chapter 6 textbook

Question 1

how could you measure loyalty (5 steps)

Answer 1

Step 1: have to identify the concept of interest
Step 2: develop a c construct
Step 3: define the concept constituively
Step 4: define the concept operatioally
Step 5: develop a measaruemtn scale
Step 6: evaluate the reliability and vadiity of the measuremtn scale

Question 2

Step 1: have to identify the concept of interest

Answer 2

A con- cept is an abstract idea generalized from particular facts. It is a category of thought used to group sense data together “as if they were all the same.

Question 3

Step 2: develop a c construct

Answer 3

Constructs are specific types of concepts that exist at higher levels of abstraction than do everyday concepts.
Not directly observable and you infer them from an indirect method

Question 4

Step 3: define the concept constituively

Answer 4

A constitutive (or theoretical, or conceptual) definition is a statement of the meaning of the central idea or concept under study, establishing its boundaries.

Question 5

Step 4: define the concept operatioally

Answer 5

operational definition: A statement of precisely which observable characteristics will be measured and the process for assigning a value to the concept.
, it assigns meaning to a construct in terms of the operations necessary to measure it in any con- crete situation

Question 6

Step 5: develop a measaruemtn scale

Answer 6

A scale is a set of symbols or numbers so constructed that the symbols or numbers can be assigned by a rule to the individuals (or their behaviours or attitudes) to whom the scale is applied
Nominal level: Scales that partition data into mutually exclusive and collec- tively exhaustive categories. The term nominal means “name-like,” indicating that the num- bers assigned to objects or phenomena are naming or classifying them but have no true number value; that is, the numbers cannot be ordered, added, or divided
Ordinal level : Scales that maintain the label- ling characteristic of nominal scales and have the ability to order data.
Interval leve: Scales that have the characteristics of ordinal scales, plus equal intervals between points to show relative amounts; they may include an arbitrary zero point.
Ratio level: Scales that have the character- istics of interval scales, plus a meaningful zero point so that magnitudes can be compared arithmetically.

Question 7

Step 6: evaluate the reliability and vadiity of the measuremtn scale

Answer 7

An ideal marketing research study would provide information that is accurate, precise, lucid, and timely.
m= a + e
a= accuracy
e= errors → can be systematic or random. Systematic error results in a constant bias in the measurements, caused by faults in the measurement instrument or process. Random error also influences the measurements but not systemati- cally. Thus, random error is transient in nature.

Question 8

Reliability

Answer 8

A measurement scale that provides consistent results over time is reliable
Thus, reliability is the degree to which measures are free from random error and, therefore, provide consistent data.
The less error there is, the more reliable the observa- tion is, so a measurement that is free of error is a correct measure

Question 9

Test retest reliability

Answer 9

Test–retest reliability is obtained by repeating the measurement with the same instru- ment, approximating the original conditions as closely as possible. The theory behind test– retest is that if random variations are present, they will be revealed by differences in the scores between the two tests
stability means that very few differences in scores are found between the first and second administrations of the test; the measuring instrument is said to be stable.
There are several problems with test–retest reliability. First, it may be very difficult to locate and gain the cooperation of respondents for a second testing. Second, the first measurement may alter a person’s response on the second measurement. Third, envi- ronmental or personal factors may change, causing the second measurement to change

Question 10

Equivalent form of reliability

Answer 10

The ability of two very simi- lar forms of an instrument to produce closely correlated resul
There are two problems with equivalent forms that should be noted. First, it is very difficult, and perhaps impossible, to create two totally equivalent forms. Second, if equivalence can be achieved, it may not be worth the time, trouble, and expense involved.

Question 11

Internal consistency reliability

Answer 11

Internal consistency reliability assesses the ability to produce similar results when different samples are used to measure a phenomenon during the same time period
Split half technique: A method of assessing the reli- ability of a scale by dividing the total set of measurement items in half and correlating the results

Question 12

Validity

Answer 12

The degree to which what the researcher was trying to measure was actually measured.
A scale or other measuring device is basically worthless to a researcher if it lacks validity because it is not measuring what it is supposed to.

Question 13

Face validiidty

Answer 13

The degree to which a measurement seems to measure what it is supposed to measure.
Weakest form

Question 14

Content validity

Answer 14

Representativeness, or sampling adequacy, of the content of the measurement instrument.
In other words, does the scale provide adequate coverage of the topic under study?
Judment matter

Question 15

Criterion related validity

Answer 15

Criterion-related validity examines the ability of a measuring instrument to predict a variable that is designated a criterion.
To sub categories: predictive, concurrent
Predictive : The degree to which a future level of a criterion variable can be forecast by a current measure- ment scale.
Concucrrenta; The degree to which another variable, measured at the same point in time as the variable of interest, can be predicted by the measurement instrument.

Question 16

Construct validity

Answer 16

The degree to which a measure- ment instrument represents and logically connects, via the underlying theory, the observed phenomenon to the construct.
Two statistical measures of construct: convergent and discirminant
Convergent: The degree of correlation among different measurement instru- ments that purport to measure the same construct.
Discirminant : A measure of the lack of association among constructs that are supposed to be different.

Convergent validity checks that different tests for the same thing agree.
Discriminant validity checks that tests for different things don’t overlap.

Question 17

Scaling

Answer 17

Procedures for assigning numbers (or other symbols) to properties of an object in order to impart some numerical characteristics to the properties in question.
Scales are either unidimensional or multi-dimen- sional.

