Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Chapter 10: Measurement and Attitude Scaling
Transcript of Chapter 10: Measurement and Attitude Scaling
Mathematical and Statistical Analysis of Scales
Reliable and Valid Index Measures
Indexes and composites:
Reliability represents how precise a measure is in that the different attempts at measuring the same thing converge on the same point. Accuracy deals more with how a measure assesses the intended concept. Validity is the accuracy of a measure or the extent to which a score truthfully represents a concept.Researchers need to know if their measures are valid.The question of validity expresses the researcher’s concern with accurate measurement.Validity addresses the problem of whether a measure (for example, an attitude measure used in marketing) indeed measures what it is supposed to measure. When a measure lacks validity, any conclusions based on that measure are also likely to be faulty.
A researcher has to know what to measure before knowing how to measure something.The prob- lem definition process should suggest the concepts that must be measured. A concept can be thought of as a generalized idea that represents something of meaning. Concepts such as age, sex, education, and number of children are relatively concrete having relatively unambiguous meanings. They present few problems in either definition or measurement. Other concepts are more abstract. Concepts such as loyalty, personality, channel power, trust, corporate culture, customer satisfaction, value, and so on are more difficult to both define and measure
What Is Measurement
Is the process of describing some property of a phenomenon, usually by assigning numbers, in a reliable and valid way. The numbers convey information about the property being measured. When numbers are used, the researcher must have a rule for assigning a number to an observation in a way that provides an accurate description.
Measurement can be illustrated by thinking about the way instructors assign students’ grades. A grade represents a student’s performance in a class. Students with higher performance should receive a different grade than do students with lower performance
A student can be assigned a letter corresponding to his/her performance as is typical of U.S.- based grading systems.
a. A — Represents excellent performance
b. B — Represents good performance
c. C — Represents average performance
d. D — Represents poor performance
e. F — Represents failing performance
A student can be assigned a number from 1 to 20, which is the system more typically used in France
a. 20 — Represents outstanding performance
b. 11–20 — Represent differing degrees of passing performance
c. Below 11 — Represent failing performance
Any other examples of grading scales??????
Researchers measure concepts through a process known as
.This process involves identifying scales that correspond to properties of the concept.
, just as a scale you may use to check your weight, provide a range of values that correspond to different characteristics in the concept being measured. In other words, scales provide
that indicate that a certain value on a scale corresponds to some true value of a concept. Hopefully, they do this in a truthful way.
Researchers use the variance in concepts to make meaningful diagnoses. Therefore, when we defined variables, we really were suggesting that variables capture different values of a concept. Scales capture a concept’s variance and, as such, the scales provide the researcher’s variables.
term used for concepts that are measured with multiple variables
Operational definitions translate conceptual definitions into measurement scales. An operational definition is like a manual of instructions or a recipe: even the truth of a statement like “Gaston likes seafood gumbo” depends on the recipe and the ingredients. Different instructions lead to different results
Levels of Scale Measurement
Marketing researchers use many scales or numbering systems. Not all scales capture the same rich- ness in a measure. Not all concepts require a rich measure. But, all measures can be classified based on the way they represent distinctions between observations of the
variable being captured.The four levels or types of scale measure-
ment are nominal, ordinal, interval, and ratio level scales. Tradition- ally, the level of scale measurement is seen as important because it determines the mathematical comparisons that are allowable. Each of the four scale levels offers the researcher progressively more power in analyzing and testing the validity of a scale.
A nominal scale assigns a value to an object for identifica- tion or classification purposes. The value can be a number, but does not have to be a number, because no quantities are being represented. In this sense, a nominal scale is truly a qualitative scale. Nominal scales are extremely useful even though some may consider them elementary.
Ordinal scales have nominal properties, but they also allow things to be arranged based on how much of some concept they possess. In other words, an ordinal scale is a ranking scale. When a professor assigns an A, B, C, D, or F to a student at the end of the semester, he or she is using an ordinal scale. Research participants often are asked to rank order things based on preference. So, preference is the concept, and the ordinal scale lists the options from most to least pre- ferred, or vice versa. In this sense, ordinal scales are somewhat arbi- trary, but not nearly as arbitrary as a nominal scale. Five objects can be ranked from 1–5 (least pre- ferred to most preferred) or 1–5 (most preferred to least preferred) with no loss of meaning.
have both nominal and ordinal properties, but they also capture information about differences in quantities of a concept. So, not only would a sales manager know that a particular salesperson outperformed a colleague, but the manager would know by how much. If a professor assigns grades to term papers using a numbering system ranging from 1.0–20.0, not only does the scale represent the fact that a student with a 16.0 outperformed a student with a 12.0, but the scale would show by how much (4.0)
represent the highest form of measurement in that they have all the properties of inter- val scales with the additional attribute of representing absolute quantities. Interval scales represent only relative meaning whereas ratio scales represent absolute meaning. In other words, ratio scales provide iconic measurement. Zero, therefore, has meaning in that it represents an absence of some concept.
are those that take on only one of a finite number of values. A discrete scale is most often used to represent a classificatory variable. Therefore, discrete scales do not represent intensity of measures, only membership. Common discrete scales include any yes-or-no response, matching, color choices, or practically any scale that involves selecting from among a small num- ber of categories.
