Special Session Summary Conceptually Etic: Methodologically Emic?

Citation:

James B. Wiley and Elizabeth Cowley (1998) ,"Special Session Summary Conceptually Etic: Methodologically Emic?", in AP - Asia Pacific Advances in Consumer Research Volume 3, eds. Kineta Hung and Kent B. Monroe, Provo, UT : Association for Consumer Research, Pages: 216-217.

Asia Pacific Advances in Consumer Research Volume 3, 1998      Pages 216-217

SPECIAL SESSION SUMMARY

CONCEPTUALLY ETIC: METHODOLOGICALLY EMIC?

James B. Wiley, University of Western Sydney, Nepean, Australia

Elizabeth Cowley, University of Western Sydney, Nepean, Australia

In the classic paper "That’s Interesting," the political scientist Murray S. Davis explored what it was that made theories and research "interesting." Davis concluded that a common characteristic of interesting work is that the work, in one way or another, denies the reader’s previously held preconceptions. One way to classify preconceptions in cross national/cultural research in consumer behavior is in terms of the emic/etic categories. Emic refers to those issues that are thought to be largely, if not completely, national/cultural specific; etic refers to those that largely, if not completely transcend national cultural boundaries. This session presents ongoing, programmatic research on topics that would seem to be prototypically etic: psychophysics, memory, and the use of numeric response scales. There are methodological issues in each of the areas, however, that can introduce emic affects into what would normally be conceived as largely etic phenomenon.

The first paper, "Pizza: Pi or Squared?" explores the perception of area. There is a vast literature on psychophysics in general and of area perception in particular. However, little research has been done in the specific context of marketing related issues and very little has been done in a cross cultural context. One problem that can arise in cross national/cultural investigations concerns the respondents’ judgements. Most of the previous work in psychology on perceptual biases in area judgements has used ambiguous terms, such as "larger," in measurement. Such terms must be interpreted by subjectsCdoes "larger" mean greater on some single dimension, or on a two-dimensional measure? The present work attempts to reduce the possibility of scale interpretation problems by avoiding ambiguous terms.

The second paper, "Should We Have Confidence in Confidence Scales?" concerns the retrieval of information from memory. While most researchers probably would conceive the process the human brain uses to retrieve information not to be much influenced by cross national/cultural influences, the consumers’ use of the information may depend on factors that are subject to national/cultural influences. For example, consumers’ confidence in the accuracy of what is retrieved may influence their willingness to make use of the information. Furthermore researches’ efforts to measure confidence may be subject to cross national/cultural influences. The objective of this paper is to develop a better understanding of these issues than prevails at the present time.

The third paper, "Cross-Cultural/National Usage of Constant Sum Scales," investigates a measurement procedure that many researchers wuld consider being among the least susceptible to cross national/cultural influences. That is, in this paper, it is the methodology itself is thought to be etic. The paper proposes a variety of ways in which emic influences might intrude on scale use. Data from a variety of sources are analyzed for evidence of cross national/cultural effects on scale usage.

 

PIZZA: PI OR SQUARED?

Robert E. Krider, Hong Kong University of Science & Technology

Priya Raghubir, University of California at Berkeley

Aradhna Krishna, University of Michigan

Pizzas, carpets, real estate, cakes, and retail shelf facings present buyers with purchase decisions involving estimates of area. Given that the same area may be manifested in a large variety of shapes and sizes, how should managers design products and packages, and how should they present the information to the customer? For example, should a pizzeria’s takeout menu give diameter or area information? Should they serve round or square pizzas? Would shape and size considerations affect the manner in which they display their pizzas in their outlets? If perceptual biases in quantity judgement do exist, do they translate to prices consumers are willing to pay, and to the amount they consume? And, do these vary by individual characteristics?

We investigate the antecedents and consequences of biases in judgements of area. Building on the literature in cognitive psychology, we propose a simple psycho physical model of how consumers may simplify the task of estimating area. This model predicts that a single linear dimension may inappropriately dominate the two-dimensional judgement task of area estimation, and that contextual contingencies affect which linear dimension dominates. Typically, the more equal two orthogonal dimensions are, the smaller the areas estimate: a square is seen as smaller than a rectangle of the same area because the longest linear dimension of the rectangle is very salient, and greater than the longest linear dimension of the square. However, a square may be judged either larger or smaller than a circle of the same size, depending on whether the context encourages the use of the diagonal or the side of the square as the single dimension used in the judgement.