Question 18

Unidimensioanl

Answer 18

Scales designed to measure only one attribute of a concept, respondent, or object.

Question 19

Multi:

Answer 19

Scales designed to measure several dimensions of a concept, respondent, or object.

Question 20

Rank order scales

Answer 20

Rank-order scales, on the other hand, are comparative scales because the respon- dent is asked to compare two or more items and rank each item. Rank-order scales are widely used in marketing research for several reasons. They are easy to use and give ordinal measurements of the items evaluated.
Rank order scales have disavantagaes: If all of the alternatives in a respondent’s choice set are not included, the results could be misleading. For example, a respondent’s first choice on all dimensions in the eye shadow study might have been Maybelline, which was not included. A second problem is that the concept being ranked may be completely outside a person’s choice set, thus producing meaningless data. Perhaps a respondent doesn’t use eye shadow and feels that the product isn’t appropri- ate for any woman. Another limitation is that the scale gives the researcher only ordinal data. Nothing is learned about how far apart

If not all possible options are included, the rankings can be misleading.
Sometimes the concept being ranked might not even apply to a respondent, leading to invalid or meaningless data.
Rank-order scales only give an ordinal measurement, which means we know the order of preferences but not how much more one item is preferred over another

Question 21

Q sorting

Answer 21

A measurement scale employing a sophisticated form of rank ordering using card sorts

Question 22

Paired comparisons

Answer 22

Measurement scales that ask the respondent to pick one of two objects in a set, based on some stated criteria.
Paired comparisons overcome several problems of traditional rank-order scales. First, it is easier for people to select one item from a set of two than to rank a large set of data. Second, the problem of order bias is overcome; there is no pattern in the ordering of items or questions to create a source of bias
On the negative side, because all possible pairs are evaluated, the number of paired comparisons increases geometrically as the number of objects to be evaluated increases arithmetically. Thus, the number of objects to be evaluated should remain fairly small to prevent inter- viewee fatigue.

Question 23

Constant sum scales

Answer 23

Measurement scales that ask the respondent to divide a given number of points, typically 100, among two or more attributes, based on their importance to him or her.
A major disadvantage of this scale is that the respondent may have difficulty allo- cating the points to total 100 if there are a lot of characteristics or items. Most research- ers feel that 10 items is the upper limit on a constant sum scale.

Question 24

Semantic difefreintial scales

Answer 24

Measurement scales that examine the strengths and weaknesses of a concept by having the respondent rank it between dichotomous pairs of words or phrases that could be used to describe it; the means of the responses are then plotted as a profile or image
The semantic differential is a quick and efficient means of examining the strengths and weaknesses of a product’s or company’s image versus those of the competition.
Dussvantages: suffers from a lack of satndardization. Number of dovusions on the scale can be an issue. Halo effect . occasional lack of universally accepted bipolar adjective

Question 25

Stapel scales

Answer 25

Measurement scales that require the respondent to rate, on a scale ranging from +5 to −5, how closely and in what direction a descriptor adjective fits a given concept Modification of the semantic differential The primary advantage of the Stapel scale is that it enables the researcher to avoid the arduous task of creating bipolar adjective pairs. The scale may also permit finer discrimination in measuring attitudes A drawback is that descriptor adjectives can be phrased in a positive, neutral, or negative vein, and the choice of phrasing has been shown to affect the scale results and the person’s ability to respond.15 The Stapel scale has never had much popularity in commercial research and is used less frequently than the semantic differential.

Question 26

Likert scales

Answer 26

The Likert scale is another scale that avoids the problem of developing pairs of dichot- omous adjectives. The scale consists of a series of statements expressing either a favour- able or an unfavourable attitude toward the concept under study. The respondent is asked to indicate the level of her or his agreement or disagreement with each statement by assigning it a numerical score. The scores are then totalled to measure the respon- dent’s attitude. A disadvantage of the Likert scale is that it could bias the respondent by putting thoughts into the respondent’s head. For example, a respondent asked to indicate the extent of agreement with a statement such as “Starbucks coffee is overpriced.” may not have previously considered Starbucks as being overpriced. But, now that the thought has been planted, the respondent may indicate a greater extent of agree- ment with the statement than the respondent may have otherwise. Moreover, the statement may lead the respondent to assume that there must be a general feeling among people that Starbucks is overpriced, which is why the researcher included that statement in the survey. This assumption could also lead to a potential bias in the response. Steps to build it: identifies the concept or activity to be scaled , assemble a large number of statements concerning how the public feels abut whatever you are seein, classify each test with faourbale and unfavourble. The respondent indicates how tehy feel like (stringly agree, agree…). You give the response a humerical weight

Question 27

Determinant attitudes

Answer 27

Those consumer attitudes most closely related to preferences or to actual purchase decisions. There are three major approaches to identifying determinant atti- tudes: (1) direct questioning, (2) indirect questioning, and (3) observation.

Question 28

to summarize

Answer 28

Steps in the measurement process : steps 1-4. Then 5 is develop a scale which can be nominal, ordinal, interval, ratio. Step 6 is evaluate the reliabultiy and valdiidty. Reliability has different types like internal consistency, equivalent form and test retest. Validity has construct, criterion related, content and face Scaling: unidiemndsional and mutli dimensainal scales Attitude measurement scales : have a bunch Considerations in selecting a scale Methods of determining buyer motives: direct, indirect and observation