Primary Data

1.0 Introduction
Sources of information are generally categorised as primary, secondary or tertiary depending on their originality and their proximity to the source or origin. For example, scientific information moves through a dissemination cycle. Initially, findings might be communicated informally by email, then presented at meetings before being formally published as a primary source. Once published, they will then be indexed in a bibliographic database, and repackaged and commented upon by others in secondary sources. The designations of primary, secondary and tertiary differ between disciplines or subjects, particularly between what can generally be defined as the sciences and the humanities. Primary sources for critic studying the literature of the Second World War are different from those for a research scientist investigating a new drug for arthritis. The critic's primary sources are the poems, stories, and films of the era. The research scientist's primary sources are the results of laboratory tests and the medical records of patients treated with the drug. You should always check with your lecturer or tutor if in doubt.

2.0 Analyzing Your Primary Data
Now that you've collected your primary data, its time to figure out what that data means and what you can learn from it. The keys when analyzing your data is to pull out information that is the most pertinent to your writing, information you can highlight and discuss, and information that will support your claims (if you are making any). Interviews Interviews are fairly easy to analyze, as you simply have to go back through the answers you received and decide how to use them within your writing. You can group the answers into categories and create a chart of how those answers may best fit within your paper or article. If you recorded the interview with a tape or digital recorder, you may want to listen to it and type a transcript of the interview. Since transcription is a tedious process, only use this option if you need to. Surveys

When analyzing surveys, you want to get the raw data into form that you can manipulate. If you were using a numerical system or yes/no answer system for your survey, you may find it helpful to enter the results into a spreadsheet program such as Microsoft Excel. If the survey was an open-ended question style, see if you can fit your answers into categories of responses.

Observations Observations are more difficult to analyze because when you are taking notes, you often write down everything that you see. Start by organizing your notes into categories or by some criteria. Once you have everything organized, see if you can make some generalizations about what you have observed. Over-generalizing your results Your first attempts at primary research will most likely include small groups of people and may not be representative of the population as a whole. It is important to remember not to over-generalize your findings--in other words, don't assume that your findings are necessarily true of every person within the group or every person in a society. Triangulation of Data One of the benefits of combining primary research with secondary research is in the area of data triangulation. Data triangulation is when a piece of data, a finding, or a generalization is able to be verified with several different research methods. This helps add to your credibility and makes your findings stronger. For example, you are studying binge drinking on campus. You find national averages that indicate that 45% of college students binge drink nationwide. You conduct your own research at the Purdue campus. You find that 47% of the individuals you surveyed drink; you also interview a counselor on campus who reports that approximately 1/3 of the students who he sees suffer from a drinking problem. Thus, your results from an interview with an expert and your own survey support the national averages.

3.0 Primary Data Sources
Primary data collection is necessary when a researcher cannot find the data needed in secondary sources. Market researchers are interested in primary data about demographic/socioeconomic characteristics, attitudes/opinions/interests, awareness/knowledge, intentions, motivation, and behavior. Three basic means of

obtaining primary data are observation, surveys, and experiments. The choice will be influenced by the nature of the problem and by the availability of time and money. 3.1 Observation Observation means that the situation of interest is checked and a person or some mechanical device records the relevant facts, actions, or behaviors. Accurate data about what consumers do in certain situations is provided by observation. Observation does not tell why it happened. Mechanical Approaches Mechanical approaches are reliable data collection instruments because they provide objective measures. Data on the factors influencing product sales, such as competitor advertising and other promotional activities can be effectively assessed. Information can be obtained on a specific store or all the stores in a system, enabling rapid and effective comparisons at various local, regional and national levels. The information is available continuously and enables firms to plan down to the individual store level. Scanner and bar coding technologies form the basis for capturing marketing information at the retail level. Scanners are electronic devices at retail checkouts that read the bar code for each item bought. They provide up-to-the-minute data on product purchases by item and also by household. Telecommunications can transmit the information directly to the manufacturer and shorten the communications cycle from weeks to minutes. With this information the manufacturer can develop a profile for each retailer and establish the optimum retail inventory for each location. The optimum inventory ensures stocking of merchandise that customers buy with a minimum amount of inventory investment for the retailer. Combining the retailer’s information with the manufacturer’s database yields local promotional mailings, fine-tuned shelf displays, and redesigned store layouts. Other mechanical devices include video cameras, Nielsen People Meters, and singlesource data systems that link consumers’ exposure to television advertising, sales promotion, and other marketing efforts with their store purchases (Behavior Scan and InfoScan of Information Resources, Inc.). Furthermore, measurements might be taken of respondents’ eye movements, pulse rates, or other physical reactions to advertisements. Personal Approaches Marketers can learn by personally observing or watching actions and situations. For example, when an organization is choosing a new location, it would observe the neighborhood conditions. Also, marketers of pet products and baby products are extremely interested in how respondents react to new products, but obviously cannot ask them to describe their opinions or to fill out surveys. They must depend on observational research. 3.2 Surveys

