1 / 42

Small Sample Quantitative Techniques

Small Sample Quantitative Techniques. COMM 3710 April 29, 2002 Tim Larson. Objectives for this session. Review of large sample quantitative techniques Introduction to scales and measurement Introduction to small sample quantitative techniques in business-to-business (B2B) marketing research

roth-orr
Download Presentation

Small Sample Quantitative Techniques

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Small Sample Quantitative Techniques COMM 3710 April 29, 2002 Tim Larson

  2. Objectives for this session • Review of large sample quantitative techniques • Introduction to scales and measurement • Introduction to small sample quantitative techniques in business-to-business (B2B) marketing research • Two small sample quantitative techniques examined • Conjoint Analysis • Multidimensional Scaling (perceptual mapping)

  3. Review of Large Sample Quantitative Methods • Survey research/Public Opinion Polling • Goal is to estimate, with as much precision as possible, the percentage of your sample that does this or that or believes one way or another. • Cost factors are fixed to sample size •Sampling error formula SE=1.95 √ (P) x (1-P) N Size of sample is based on how much diversity or homogeneity there is in the population or market, or how many ways you will want to cross-section (segment) the sample.

  4. Levels of data or measurement • Nominal (chi-square) • Classification of attributes into mutually exclusive categories that have no quantitative relationship (Yes/No, male/female) • Ratio • Quantify attributes by scaling them along a continuum • Usually an absolute zero (distance, weight, purchases)) • Interval measurements (t-test) • Variant of of the ratio scale • No real zero --arbitrary zero points (temperature, an index) • Ordinal measurements • Categorical in nature and different from nominal • Rank ordered (Education, telescoped ratio data)

  5. Review of Criteria for Measurement Quality • There are many ways to operationalize a variable or concept. How do you know when it is a quality measurement? • Three criteria to consider • Precision (the more the better) • Validity (face,content, construct) • Reliability (consistency and dependability)

  6. Measurement and Scales • Most scales are probably at best interval-level measurement because the end point that anchors the continuum is somewhat arbitrary and is not necessarily standardized across every individual • No absolute zero and it may not mean the same to everyone • Intervals between numbers cannot be assumed to be equal.

  7. Scales and Use of Sophisticated Analytical Procedures • If you can assume that the differences in the way people standardize their perceptions and the widths of the intervals will average themselves out --in other words, those who are more stringent balance out those who are less stringent-- then you can treat scales as though they were ratio scales and calculate averages using the more powerful analytical quantitative statistical procedures. • For most marketing and B2B market research purposes, attitude scales are used as though they are ratio- and equal-interval scales.

  8. Small Sample Quantitative Techniques Two choices when it comes to conducting research Collect a lot of information from a few individuals Collect less information across more individuals. Rule of Thumb The more information you collect from any single individual, the fewer the number of cases you will need in order to achieve a stable sample estimate.

  9. Small Sample Quantitative Techniques • Used in B2B research • Large sample quantitative survey research is not always feasible for business-to-business marketers. • Small sample qualitative techniques not always sufficient for business decisions. • Two small sample quantitative techniques explored today work incredibly well with small samples & yield incredibly useful, quantifiable information for a variety of B2B and other marketing research. • CONJOINT ANALYSIS • MULTIDIMENSIONAL SCALING

  10. CONJOINT ANALYSIS & MULTIDIMENSIONAL SCALING Rule of Thumb Again: •The more information you collect from any single individual, the fewer the number of cases you will need in order to achieve a stable sample estimate. • Both CA and MDS apply this rule of thumb by measuring a phenomenon systematically in every possible light; hence, the statistical stability without the large, expensive samples.

  11. Conjoint Analysis(CA) • A technique that quantifies people’s preferences or priorities when faced with the task of evaluating a set of products or services and choosing the most preferred alternative. • Often labeled trade-off analysis • Parallels a purchase situation • Product purchase example • Computer Brands and Descriptors: HD memory, monitor size, RAM, OS, service, software, peripherals

  12. CA Applications • Can effectively answer some of the following questions • What value does my customer or the market put on the various features of my product? • How could I improve my product or service? Or how would I alter the competitiveness of my product or service if I change this or that? • If my competitor changed this or that aspect, how might the market respond? • What is the price elasticity for my brand? • How can I segment the market based on differences in what people think is important?

