Measuring Response – Scaling Techniques

In this article we are highlighting some aspects of Market research, (MR) data collection and scaling techniques. MR is required for numerous applications in industry and business particularly in the areas of Marketing, Finance and even Human resources. First the requirement or problem has to be identified and then data needs to be collected from trained researchers and results correlated based on the requirements.

MR is being used to help create and enhance brand equity. This is a new role and de3cidedly different from the conventional one where it was used for just studying buyer behavior of for conducting feasibility studies.

Primary data is original data collected by the researcher first hand. If the secondary data is found inadequate, the researcher goes for secondary data.

Research instrument for collecting primary data: Basically the research instrument used for collecting primary data from respondents is either a

1. Questionnaire, or a
2. Mechanical /electronic device

Telephone interview: This method is extensively used in developed countries. In India, the scope for employing this method is relatively less. It can only be used for certain types of products.

Personal interview: The third method is the personal interview. This method is extensively used in surveys. The, personal interview method enables better control of the sample and ensures answer from the respondents. It also provides for a tactful approach to the respondent since it is based on person-to-person talk. But this method is generally more expensive and time consuming.

Questionnaire: Questionnaire development is a critical part of the primary data collection job. It is an art that calls for a lot of expertise and resourcefulness on the part of the researcher. The questionnaire should be so structured as to collect all relevant information. It often sets the framework as well as the tone of the survey. The research data is built up on the framework of the questionnaire. If the questionnaire is faulty, it will generate incorrect information and no amount of analysis and interpretation can set it right. The choice of words in the questionnaire should also be appropriate and should be unequivocally understood by all respondents.

Fortunately, appropriate scaling techniques are available for measuring qualitative responses. The merit of these techniques is that with the use of scaled answers/responses, qualitative information gets recorded pin a quantifiable and measurable form. Let us consider a few examples of scaling techniques.

Ordinal scale: These scales order or rank the item with reference to a particular attribute. For example, bath soaps can be ranked with reference to the attribute of ‘freshness’ Supposing that there are five different brands of bath soaps and a respondent ranks them based on ‘freshness’ he will assign rank 1 to the brand having maximum freshness according to home rank 2 to the second best brand and so on. When this is done, an ordinal scale results Instead of numbers, letters or symbols too can be used to constitute an ordinal scale. If there are four different lots of rice and if they are ordered based on quality, they can be ranked into grade A, grade B grade C and grade D. It is again an ordinal scale.

The semantic differential: The semantic differential, developed by Osgood and his associates, probes, the intensity and content of a respondent’s attitude towards, say a company’s image or a brand’s image. The evaluation is made using a scale of adjectives, which are polar opposites, ranging from one extreme to opposite extreme, separated by seven equal intervals.

The Likert scale also known as ‘the method of summated ratings’ is a five point scale ranging from ‘agreement’ to ‘strong disagreement’. The respondent is given a statement about a topic and he records the degree of his agreement or disagreement with the statement in the five point scale.

Multi dimensional scaling: The scales discussed above are uni-dimensional scales. They measure attitudes along a single dimension. Often, a person’s decision to buy a particular product is the outcome of not one, but a variety of stimuli — the product features the company image, the advertising message, etc. Techniques have been developed to scale and measure such multi dimensional stimuli. Grouped under the broad term of ‘multi-dimensional scaling’ (MDS) these techniques have helped improve the precision of psychological scaling MDS techniques are now finding increasing application in a variety of marketing research situations.