Are those assigning values anywhere along some scale range in a place that corresponds to the intensity of some concept. Ratio measures are continuous measures. Thus, when we measure sales for each salesperson using the dollar amount sold, we are assigning con- tinuous measures. A number line could be constructed ranging from the least amount sold to the most and a spot on the line would correspond exactly to a salesperson’s performance
An index measure assigns a value based on how characteristic an observation is of the thing being measured. Indexes often are formed by putting several variables together. For example, a social class index is based on three weighted variables: income, occupation, and education. Composite measures also assign a value based on a mathematical derivation of multiple vari- ables. For example, restaurant satisfaction may be measured by combining questions such as “How satisfied are you with your restaurant experience today? How pleased are you with your visit to our restaurant? How satisfied are you with the overall service quality provided today?” For most practical applications, composite measures and indexes are computed in the same way. However, composite measures are distinguished from index measures in that the composite’s indicators should be both theoretically and statistically related to each other.
Computing Scale Values
A summated scale is created by simply summing the response to each item making up the composite measure. In this case, the consumer would have a trust score of 13 based on responses to five items. A researcher may sometimes choose to average the scores rather than summing them. The advantage to this is that the composite measure is expressed on the same scale as are the items that make it up. Sometimes, a response may need to be reverse-coded before computing a summated or aver- aged scale value. Reverse coding means that the value assigned for a response is treated oppositely from the other items
Reliability is an indicator of a measure’s internal consistency. Consistency is the key to under- standing reliability. A measure is reliable when different attempts at measuring something con- verge on the same result. If a professor’s marketing research tests are reliable, a student should tend toward consistent scores on all tests. In other words, a student that makes an 80 on the first test should make scores close to 80 on all subsequent tests. If it is difficult to predict what stu- dents would make on a test by examining their previous test scores, the tests probably lack reliability.
performed by taking half the items from a scale (for example, odd-numbered items) and checking them against the results from the other half (even-numbered items). The two scale halves should correlate highly. They should also produce similar scores. However, multiple techniques exist for estimating scale reliability.
the most commonly applied estimate of a composite scale’s reliability. Coefficient estimates internal consistency by computing the average of all possible split-half reli- abilities for a multiple-item scale. The coefficient demonstrates whether or not the different items converge.
The three basic aspects of validity are face or content validity, criterion validity, and construct validity
refers to the subjective agreement among professionals that a scale logically reflects the concept being measured. Simply, do the test items make sense given a concept’s definition?
addresses the question, “Does my measure correlate with measures of the similar concepts or known quantities?” Criterion validity may be classified as either concur- rent validity or predictive validity depending on the time sequence in which the new measurement scale and the criterion measure are correlated
exists when a measure reliably measures and truthfully represents a unique concept. Construct validity consists of several components, including:
• Face or content validity
• Convergent validity
• Criterion or validity
• Discriminant validity
Attitudinal Rating Scales
Researchers face a wide variety of choices in measuring attitudinal concepts. One reason for the variety is that no complete consensus exists over just what constitutes an attitude or an attitudinal variable. Researchers generally argue that the affective, cognitive, and behavioral components of an attitude can be measured by different means.
This category scale measures attitude with greater sensitivity than a two-point response scale. By having more choices, the potential exists to provide more information. However, a researcher will create measurement error if he/she uses a category scale for something that is truly bipolar (yes/ no, female/male, member/non-member, and so on).
The Likert scale may well be the most commonly applied scale format in marketing research. Likert scales are simple to administer and understand. Likert scales were developed by and named after Rensis Likert, a 20th century social scientist. With a Likert scale, respondents indicate their attitudes by checking how strongly they agree or disagree with carefully constructed statements. The scale results reveal the respondent’s attitude ranging from very positive to very negative. Indi- viduals generally choose from multiple response alternatives such as, strongly agree, agree, neutral, disagree, and strongly disagree. Researchers commonly employ five choices although they also often use six, seven or even more response points.
A constant-sum scale demands that respondents divide points among several attributes to indicate their relative importance. Suppose United Parcel Service (UPS) wishes to determine the impor- tance of delivery attributes such as accurate invoicing, delivery as promised, and price to organiza- tions that use its service in business-to-business marketin
A graphic rating scale presents respondents with a graphic continuum. The respondents are allowed to choose any point on the continuum to indicate their attitude
Graphic Rating Scale
Attitudes and Intentions
Multi-attribute Attitude Score
Attitudes are modeled with a multi-attribute approach by taking belief scores assessed with some type of rating scale like those described and multiplying each belief score by an evaluation also supplied using some type of rating scale, and then summing each resulting product. For instance, a series of Likert statements might assess a respondent’s beliefs about the reliability, price, service, and styling of a Honda Fit.
According to reasoned action theory, people form intentions consistent with the multi-attribute attitude score. Intentions represent the behavioral expectations of an individual toward an attitu- dinal object. Typically, the component of interest to marketers is a buying intention, a tendency to seek additional information, or plans to visit a showroom. Category scales for measuring the behavioral component of an attitude ask about a respondent’s likelihood of purchase or intention to perform some future action, using questions such as the following:
How likely is it that you will purchase a Honda Fit?
• I definitely will buy.
• I probably will buy.
• I might buy.
• I probably will not buy.
• I definitely will not buy.