  13. CA Best Uses • Serves as foundation for a segmentation scheme. • Good for products or services that are multi-attribute in nature. • Especially good in B2B research • Most appropriate in situations where the decision making is dominated by a single individual because the CA research technique is conducted among individuals. • Helps you customize stakeholder messages. • See Exhibit 5.3 CA Application

  14. Conjoint Basics • Involves presenting respondents with alternative choice situations and having them rank the alternatives. • Computer model “decomposes” these preferences by analyzing what features have been consistently present or traded off in the way choices were rank-ordered. • Output of model is a set of numerical values associated with every feature and feature level, which portrays the relative importance of each to the individual.

  15. FAX Machine Example • Utilities or “”PARTWORTH” scores • Each score shows what a particular feature level is worth, relative to others, in in its contribution to an individual’s overall preference choice. • Higher scores indicate more importance. • By adding together the utilities for each feature of an actual fax machine, you theoretically have the overall preference value for that particular machine. • By comparing one fax machine’s total utility to that of another fax machine would tell you the relative preference of one over the other.

  16. FAX Example •Price is the most important feature in the purchase decision for this one individual, and the lowest price of $499 is the most preferred price point. •Print speed is next in importance, and the highest speed of 10ppm speed is the most preferred speed point. •Color and brand name have less impact on preference because the utilities are much lower for these features •See Exhibit 5.4 Hypothetical Conjoint Output for FAX Machine

  17. 5.4 Hypothetical Conjoint Output for FAX Machine Partsworth Difference Relative Weight Paper cutter Yes .6 No .01 .59 18% Telephone handset Yes .5 No .1 .4 12 Answering machine Yes .8 No .5 .3 9 Print speed 5ppm .3 8ppm .9 10ppm 1.0 .7 21 Brand name Sharp .4 Panasonic .5 Brother .5 HP .6 .2 6 Color Black .1 Putty/Beige .2 Gray .2 .1 6 Price $499 1.5 $699 .9 $899 .5 1.031 Total Utility 3.29 100%

  18. FAX Example Partworth Diagrams for Hypothetical FAX Example • A change in price from $499 to $699 greatly reduces preference, and with a price point of $899, preference dips considerably lower. • Difference between highest and lowest utility is an indication of the impact of this feature on overall preference. • Note that the “elasticity” of the color feature is small and flat while price elasticity is steep. • See Exhibits 5.5 & 5.6

  19. 5.5 Partworth Diagrams for Hypothetical FAX Example Preference Utility Preference Utility Preference Utility Black Beige Gray Color Yes No Paper Cutter Yes No Answering Machine Preference Utility Preference Utility Preference Utility Sharp Panasonic Brother Hp Brand Name Yes No Telephone Handset $499 $699 $899 Price Preference Utility 5ppm 8ppm 10ppm Print Speed

  20. FAX Example Observations • Differences among individuals allows for the application of CA as a market segmentation and message customization tool. (Price with price, etc.) • Important first step in CA is to determine the appropriate features to test for. (Qualitative research) • Second step is to determine an appropriate number of realistic levels for each feature.

  21. Trade-Off CA or “Two -Factor Approach” • Respondents are presented with a series of combinations based only on pairs of features • Requires respondent to make choices about every combination of features and levels. • This intense data collection adds to the stability of the utilities for even small samples. • Works best when features do not interact with one another, or in others words, the preference or utility of one variable compared to another does not depend on the circumstances of a third feature. • Computer model analyzes the data to get utility values • See Exhibit 5.8 Trade-Off Conjoint Matrices

  22. 5.8 Trade-Off Conjoint Matrices Telephone Price Price

  23. Full Profile Conjoint Analysis or “Multi Factor Approach” • Respondents are presented with a complete profile of alternative products, each alternative profiled in terms of information for each and every feature of interest. • Rather than pairs of features, in full profile CA the individual is confronted with many alternatives for which he or she must consider all the various features before indicating a rank-ordered preference among them. • Using computers to model and crunch the data, full-profile CA is the dominant method used today. • See Exhibit 5.9 - Product Descriptions in a Full Profile CA

