Primary Data & Secondary Data

 

  Two types of Data & Methods of Collecting it.

In addition to deciding who should supply his marketing research needs the manager will also want to contribute to the decision as to what type of data is most appropriate, i.e. primary data or secondary data

1.     Secondary data

The term 'secondary data relates to data which has been collected by individuals or agencies for purposes other than those of a given research study.

Characteristics of secondary data

• Secondary data may be sufficient to solve the problem. On occasion it happens that adequate data may be available to the extent that primary data collection unnecessary.
• Data collection costs are substantially lower for secondary data in comparison to primary data. A thorough search of secondary sources can be completed at a fraction of the cost incurred in even a modest primary data collection exercise.
• The time involved in searching secondary sources is far less than that needed to complete primary data collection. A systematic search of secondary sources can be completed in a fraction of the time it takes to complete primary data collection.
• Secondary sources of information can yield more accurate data than that obtained through primary research. This is not always true but where a government or international agency has undertaken a large scale survey, or even a census, this is likely to yield far more accurate results than independent surveys when these are based on relatively small sample sizes.
• Secondary data helps define the research problem and to formulate hypotheses. The assembly and analysis of secondary data almost invariably improves the understanding of the marketing problem, the various lines of inquiry the study could take and the alternative course of action which might be pursued.
• Secondary sources help define the population. Secondary data can be extremely useful both in defining the population and in structuring the sample to be taken. For instance, government statistics on a countries' agriculture will help decide how to stratify a sample and, once sample estimates have been calculated, these can be used to project those estimates from the sample to the population.

Whilst the benefits of secondary sources are considerable, their shortcomings have to be acknowledged. The main problems may be categorised as follows:

Problems with secondary sources

Definitions: The researcher has to be careful, when making use of secondary data, with regard to the definitions used by those responsible for its preparation. 

Measurement error: 

Source bias: Researchers have to be aware of vested interests when they consult secondary sources. 

Reliability: The reliability of published statistics may vary over time

 

Time scale: Published sources may be out-of-date at the time the researcher wants to make use of the statistics.

Sources of information

Secondary sources of information may be divided into two categories; internal sources and external sources.

1. Internal sources of information

All organisations collect information in the course of their everyday operations. Orders are received and delivered, costs are recorded, sales personnel submit visit reports, invoices are sent out, returned goods are recorded and so. 

For example, consider how much information can be obtained from sales invoices:

• sales by territory
• sales by customer type
• average size of order by customer
• customer type, geographical area
• average sales by sales person
• sales by pack size and pack type.

2. External sources of secondary information

The main external sources of secondary data are (1) government departments (2) trade associations (3) domestic and international commercial information services (4) national and international development organisations institutions.

Government Statistics: Federal, state and local government departments usually publish a wide range of statistics. These may include all or some of the following:-

• population censuses
• social surveys, family expenditure surveys
• import/export statistics
• production statistics
• agricultural statistics

Trade Associations: Trade Associations differ widely in the extent of their data collection and information dissemination activities. However, it is worth checking with them to determine what they do publish. At the very least one would normally expect that they would produce a trade directory and, perhaps, a yearbook. Chambers of commerce could also prove useful as an information source.

Commercial Services: Published market research report and other publications are available from a wide range of organisations who charge for their information. Typically, marketing people are interested in media statistics and consumer information which has been obtained from large scale consumer or farmer panels. The commercial organisation funds the collection of the data, which is wide ranging in its content, and hopes to profit from selling this data to interested parties.

National and International Institutions: Bank economic reviews, University research reports, journals and articles are all useful sources to contact. International agencies such as World Bank, FAO, UNDP, ITC and ILO produce a plethora of secondary data which can prove extremely useful to the marketing researcher.

2.     Primary research / Primary Data

Primary research is that which has been specifically designed to address particular marketing problems or questions. Perhaps the approach most readily associated with marketing research is the survey but as we are about to see this is but one, and not invariably the most appropriate, approach. The principal approaches to primary marketing research are:

• Survey research
• Qualitative research
• Observation
• Experimentation
• Continuous research

Survey research: Surveys are characterised by a relatively large number of respondents and the desire to project the results obtained from a sample to a population. If the sample is drawn using a probabilistic method then we can place confidence levels on the inferences we make about the population. Where a non-probabilistic method is used we cannot say how certain or uncertain we are about our inferences. Nonetheless if the sample is reasonably large, and comprised of a good cross-section of the target population, then marketing researchers tend to assume that the sample results are representative of the population. The main forms which surveys take are depicted in figure 11.4.

