Sunday, 4 September 2011

Factor Analysis - A Rather Lengthy Anecdote of Theory

Factor analysis is a statistical method used to unearth the underlying variables, or factors, that drive change in a group of observable attributes. An analyst might want to explain what motivates respondents to answer a survey in particular ways. He or she would note that questions about service, product quality, and other discrete topics correlate. The analyst’s job is to figure out how many, and what kinds, of factors influence these correlations –and to form a plausible theory as to what each factor means.
Factor Analysis is used to understand the underlying motives of consumers who buy a product category or a brand i.e. to find out the key factors responsible for purchase of a brand.
There are basically two types of factor analysis: exploratory and confirmatory.
Exploratory factor analysis (EFA) attempts to discover the nature of the constructs influencing a set of responses.
Usage of exploratory factor analysis in marketing:
           customer satisfaction surveys
           measuring service quality
           personality tests
           image surveys
           identifying market segments
 Confirmatory factor analysis (CFA) tests whether a specified set of constructs is influencing responses in a predicted way
Companies facing an overwhelming amount of marketing research information, particularly survey data such as customer satisfaction results, attitudes and usage data, or employee feedback, can use factor analysis to reduce the data into much more manageable and meaningful intelligence. Factor analysis is foremost a data reduction technique: It takes the entire data set and whittles the individual variables down to underlying factors. Factor analysis can analyze "interdependence" by tying together variables highly correlated with one another to reveal patterns flowing just beneath the surface, allowing new insights into the data under study.
For example, say you have a battery of 100 statements that can be used to describe a product x. You could just present a ranked list of statements or you could FA it and determine what underlining factors really exist that explain as much of those 100 statements in as few factors as possible. It's obviously easier to describe a product in 5-6 factors rather than 100. So, it we were talking about soft drinks there could be a "sweetness" factor, a "colour" one, "taste" and so on.
Factors in Marketing
The variables that affect marketing include, but are not limited to, product, product size, product color, packaging, weight, etc.
Factor analysis in marketing studies how the product variables affect the customer's perception and/or purchase of the product. The goal is to create the ideal attributes to create customer purchases.
Information Gathering:
The information for factor analysis in marketing can be gathered through surveys and focus groups.
In essence, marketing factor analysis is changing one marketing variable to see what affect, if any, the change has on the outcome. The change in sales also affects the bottom line of the company, so factor analysis in marketing helps companies determine which marketing efforts it should pursue, which efforts need work and which marketing efforts may meet the cutting room floor.
Factor analysis in marketing requires an evaluation of how changing one marketing point, such as price, changes the sales of the product. In order to measure how the factor changes the sales, it requires that only one marketing variable is changed at a time in order to measure the relationship between the variables and the outcome. In marketing, changing one variable can be significant because it may cause an increase or decrease the sales of the product.
The objective of factor analysis:
The objective of Factor Analysis is to find underlying "factors" which are fewer in number than the original set of variables, but would be linear combinations of some of the original variables. It is useful to use the factors instead of the original variables for further marketing mix decisions.
Where it can be applied in Industry:
Managers of consumer electronics companies, banks, truck operators, retail stores, or automobile companies can all benefit from the use of factor analysis to understand customer motives in the form of broad underlying factors instead of numerous variables which may actually be measuring the same underlying factors.
Using factor analysis helps a marketer to understand his customers and plan his marketing strategy. For instance, a car manufacturer may want to release a small car into the market and understand the factors responsible for purchase of a small car, a cigarette manufacturer may need to determine which variables his potential customers think about when they consider his product, an airline company could want to find out the factors responsible for selection of its airline against its competitor's airline. All these people would need factor analysis to get them the information they require.
When running factor analysis and examining results, keep in mind a few rules of thumb:
1. Be sure the statistician is comfortable with the sample size (usually 50 is a safe minimum).
2. You can keep factors explaining more of the response variance than an average single variable explains (an eigenvalue higher than 1.0).
3. Factor analysis works best if variables are correlated with one another. A Measure of Statistical Adequacy (MSA) of less than .5 demonstrates the correlation between variables may be too weak for factor analysis to work.
4. Usually, a variable needs a factor loading with a number of .4 or more (1.0 would be a perfect correlation) to be included in a factor, so don't be surprised if some variables get left out, because they don't correlate with anything.
5. Results of factors are listed in a descending order: the first factor explains the most variance observed in the variables
1.         Advertising
           to better understand media habits of various customers
2.         Pricing
           to identify the characteristics of price-sensitive and prestige-sensitive customers
3.         Product
           to identify brand attributes that influence consumer choice
4.         Distribution
           to better understand channel selection criteria among distribution channel members
5.         Reduce a body of data to a few key dimensions, making insights easier to access.
6.         Explain how a collection of dependent variables relate to each other, even as you examine in detail the relationship of each to an independent variable.
7.         Use the relationship between attributes to create perceptual maps and other visual aids that explain data to end users.
8.         To cluster variables or individuals for classification and segmentation. To classify variables, the R-type factor analysis is used; to classify individuals, the Q-type factor analysis is done
9.         Identifying the underlying variables on which to group Identifying the underlying variables on which to group the customers ( new car buyers may be grouped the customers ( new car buyers may be grouped based on the relative emphasis they place on based on the relative emphasis they place on economy, convenience, performance, comfort and economy, convenience, performance, comfort and luxury.) luxury.)
10.       Determining the brand attributes that influence Determining the brand attributes that influence consumer choice. (toothpaste brands might be consumer choice. (toothpaste brands might be evaluated in terms of protection against cavities, evaluated in terms of protection against cavities, whiteness of teeth, taste, fresh breath and price) whiteness of teeth, taste, fresh breath and price)
11.       Understand the media consumption habits of the Understand the media consumption habits of the target market. The users of ready to eat food may be target market. The users of ready to eat food may be heavy viewers of cable TV, see a lot more movies, and heavy viewers of cable TV, see a lot more movies, and listen to western music listen to western music
12.       Identifying the characteristics of price Identifying the characteristics of price- -sensitive sensitive consumers (these consumers might be methodical, consumers (these consumers might be methodical, economy minded and home centered) economy minded and home centered
An end user can benefit from the results of factor analysis, as well.
           Segment customers not just along demographic lines but in terms of which factors shape their choices (are they sensitive to product packaging, ease of use, convenience, etc).
           Position a product or service according to which factors influence targets’ decisions, so as to fully utilize the available sales opportunities.
           Gauge how one attribute enhances or detracts from another. Create an offering whose elements work together in an optimum whole, driving sales and securing loyalty.


No comments:

Post a Comment