Discuss the factors which have stood in the way of acceptance of marketing research in india

Classifications of observational research

Observation can be used as both a quantitative and a qualitative research methodology. In the case described in this study, observation was mainly used qualitatively as the research was highly exploratory in nature. On the other hand, observation, if structured, can generate detailed quantitative findings. Data, for example, generated via EPoS tracking (a machine-based observational tool) is highly statistical in nature. Whether findings generated by observation are quantitative or qualitative in nature depends on whether the research is structured or unstructured - which, in turn, often depends on the stage of the research project.

Observational research has been classified in terms of various dichotomies (see, for example, Malhotra, 1996; Rebello, 1977). There are several widely used classifications, of which the most popular are:

• human versus machine;
• structured versus unstructured;
• overt versus covert;
• natural versus disguised;
• participant versus non-participant.

Each of these classifications are discussed below.

Human versus machine observation

Observation is often analysed in terms of who (or what) is doing the observation (McDaniel and Gates, 1995). Where the observation is less structured, particularly in the initial stages of a research project, a human-being, rather than a machine, is often considered the appropriate observer because, while a machine is superior in terms of speed, a human-being is superior in terms of interpretation. Large-scale, statistically-driven observation often necessitates the use of a machine as the "observer" - such as EPoS equipment which is utilised to track product sales. It is also the case that certain observable phenomena may not be visible to the naked eye - hence the development of equipment such as the pupilometer and the psychogalvanometer (Zikmund, 1989).

Structured versus unstructured observation

The degree to which a period of observation is structured largely depends on the stage of the research project. In the early stages of research, before the development of working hypotheses, observation tends to be unstructured (or ad-libitum) in an attempt to identify variables to investigate more scientifically (i.e. in a more structured fashion). Altman (1974, p. 235) refers to the unstructured recording of behaviour as "typical field notes", with the researcher "recording as much as he can or whatever is most readily observed...". Structured observation, on the other hand, will usually involve time sampling - i.e. the detailed (usually quantifiable) recording of a phenomenon over a given time-frame.

Overt versus covert observation

Often discussed in the sociological literature in terms of ethics (see, for example, Adler and Adler, 1994), observation can be either overt - where people are aware of being observed - or covert - where the observer disguises the fact that observation is taking place. Covert observation is often used in focus group discussions, where, behind a two-way mirror, a second moderator records reactions and body language in some detail (Zikmund, 1989).

Natural versus contrived observation

This classification is concerned with where the observation takes place: either in the observees' natural physical environment or in a contrived setting such as an artificially-created supermarket set up with sensors and cameras. It is argued by many academics that observation should, as far as possible, take place in natural environments because observees tend to be more relaxed, thus giving the researcher a truer picture of the phenomenon, as "reactivity" is minimised (Bryman, 1992).

Participatory versus non-participatory observation

A dividing line is drawn between those who participate in the context of the research and those who do not. Those who do not participate are known as non-participatory observers. Participant observers, on the other hand, have been further subdivided by Gold (1958) into the complete participant, the participant as observer, the observer as participant, and the complete observer. The first conceals his research role, the next two reveal their research roles while participation varies from a great deal to a minimum, and the complete observer is entirely removed from social interaction with his informants and observes in unobtrusive ways (Rebello, 1977).

Criteria for using observation as a market research technique

Observation is an appropriate methodology for conducting market research when at least one of the following four criteria is met:

1 the phenomenon under investigation is easily observable;

2 the phenomenon under investigation is a social process or a mass activity;

3 the phenomenon under investigation occurs at a subconscious level;

4 the consumers under investigation are either unable or unwilling to communicate directly with the researcher.

First, and perhaps most obviously, the phenomenon being investigated must be able to be observed. This pre-supposes that the consumer behaviour being studied is public, rather than private, behaviour. One of the main advantages of observational research is the way it gives direct access to social interaction. It is a flexible technique, and can be used very effectively to enrich and supplement data gathered by other methods (see Simpson and Tuson, 1995). It can also be used profitably at the start of a study to uncover promising areas of investigation for further study (Bryman, 1992). From the respondents' point of view, it is probably the most convenient form of data gathering as it requires no effort on their part. In the case of covert participant observation, however, there may be an ethical dimension to be considered if the actions of the individual being observed are in any way criminal or otherwise reprehensible.