We further demonstrate that, for a given shape, people’s estimates of an increase in area lag the actual increase. These results are also explained in terms of the use of a single dimension to judge area. We also show that shape and size biases affect the price people are willing to pay for a product, and the amount they are likely to consume.

Most of the previous work in psychology on perceptual biases in area judgements has used ambiguous terms, such as "larger," in measurement. Such terms must be interpreted by subjectsCdoes "larger" mean greater on some single dimension, or on a two-dimensional measure? Our work attempts to reduce the possibility of scale interpretation problems by avoiding ambiguous terms.

Implications for the manner in which visual cues affect spatial judgements are offered, and managerial implications for pricing, product design, packaging information, shelf layouts, and retailing are suggested.

 

SHOULD WE HAVE CONFIDENCE IN CONFIDENCE SCALES?

Elizabeth Cowley, University of Western Sydney

James B. Wiley, University of Western Sydney

Consumers must retrieve brand information from memory when making purchase decisions. Unfortunately, consumers do not know whether they have accurately retrieved information from memory, instead they rely on a feeling of confidence. Retrieval Confidence, or the confidence one has in their ability to retrieve a particular piece of information from memory, has been considered in three general areas in psychology: eyewitness testimony, signal detection theory and memory storage-retrieval models. In each area, the measures used to capture confidence vary in how thy measure confidence. The research question here is whether the measures also vary in what they are measuring.

Eyewitness testimony. A growing literature investigating eyewitness testimony considers the correlation between retrieval confidence and retrieval accuracy when recalling information from traumatic or violent crime scenes. Confidence is usually measured by: 1] asking study participant to choose between two descriptions of the event, and to rate their confidence on a scale from 50% sure (guessing) to 100% (sure), or 2] asking them to make a number of judgments and then to ask how many they believe they have made correctly (a memory monitoring task).

Signal Detection Theory. SDT is generally used to test theories of perception, however, confidence measures are used to specify decision or judgment criteria when retrieving information from memory. Confidence is measured by showing participants information and asking whether they saw it before or not. Participants rate their confidence on a scale ranging from "most sure new" (1) to "most sure old" (10). Another measure in signal detection theory is to ask for a "old" or "new" response and then to ask for a measure of confidence on a 50% sure to 100% sure scale.

Reconstructive memory. Confidence has also been measured when investigating whether information from memory is retrieved or reconstructed. In this case, confidence is measured with response time, or by asking the respondent whether they remember or know that they were exposed to the information earlier.

Finally, confidence has also been considered as an individual difference variable. Some people are more likely to be confident, or to rate themselves highly on confidence scales. This paper discusses the results of a study measuring accuracy and retrieval confidence using five different confidence scales in two processing conditions: memorize and evaluate. Subjects were also asked to indicate their confidence in their responses to a number of unrelated variables to measure their propensity to record a high degree of confidence. The data will be presented with the objective of better understanding how to measure confidence in memory in a marketing context.

 

CROSS-CULTURAL/NATIONAL USAGE OF CONSTANT SUM SCALES

Gerald Albaum, University of Oregon

Robert Peterson, University of Texas-Austin

One of the most widely used attitudinal measurement scale formats in marketing research is the constant sum scale. In constant-sum methods the respondent is given some number of points and asked to distribute them over a set of alternatives in a way that reflects the relative (and maybe even absolute) magnitude of some attitudinal or behavioral characteristic.

Since most of what is known about the properties of constant sum scales is based on research done in the United States, the question of whether scale formats are equivalent across cultures/nations arises? In other words, the question of whether the constant sum scale approach is etic or emic is unanswered.

This paper examines the use of constant sum scales across cross-cultural/national contexts. The objective of the paper is to provide material relevant to the following types of questions.

What should be the number of points allocated? What effect does the number of items have on usage? Should all items have to be assigned points? Should ties be allowed?

Are certain numbers used more often than others in different cultural/national contexts? Are some numbers never used?

What process is used by respondents to assign numbers from the perspective of the order in which items are assigned numbers?

At the present time data is available from the U.S., Hong Kong, and Korea. We anticipate that additional studies will be done in areas like China, Australia, Indonesia an one or more countries of Europe.

----------------------------------------