 Qualitative research: In situations where the researcher is primarily interested in why people thing and/or behave in a particular way rather than in being able to quantify things, then qualitative research methods are likely to be employed. Qualitative methods have at least four distinguishing characteristics:

• Small numbers of respondents. The idea is to devote a considerable amount of time on each interview to get to the heart of a matter.
• Unstructured question formats. That is, the questions are not completely predetermined and the interviewer is free to probe for all details and underlying feelings.
• Indirect measurement of respondents' feelings and beliefs. Respondents provide descriptive information about their thought and feelings. These are not easily projected to the population.
• Direct observation. The interviewer not only records answers but observes how questions affect interviewees. Hesitant answers, agitation, smiling, sweating, calmness, boredom etc. are all observable and all tell us something about the individuals state of mind.

Three commonly employed qualitative marketing research methods are projective techniques, focus groups and depth interviews.

 Focus groups: Each focus group generally involves six to eight people who meet with a moderator for a discussion. The discussion is focused, by the moderator on a particular topic. Typically, a group session will last one to two hours. The discussion is free ranging with the moderator intervening only periodically to stimulate the discussion in a particular direction. The moderator uses a discussion guide rather than a questionnaire. This guide is simply an agenda of the topics which the group should cover. Thus, the focus of the discussion, at any point in time, is subtly controlled by the researcher (hence the term moderator). Participants in the groups are chosen on the basis that they belong to the target market.

Any number of focus groups may be held in connection with a particular marketing problem but the results are not strictly projectable to the population since the selection of participants is in no way probabilistic.

Depth interviews: Depth interviews are like lengthy psychoanalytic sessions between a single respondent and a highly skilled interviewer. The idea is to get to the deep, hidden underlying attitudes and feelings the respondent has towards a product, service, company or problems which a product is trying to solve.

Depth interviews are of most value where a study deals with (1) a confidential, emotionally charged or embarrassing matter; (2) a behaviour for which socially acceptable norms exist and the need to conform in group discussions influences responses; (3) a complex behavioural or decision-making process that requires a detailed idiosyncratic, step-by-step description; and (4) when group interviews are difficult to schedule for the target population.

Projective techniques: On occasion, the interests of the research are best served by obtaining information on respondents' beliefs and feelings indirectly. Projective techniques presume that respondents cannot or will not communicate their feelings and beliefs directly. Instead, respondents are encouraged to respond indirectly by projecting their own feelings and beliefs into the situation as they interpret the behaviour of others. The most common projective techniques are:-

Thematic apperception tests	Respondents are presented with a series of pictures or cartoons in which consumers and products are featured. Participants are asked to study the situation depicted and to comment on what is happening or what might happen next. In this way, respondents are encouraged to project their own feelings and beliefs onto the situation portrayed in the pictures or cartoons. The term thematic apperception test is used because themes (thematic) are elicited based on the perceptual-interpretive (apperception) use of pictures and cartoons.
Word association	Respondents are presented with a series of words, one at a time, and asked to indicate what word comes immediately to mind. The respondent's response and time to respond are recorded. Elapsed time and associations are the key measures. Word association is commonly used in the testing of brand names.
Sentence completion	Sentence completion tests are similar to word association. Respondents are asked to conclude a number of incomplete sentences with the first word or phrase that comes to mind. Responses are then analysed for content and meaning.
Scenario/story completion	Respondents are asked to complete the end of a story or supply the motive for why one or more actors in a story behaved as they did.
Third person/role playing	Respondents are presented with a visual or verbal situation in which they are asked to relate the feelings and beliefs of a third person - for example, a friend, neighbour, another farmer or ‘typical’ person - to the situation, rather than to directly express their own feeling/beliefs about the situation. In this way the individual reveals his/her own inner most feelings, attitudes and motives.

As was said earlier, qualitative research methods are, best employed where the task is to address ‘why’ questions. However, the results of qualitative research are rarely projectable to the population at large. Moreover, they must be carried out by interviewers trained in psychology and/or sociology.

Observation: Methods of data collection involving directly or indirect, human or mechanical measurement of behaviour, are termed observational methods. These can be particularly useful in situations where the respondent is either unable or unwilling to report past behaviour, or in cross-cultural research where it is possible that imperfect translation of the questions can occur. Observation methods are also called naturalistic inquiries because, in their purest form, such studies demand a natural setting. This is because behaviour takes its meaning as much from their context as they do from themselves⁴. Examples of observation methods include pantry and dustbin audits, and physiological measurements.