources.

 For help and advice on gathering secondary data contact Steve at Bax Interaction now. Click here