  24. Product Description in a Full Profile Conjoint Example Card Print Speed Handset Price 5ppm Yes $499 5ppm Yes $699 5ppm Yes $899 4 5ppm No $499 5 5ppm No $699 6 5ppm No $899 7 8ppm Yes $499 8 8ppm Yes $699 9 8ppm Yes $899 10 8ppm No $499 11 8ppm No $699 12 8ppm No $899 13 10ppm Yes $499 14 10ppm Yes $699 15 10ppm Yes $899 16 10ppm No $499 17 10ppm No $699 18 10ppm No $899

  25. Multidimensional Scaling (MDS) and Preference Mapping Small-sample quantitative approaches • MDS quantifies the perceptions and images that people have about a market, a brand, or any concept by assessing how similar or dissimilar people perceive them to be from various other related concepts. • Often called perceptual mapping because, in addition to deriving the perceptual “distances” among brands or concepts, the program will plot these concepts as points in a geometric space so that they are visually shown as a picture --a “perceptual map.”

  26. Multidimensional Scaling (MDS) and Preference Mapping •Preference mapping is a related procedure that will measure people’s preferences for the same concepts as used for the MDS, and derive from this the most preferred point, mapping this “ideal” point back into the perceptual map. •Helps marketer not only know how their brand is currently perceived, but also how it falls short of the ideal. •See Exhibit 5.11 -Multidimensional Scaling “Perceptual Map”

  27. 5.11 Multidimensional Scaling “Perceptual Map” A c b G •IDEAL POINT E F H I D

  28. MDS Applications • Some questions MDS can answer • What is my brand/company/service’s image or perceived position in the marketplace? • Who are my key competitors in the customer’s eyes? • What are my perceived strengths and weaknesses? • How do I best position my product? • What are my best message strategies? • Are there different perceptual segments in the market? • How has my image changed over time? • Are there untapped opportunities in the marketplace that are as yet unmet by the competition?

  29. 5.12 Example MDS Concepts for the Copier Category Brands 1. Canon 2. Kodak 3. Konica 4. Minolta 5. Mita 6. Panasonic 7. Ricoh 8. Sharp 9. Toshiba 10. Xerox Brand and Descriptors 1. Affordable 2. Aloof 3. American 4. Approachable 5. Bureaucratic 6. Conservative 7. High Quality 8. Innovative 9. Intimidating 10. Me 11. Prestigious 12. Reliable 13. Canon 14. Sharp 15. Xerox

  30. MDS Procedure and Interpretation • Begin with selecting the concepts to be included in the map, or those brands, adjective descriptors, and other key concepts and/or phrases that define the topic or market under study. • See Exhibit 5.13 Examples of MDS Similarity Measures • Obtain “similarities” or “proximities” data on each concept in relation to every other concept. • See Exhibits 5.14, 5.15, 5.16 & 5.17 • Examples of MDS Similarity Measures • Copier maps with Brands and with Descriptors • Perceptual map with Preference Mapping

  31. 5.15 Copier MDS Map with Descriptors 5.14 Copier MDS Map with Brands Only Intimidating High Quality Prestigious Innovative Bureaucratic Xerox Aloof Conservative Reliable Mit a Lanier Xerox Sharp Canon Approachable My Business Affordable American Ricoh Konica Toshiba Minolta Panasonic Sharp Canon Kodak 5.16 Perceptual Map with Preference Mapping 5.17 Message Strategies for Copier Example Copier Specialist Bureaucratic XEROX Aloof Conservative Intimidating High Quality Prestigious Innovative Mit a Lanier Xerox Ideal Point American Foreign American Sharp Canon Approachable MY BUSINESS Affordable Ricoh Konica Toshiba Minolta Panasonic Sharp Canon Kodak General Electronics Company

