Ch12 - Asking Good Questions

Which of the following is the most complete description of permissible measures of central tendency for an interval scale? a. Arithmetic mean, median, mode b. Geometric mean, mode, median c. Median, mode d. Arithmetic mean, geometric mean e. Mode, arithmetic mean

a. Arithmetic mean, median, mode

Which of the following is FALSE? a. Interval scales allow researchers to compare the absolute magnitude of numbers. b. A ratio scale differs from an interval scale in that the ratio scale has a natural zero. c. Determination of absolute zero is difficult with measurements in marketing. It is debatable whether many of the constructs used in marketing even possess this property. d. All of these are false. e. None of these are false.

a. Interval scales allow researchers to compare the absolute magnitude of numbers.

An attitude rating scale that uses a set of faces, varying systematically in terms of whether they are frowning or smiling, to capture a person's satisfaction or preference is an example of which of the following? a. Itemized-ratings scale b. Semantic-differential scale c. Graphic-ratings scale d. Summated-ratings (Likert) scale e. Comparative-ratings scale

a. Itemized-ratings scale

Code words, letters, or numbers used by security and intelligence organizations to form secret codes are examples of what type of scales? a. Nominal b. Ordinal c. Interval d. Ratio e. Conscious

a. Nominal

Which of the following scales is reflected by an attribute of an object that represents a non-ordered classification? a. Nominal b. Ordinal c. Interval d. Ratio e. Conscious

a. Nominal

What type of scale is the following? "What is your gender? 1 Male 2 Female" a. Nominal scale b. Ordinal scale c. Interval scale d. Ratio scale e. Likert scale

a. Nominal scale

What factor determines which level of measurement is possible? a. Properties of the attribute b. Objects being measured c. Quantities of attributes of objects d. Quantities of objects being measured e. Variance among attributes

a. Properties of the attribute

A ratio scale allows the comparison of absolute magnitudes but not the differences in magnitude. a. True b. False

F

Consistency is the hallmark of validity. a. True b. False

F

In general, when using itemized-rating scales, using five to nine categories works best. a. True b. False

F

It is possible to compare the absolute magnitude of numbers when measurement is made on the basis of an interval scale. a. True b. False

F

Random error affects the measurement in a constant way. a. True b. False

F

The four types of scales used to measure an attribute are nominal, ordinal, interval, and ratio. a. True b. False

F

Validity is the ability of a measure to obtain similar scores for the same object, trait, or construct across time, across different evaluators, or across the items forming the measure. a. True b. False

F

You should use the lowest level of measurement possible when developing a measure for some attribute. a. True b. False

F

A ratio scale differs from an interval scale in that it possesses a natural or absolute zero that reflects the complete absence of the attribute being assessed. a. True b. False

T

A useful property of all scales above the nominal level of measurement is that of order. a. True b. False

T

Error in measurement is also known as constant error because it affects the measurement in a constant way is called systematic error. a. True b. False

T

Ratio scales can be used to do everything that interval, ordinal, and nominal scales can do. a. True b. False

T

Reliability is necessary, but not sufficient, for establishing the validity of a measure. a. True b. False

T

The four types of scales used to measure an attribute are nominal, ordinal, interval, and ratio. a. True b. False

T

The key features of the summated-ratings scale are a set of statements with which respondents indicate a level of agreement. a. True b. False

T

The most common approach to measuring attitudes is to obtain self-reports from respondents. a. True b. False

T

When using the ordinal scale, the chosen number scale reflects the relative standing of the different options for that particular individual. a. True b. False

T

You should use the lowest level of measurement possible when developing a measure for some attribute. a. True b. False

T

Which of the following statements is INCORRECT with respect to ordinally scaled variables? a. The difference between the first- and second-ranked people is the same as the difference between the second- and third-ranked people. b. Both the median and mode are permissible measures of average. c. We can transform an ordinal scale in any way we wish as long as we maintain the basic ordering of the objects. d. The ordinal scale implies identity. e. The ordinal scale implies relative standing on the measured attributes.

a. The difference between the first- and second-ranked people is the same as the difference between the second- and third-ranked people.