Second, observation is useful in the analysis of large-scale social processes, or mass-activities, where a pattern of activity is under investigation, and aggregate statistics are all that is required. Examples would include the analysis of licence plates of parked cars to assess shop catchment areas (see the work of Urban Science Applications and R.L. Polk - as cited in Zikmund, 1989) and the study of household waste (Cote et al., 1985). The sheer scale of data required in such cases necessitated the use of structured observation - which was regarded as the quickest, most economical, methodology to uncover patterns of behaviour.

Third, observation is useful for examining subconscious influences on consumer behaviour, for example, the studies linking music to shopping behaviour. Milliman (1982) reported a link between the speed of music played in supermarkets, the speed of shopping and daily takings. The faster the music, argued Milliman, the faster the shop. Furthermore, Leicester University's Music Research Group reported that the type of music played can influence the type of product sold (Hawkes, 1997). Such research is best investigated by observation because respondents to a questionnaire would probably regard themselves as being too sophisticated to allow music to influence their speed of shopping and the type of products they buy. Observational techniques, being unobtrusive, can uncover linkages between subtle influences and subconscious changes in shopping and buying behaviour; linkages that may not be uncovered through other research methodologies. As the title of the article suggests, what people say is often different from what they do. Observation can provide an accurate record of individuals' actions in situations where they do not wish to reveal their behaviour or where they genuinely do not have a "conscious" reason for their behaviour.

Fourth, in certain circumstances, observation may be the only methodology available. This is true in research investigations involving consumers who speak a foreign language, very young children, or the mentally handicapped. For example, Children's Television Workshop (CTW), the makers of Sesame Street, deployed observational techniques to gauge children's reactions and attention spans regarding various aspects of the programme in order to assess which parts were considered particularly (un)interesting (Waterman, 1987). Obviously, for a programme such as Sesame Street, which is aimed at pre-school-aged children, any research aimed at gauging their reactions would have to be observationally based because the children themselves are still learning to communicate. More information could be gleaned by observing the children watching the programme than by attempting to ask them even simple questions.

Within the "only available methodology" justification, one can also include the analysis of voice pitch and content. By their very nature, the study of both voice patterns and the content of written material presuppose that observation is the only appropriate research methodology. In the case of voice pitch analysis (see, for example, Nighswonger and Martin, 1981), the researcher has to be present to observe the vocal response to a stimulus; any other market research methodology, such as an interview, where the information comes to the interviewer second hand, is obviously futile. In content analysis (see especially Kassarjian, 1977), the researcher has to employ observation to "systematically evaluate the symbolic content of all forms of recorded communications" (Kolbe and Burnett, 1991, p. 243).

Methodology

Background

Much has been written in the retailing literature on the overriding importance of location. The bulk of academic attention has concentrated at the macro (i.e. regional or city) level often with the intention of creating mathematical models to predict the likely success of a particular store in a particular region (see, for example, Craig et al., 1984; Davies and Rogers, 1984; Mahajan et al., 1985; Rust and Donthu, 1995). It has been argued, especially by Brown (1993) that micro-level (i.e. individual site) aspects of locational decision making have not received sufficient attention in the literature. Brown (1993, p. 10) is concerned that too much research pertains to "national, regional and urban scales of analysis". It is the aim here to demonstrate the applicability of observation as a research technique to micro-level site location analysis.

The aim of the study was to develop guidelines for the siting of middle-market restaurant outlets in order to maximise the number of consumers passing by at lunch-times. The location of the study was three interconnecting streets in South London. The study had two parts. The first involved consumer movements around the high-street and the second involved the researchers in visiting all the restaurants in the role of consumers at lunch time and in the evening.

Phase one

In the first phase, the following factors were assessed to see what impact they had on high-street consumer traffic flows in general and on restaurant usage in particular:

1 Topography:

• a curve in the road;
• height of buildings;
• the sunny side of the street;
• pedestrian crossings;
• public transport sites;
• gradient of the street.