  32. Developing Message Strategies • Identify key concepts that might be used in creating alternative message strategies with the purpose of moving your brand closer to the ideal point or “My Business.” • Alternatively, you could strategize how to move the ideal point or “My Business” in the direction of your brand and its image. •Easier to change one’s brand than to alter respondent’s perceptions of who they are and what they prefer. •See Exhibit 5.17- Message Strategies for Copier Example • Emphasize reliability, approachable, American, affordability, and innovativeness

  33. Metric and Non-metric Measurement • Metric measures are those that use higher-level measurement, such as ratio scaling techniques that require respondents to estimate actual distance between concepts with a number, or to mark on a line scale that can be directly related to a number. See Exhibit 5.13 - measuring scales A, B, and C • Non-metric measures are those that do not require a direct estimate of perceptual distance. Respondents are asked only to make rank-order judgments about differences between the pairs of concepts. See Exhibit 5.13 parts D and E • Both have advocates and detractors

  34. 5.13 Examples of MDS Similarity Measures A. How similar or different are Canon and Xerox Very Similar ------------------------- Very Different Cannon and American Very Similar-------------------------- Very Different B. How similar or different are Canon and Xerox Very Similar 1 2 3 4 5 6 7 8 9 10 Very Different Canon and American Very Similar 1 2 3 4 5 6 7 8 9 10 Very Different On a scale from 0 to 100, how would you rate the similarity of the following pairs of concepts? If you think two concepts are very similar of seem to go together, or are often associated with each other, then write a small number. On the other hand, if you think the concepts are very different, do not seem to go together, or are seldom associated with one another, then write a large number. Thus a zero means that you believe the two concepts are identical, while a 100 means that you believe the two concepts are as different from each other as the possibly could be. D.Which pairs of words seem to be more similar or seem to go together Better? Choose 1. 1. Canon and Xerox Canon and Sharp 2.Canon and American Canon and M Business E. Please rank-order the following concept pairs based on how similar the word pairs seem to be. The pair which you think seems the most alike, rank first, the pair you think seems next most similar, rank second, and so on. The pair which is the most different in your opinion should be ranked last. Canon and Xerox Canon and American Canon and Sharp Canon and My Business

  35. Perceptual Mapping and Qualitative Research • Software is available that will computer-analyze interview transcripts and, by content analyzing the words and word sequences that are used repeatedly within the interview, quantify the strength of word associations in order to create a map of those conversations. • The technique is based on the principle of a “neural network” and is patterned after the way your brain works. • See the Hypothetical Beverage Category from a previous lecture.

  36. Computerized conjoint and MDS Models • Low cost computing makes CA & MDS small-sample quantitative techniques feasible. • Conjoint software by SPSS and SAS • MDS Software by SPSS, SAS, Sawtooth, PC-MDS • Galileo System available through Terra Research and Computing. offers an MDS package complete with software to develop message strategies.

  37. CA & MDS Software Suppliers SPSS Inc. 444 N. Michigan Avenue Chicago, IL 60611 800 543-6609 SAS SAS Campus Drive Cary, NC 27513 919 677-8000 Sawtooth Software 1007 Church Street, Suite 402 Evanston, Il 60201 708 866-0870

  38. CA & MDS Software Suppliers PC-MDS Attn: Scott Smith BYU Provo, UT 84602 801 378-5569 Terra Research & Computing 261 E. Maple Birmingham, MI 48009 810 258-9657

  39. Book Source Martin P. and Tamara S. Block, Business-to-Business Market Research: Identifying, Qualifying and Understanding your Customers, Chicago, Probus Publishing, 1995.

  40. Assignment 4.2 Cell Phone Trade-Off Conjoint Matrices COMM 3710 April 29, 2002 Larson/Carlson Names:

  41. Assignment 4.2: MDS Concepts for Utah Colleges Brands Snow College SLCC Westminster U of U BYU UVCC USU U of Phoenix Research Brand and Descriptors 1. Affordable 2. Aloof 3. Approachable 5. Bureaucratic 6. Conservative 7. High Quality 8. Innovative 9. Intimidating 10. IDEAL COLLEGE 11. Prestigious 12. Caring 14. High Tech 15.Student Focused Liberal Conservative Teaching

More Related