Which of the following statements about nominal scales is TRUE? a. The numbers identify the objects. b. The numbers order the objects. c. The median can be used as the measure of average. d. The mean can be used as the measure of average. e. None of these are true.

a. The numbers identify the objects.

An advertiser has developed a television advertisement designed to create a "feeling of warmth" on the part of viewers because she believes that this will lead to positive attitudes toward the product and ultimately to increased sales. In order to determine if the ad does produce the desired feelings in viewers, she has devised a three-item scale to measure the warmth construct. After viewing the commercial (independently), Joe had a score of 15 on the warmth scale, and Perry had a score of 9. Which of the following statements is FALSE? a. The scale used is likely to be at least as reliable as a ten-item scale because a ten-item scale is more likely to inadvertently tap into constructs other than warmth. b. If Joe and Perry had viewed the commercial and listened to instruction together at the same time and then had completed the scales, at least one source of variation might have been greatly reduced. c. It is possible that Perry's lower score might have resulted from accidentally checking the wrong response. d. Some relatively stable characteristic of an individual, such as Perry's unwillingness to share the actual intensity of his feelings, might have contributed to the difference in the two scores. e. Joe did not necessarily have "warmer" feelings toward the ad than Perry.

a. The scale used is likely to be at least as reliable as a ten-item scale because a ten-item scale is more likely to inadvertently tap into constructs other than warmth.

Which of the following is FALSE? a. You are studying the product life cycle phenomenon and attempting to classify certain products into the following stages: introduction, growth, maturity, and decline. This represents nominal scaling. b. The above products can be further classified according to whether they are consumer or industrial goods. This is an example of nominal scaling. c. The respondent's gender is an example of a nominal scale. d. All of these are false. e. None of these are false.

a. You are studying the product life cycle phenomenon and attempting to classify certain products into the following stages: introduction, growth, maturity, and decline. This represents nominal scaling.

When the scale type is nominal, the appropriate measure of central tendency is ____. When the scale type is interval or ratio, the appropriate measure of central tendency is ____. a. mode; mean b. median; mean c. mean; mode d. mean; median e. mode; median

a. mode; mean

The interchanging or switching of scale values among and between objects does not affect the originally intended purpose of the scale when the scale is a. nominal. b. ordinal. c. interval. d. ratio. e. the interchanging of scale values always affects the purpose.

a. nominal.

The number of "outs" in a baseball game is measured on a(n) a. ratio scale. b. interval scale. c. ordinal scale. d. nominal scale. e. lambda scale.

a. ratio scale.

The highest level of measurement from those in the following list is a. ratio. b. ordinal. c. interval. d. nominal. e. conscious.

a. ratio.

One hundred sports writers are given a survey and asked to rank basketball players on categories such as rebounding, points scored, and assists. The results of the survey indicate that most sports writers agree on the top ten players within the given categories. This is evidence of a. reliability. b. validity. c. random error. d. response error. e. systematic error.

a. reliability.

People magazine conducts an annual survey across 25 Asian countries and asks respondents to rate automobiles across 30 attributes. The questions ask the consumers to rate each brand from 1=low to 5=high on quality, value, image, price, etc. This is an example of a. self-report attitude measures. b. measuring attitude using physiological reaction. c. indirect technique to measure attitude d. direct observation of attitude-related behavior. e. performance of objective tasks to determine attitudes.

a. self-report attitude measures.

Due to a misunderstanding of his instructions, a research assistant consistently entered the upper value of an attitude scale as a "6" instead of a "7" when keypunching data for his boss. This is an example of a. systematic error. b. researcher error. c. random error. d. scaling error. e. marginal error.

a. systematic error.

As defined in the text, measurement consists of a. the assignment of numbers (representing quantities of attributes) to objects or events according to rules. b. using a given scale to make comparisons. c. determining the frequency of a particular attribute. d. assigning numbers solely for identification purposes. e. numbers which can be assigned to the physical properties of objects.

a. the assignment of numbers (representing quantities of attributes) to objects or events according to rules.