Surveys or questioning involve using a questionnaire (data collection instrument) to ask respondents questions to secure the desired information. Questionnaires may be administered by mail, over the telephone, by computer, or in person. Limitations of surveys include opportunities for error in construction and administering of a questionnaire, expense, and time needed to conduct a survey. Respondents may not respond, may be unable to respond, or may give misleading responses.

Mail Mail interviews can be used to collect large amounts of data and have a low cost per respondent. Respondents can see a concept, read a description, and think about it at their leisure. There is no interviewer bias. However, the questionnaires are not flexible, cannot be adapted to individual respondents, and generally have low response rates. The researcher has no control over who completes the questionnaire. Telephone Telephone interviews are easy to administer and allow data to be collected quickly at a relatively low cost. The interviewer can clarify the questions. Response rates tend to be higher and telephone interviewing allows for greater sample control. However, it is more expensive than a mail questionnaire. The presence of an interviewer on the phone may bias responses since respondents may be unwilling to discuss personal information. Also, respondents can’t see product. A major limitation is that they must be short. Computer Advances in computers and technology have led to sophisticated data collection methods. Computer and online interviewing allow rapid data collection from dispersed populations at a low cost. Personal Personal interviews may be conducted one-on-one or with a focus group. A personal interview is a direct, face-to-face interview between the interviewer and the respondent. In the past, personal interviews were conducted door-to-door. Today, most personal interviews (one-on-one) are conducted in malls and are referred to as mall intercept. Personal interviews are the most flexible since interviewers can clarify questions and probe for answers. Respondents can see a concept as well as read a description. More information can normally be obtained through observation of the respondent’s surroundings. Personal interviewing is expensive, yet it offers a great deal of flexibility and allows for visual stimuli. A focus group is a small group of people, carefully selected, who represent a specific target audience. They are used to generate concepts and hypotheses. The strength of

focus groups is found in the group discussion and interaction. Focus group interviews are a popular way of gaining insight into consumer thoughts and feelings about a product. In the past, focus groups were regarded mainly as a simple and quick way of asking any group of respondents, usually in someone's home, to answer questions about a product. Today, focus groups are an important source of qualitative research. Advance preparation ensures that the facility, moderator, and respondents are of high quality. An example of a technique used in a focus group is a projective technique in which a People Board is used to obtain attitudes through photograph associations and forced relationships. Participants indicate which of several images in a category relate to the subject at hand. The findings from a focus group are useful for general information but do not suffice to give absolute quantifiable information. A panel is a fixed sample of individuals from who repeated measurements are taken over time with respect to the same variables. An example is MRCA, which has a 12,000 household panel that is representative of the national census in terms of significant demographics. Surveys of the panel conducted frequently throughout the year provide a means to measure relatively small changes in household purchases and product usage. Another example is the consumer diary. Diaries are especially appropriate for answering questions on brand penetration and loyalty. This approach indicates which factors influence purchasing behaviors, such as price and advertising, and where purchases are made -- supermarkets, discount stores, drugstores. Firms in packaged goods, apparel, home furnishings, financial services, travel, and entertainment use this method. 3.3 Experiments In an experiment, a researcher selects matched groups, gives them different experimental treatments controlling for other related factors, and checks for differences in the responses of the experimental group and the control group. Experimental research attempts to explain cause-and-effect relationships. Data in an experiment may be collected through observation and surveys. An experiment can be done in either a laboratory or field setting. In a laboratory experiment, the researcher has complete control during the experiment. A field experiment is conducted under more realistic conditions. For example, if a charitable organization wanted to see whether inclusion of return-address labels affected donors’ responses to a mail solicitation, it could select similar sets of donors and send them the donation solicitation with and without labels to see if one method is more effective than the other is.

4.0 References