2 Shop type:

• banks with cash-withdrawal facilities;
• general shop type in immediate vicinity of a restaurant.

Both unstructured and structured observation were deployed as the key research techniques. Unstructured (or ad libitum) sampling was used in the very early stages of the research to assess if there was any evidence to support the factors hypothesised above that may influence high-street consumer traffic. Structured (or time-sampling) observation was used to assess the hypothesised linkages more scientifically. Consumer traffic counts were conducted at various points along the three streets. Count locations all had a strategic purpose as they were conducted:

• at far edges and the centre of the streets;
• adjacent to pedestrian crossings;
• on the outer and inner edges of curves in the road;
• at the top and the bottom of a hill;
• near banks with cash withdrawal facilities;
• outside large department stores and small specialist outlets.

Consumers at each count location were segmented by group size. Counts were conducted for 15-minute periods and were undertaken several times in an attempt to gain "theoretical saturation" (Glaser and Strauss, 1967).

Phase two

All Whitbread's restaurants in the study location were visited four times between 12.30 p.m.-2 p.m. over a four-week period. Detailed notes were made of the customers using the following classifications:

1 Dining size/type:

• individuals;
• couples;
• families;
• similar-aged groups.

2 Dining "purpose":

• business;
• family treat;
• celebration;
• romantic one-to-one.

3 Style of dress:

• formal;
• casual.

4 Mode of transport:

• walk;
• taxi;
• car;
• bus.

By analysing the clientele of the restaurants, it was possible to assess if there was a congruence between the "type" of consumers on the streets on which the restaurants were situated and the "type" of consumer in the restaurants themselves. In other words, the aim was to assess whether the restaurant was situated in the correct place to maximise its potential lunch-time trade and to identify whether the actual clientele matched the intended market segment.

Discussion

Observation was considered the most appropriate research methodology in the case of consumer high-street traffic flows because:

• The phenomenon is easily observable.
• Shopping on the high-street is a mass activity and a social process, so it was easier and more economical for the researcher to observe the phenomenon than interview a percentage of those taking part. Moreover, as soon as an individual shopper is stopped and interviewed, his/her individual pattern of shopping is immediately disrupted. Therefore, an interview approach was considered inappropriate.
• Some of the factors affecting high-street consumer traffic flows - such as curves in the road - may have only a subconscious affect on shoppers. For example, asking a shopper why he/she crossed over the road where he/she did - was it perhaps to do with the side they crossed over to being on the inner part of a curve in the road - was likely to precipitate a response such as "I always cross here". Therefore, establishing linkages between topography and consumer high-street traffic flows necessitated observing many people over a long period.

Within each restaurant, observation was used to match the profile of actual users and usage situations with the intended profile. Observation has the key advantage of unobtrusiveness, and at this stage of the research project, more detailed customer information was not required.

More specifically, it was not possible to use observation to uncover the underlying reason for the dining purpose. The most oft-repeated criticism of observation is that it can only uncover patterns of, not motives for, behaviour. In the study under discussion, it was not possible to ascertain, just by using observation, why certain consumers had arrived by taxi or why they were celebrating. To uncover such information, direct questioning of consumers would have been required.

Referring to the classifications of observational studies given earlier, it is possible to classify the study under discussion in several ways. First, in terms of what, or who, was doing the field research, the study was a human, rather than machine-based observational exercise.

Second, with regard to the degree of structure, both structured and unstructured observation were used: unstructured observation was used at the outset of the study in order to discover possible relationships between phenomena - which were then tested in a more structured way using time-sampling.

Third, the research used both covert and overt observation. On the high-street itself, during the time-sampling stages, the researchers stood on the streets with clipboards recording passing consumer traffic: no attempt was made to disguise the fact that systematic observation was taking place. Inside the restaurant, however, the observation was done discretely so as not to make the consumers feel uncomfortable.

Fourth, the observation was conducted in a natural, rather than a contrived setting. The high-street and the restaurants which were analysed were not artificially created especially for the study. They were places where consumers were