Experiments in Marketing Research

Résumé du document

An experiment involves manipulating or changing a variable or variables to see how the changes affect some other variable or variables, while eliminating or controlling the effects of other extraneous or confounding variables. The variable manipulated is called the independent variable. The variable upon which the possible effect of the change is measured is called the dependent variable. An experimental design is a procedure for ensuring that a change in a dependent variable may be attributed only to a change in an independent variable. Essentially, experimental design arrangements are intended to ensure that nothing other than the independent variable could have caused the observed changes in the dependent variable.

Extraits

[...] This allows demographic and lifestyle information on panel members to be related to the purchase or non- purchase of targeted goods. Advantages are speed, confidentiality, lower cost and more detailed information about purchasers. The main dis-advantage is the possible non representative character of the panel members. These are usually recruited in a small number of areas believed to be typical of the population at large. Simulated Test Markets:Here a sample of respondents screened for product usage are recruited in a shopping mall or some similar location. [...]

[...] The selected individuals are given an opportunity to buy the new product in a real life situation or in some laboratory type situation. Those who agree to purchase the product are interviewed on how they evaluate the product and whether or not they intend to purchase the product again. The information on re-purchase is combined with anticipated promotional spend and the firm's distribution intentions and using computer based mathematical models, sales figures are predicted. [...]

[...] The variable manipulated is called the independent variable. The variable upon which the possible effect of the change is measured is called the dependent variable. Experimental Design An experimental design is a procedure for ensuring that a change in a dependent variable may be attributed only to a change in an independent variable. Essentially experimental design arrangements are intended to ensure that nothing other than the independent variable could have caused the observed changes in the dependent variable. Notation By convention certain symbols are used to refer to various features of experiments. [...]

[...] Marketing experiments can be carried out over quite a long time period such as several weeks. If individual cases or units (be they people or shops or some other kind of unit) are unable or unwilling to continue to participate in the experiment and drop out, control and experimental groups may no longer be equivalent. This may distort the supposed effect of the independent variable on the dependent variable. The above six threats to validity are referred to as Internal Validity Concerns in that they relate to whether or not the independent variable can be truly said to have caused the observed change in the dependent variable. [...]

[...] There are several types of Test Marketing experiments: Standard Test Market: A product is tested through the company's normal distribution channels. The advantage of the Standard Test market is that the product is tested in realistic market conditions, little different from actual market situations. However, there are several dis-advantages: Time consuming and very expensive; can take several months or more, depending on the nature of the product tested.(b) Conducting the Test in actual market situations means that competitors are rapidly aware of the new product or changes to an existing product. [...]