With an interval scale, a. we cannot compare the absolute magnitude of numbers. b. we cannot state that the difference between 0.25 and 0.50 is the same as the difference between 37.75 and 38.00. c. there is a naturally-occurring zero point. d. the median and the mode are the only permissible measures of average. e. All of these statements concerning an interval scale are false.

a. we cannot compare the absolute magnitude of numbers.

Which of the following is TRUE? a. The recording of a respondent's social class using the categories upper, middle, and lower represents the use of a nominal scale. b. An ordinal scale can be transformed in any manner possible provided that the basic ordering of the objects is maintained. c. When the scale has an arbitrary zero point, it makes sense to say that A is twice as much as B. d. Interval scales possess an absolute zero point whereas ordinal scales do not. e. All of these are false.

b. An ordinal scale can be transformed in any manner possible provided that the basic ordering of the objects is maintained.

Which of the following is an example of an interval scale? a. Female/Male b. Attitude toward an advertisement c. Social class d. Income e. Number of purchasers

b. Attitude toward an advertisement

Which of the following is FALSE? a. It is true about psychological measurement that the situation in which the measurement is obtained can affect the score obtained. b. Random error is also known as constant error. This is due to the fact that there are typically an equal number of positive and negative errors that yield a constant sum. c. The validity of a measurement instrument refers to the extent to which differences in scores reflect true differences among entities on the characteristic that the instrument seeks to measure. d. All of these are false. e. None of these are false.

b. Random error is also known as constant error. This is due to the fact that there are typically an equal number of positive and negative errors that yield a constant sum.

Which attitude rating scale uses pairs of bipolar words or phrases and is particularly useful in corporate, brand, and product-images studies? a. Itemized-ratings scale b. Semantic-differential scale c. Graphic-ratings scale d. Summated-ratings (Likert) scale e. Comparative-ratings scale

b. Semantic-differential scale

Which of the following about ordinal data is TRUE? a. An ordinal scale implies order but not identity. b. The attribute being measured must possess the ordinal property to allow ordinal scaling that is meaningful. c. With an ordinal scale we can say the difference between the first and second is the same as the difference between the second and the third. d. The calculation of means is appropriate with ordinal data. e. Grade point average is a good example of an ordinal scale.

b. The attribute being measured must possess the ordinal property to allow ordinal scaling that is meaningful.

Which of the following statements is INCORRECT with respect to a nominally classified variable? a. Counting is the only permissible operation. b. The median is a permissible measure of central tendency. c. The only property conveyed by the number is identity. d. When members of a sample have been classified along a nominal scale, it is possible to say what percentage of the sample is classified a certain way. e. The mode is a permissible measure of central tendency.

b. The median is a permissible measure of central tendency

Which of the following is NOT true when a scale is a ratio scale? a. The scale allows for comparison of the absolute magnitude of numbers. b. The zero point is established arbitrarily. c. If "a" is greater than "b", and "b" is greater than "c", then "a" is greater than "c". d. If "a" equals "b", and "b" equals "c", then "a" equals "c". e. The mode and the median are permissible measures of central tendency.

b. The zero point is established arbitrarily.

If s/he wanted to use the highest level of measurement possible, the researcher measuring respondent age would most likely suggest using ____ and a(n) ____ scale. a. an open-ended item; interval b. an open-ended item; ratio c. age range categories; ordinal d. age range categories; interval e. an open-ended item; nominal

b. an open-ended item; ratio

The first step in measuring an attribute is to a. assign numbers that reflect the properties of the attribute. b. determine the properties of the attribute. c. measure the object first, then the attributes. d. decide which type of analysis you wish to perform on the measurement data. e. use the lowest level of measurement possible.

b. determine the properties of the attribute.

A study was done on the campus of a major state institution to measure attitudes of students to-wards the recent fee increase. One question asks respondents to rank the five fees they are willing to pay from 1=most willing to 5=least willing. This is scaled as a(n) a. nominal scale. b. ordinal scale. c. interval scale. d. ratio scale. e. None of these are correct.

b. ordinal scale.

A researcher wishes to rank the order of preference consumers have for a particular set of products. The lowest measurement scale the researcher could use to measure this is a. ratio. b. ordinal. c. interval. d. nominal. e. conscious.

b. ordinal.

The attitude ratings scale with which we might use a snake diagram is a. the itemized-ratings scale. b. the semantic-differential scale. c. the graphic-ratings scale. d. the summated-ratings (Likert) scale. e. the comparative-ratings scale.

b. the semantic-differential scale.

If a respondent says his income is "Over $100,000 per annum" when, in fact, it is "$50,000 per annum", the difference between the true and actual measurement of income in this case could be said to be LACK of a. reliability. b. validity. c. All of these are correct. d. None of these are correct. e. nominal scaling.

b. validity.

While completing a survey on attitudes toward house ownership, John was hit by an attack of indigestion. Which of the following best accounts for the difference between John's and another respondent's scores? a. A true difference in attitudes b. A true difference in personality between John and Bob c. A difference due to temporary personal factors d. A difference due to situational factors e. A difference due to variations in administration.

c. A difference due to temporary personal factors

Which of the following statements about nominal data is TRUE? a. Possible operations with nominal scales are few, but due to the fact that the properties of "greater than" or "less than" apply, the addition of numbers is permissible. b. The median is an appropriate measure of central tendency with nominal data. c. A nominal scale is associated with the concept of identification and categorization. d. Computing the mean is an acceptable statistical operation when one is dealing with nominal data. e. All of these are false.

c. A nominal scale is associated with the concept of identification and categorization.

Which attitude rating scale offers respondents the greatest degree of freedom in providing answers? a. Itemized-ratings scale b. Semantic-differential scale c. Graphic-ratings scale d. Summated-ratings (Likert) scale e. Comparative-ratings scale

c. Graphic-ratings scale

Which type of scale assumes an arbitrary zero point? a. Ratio b. Median c. Interval d. Nominal e. Ordinal

c. Interval

You have been assigned the task of evaluating consumer acceptance of and favorability toward a new product that a company has just test marketed in your area. In order to determine the degree to which consumers hold favorable attitudes toward the product, what is the lowest level of scale that you can use? a. Nominal b. Ordinal c. Interval d. Ratio e. Need more information

c. Interval

Which of the following is(are) permissible measure(s) of central tendency with ordinal scales? a. Arithmetic mean, median, and mode b. Geometric and harmonic means c. Median and mode d. Arithmetic mean and median e. Arithmetic mean

c. Median and mode

Which of the following is FALSE? a. If a measure is reliable, it is not heavily influenced by transitory factors that cause random errors. b. If an item lacks correlation with other scale items measuring the same construct, there is evidence that the item does not belong and should be deleted. c. Systematic error is due to temporary aspects of the person or measurement situation, which affects the measurement in irregular ways. d. Reliability is necessary, but not sufficient, for establishing the validity of a measure. e. If a set of items all measure the same construct, the responses to the items should be correlated.

c. Systematic error is due to temporary aspects of the person or measurement situation, which affects the measurement in irregular ways.

Which of the following about interval scales is FALSE? a. An interval scale exhibits the property of order. b. Absolute magnitudes cannot be compared using an interval scale because the zero point is established arbitrarily. c. The number of years the respondent has lived at a particular address is an example of an interval scale. d. All of these are false. e. None of these are false.

c. The number of years the respondent has lived at a particular address is an example of an interval scale.

Which of the following statements concerning measurement is FALSE? a. Properties of the attributes must first be determined before measurement begins. b. Numbers are assigned to accurately reflect properties of an attribute. c. Typically, researchers recognize five "levels" of measurement. d. None of these are correct. e. All of these are correct.

c. Typically, researchers recognize five "levels" of measurement.

Which of the following is FALSE? a. With a ratio scale, it is possible to have zero amount of some attribute. b. There are few psychological constructs that can reasonably be assumed to have a natural or absolute zero. c. When using a ratio scale, it is inappropriate to calculate a mean score. d. None of these are false. e. All of these are false.

c. When using a ratio scale, it is inappropriate to calculate a mean score.

A question that asks "How much you would be willing to pay in annual premium for a $100,000 home insurance policy" in the following way: "Check appropriate category—____ < $25 per month/____ $26 to $40 month/____ $41 to $55 month/____ > $55 per month" would be scaled as a(n) a. nominal scale. b. ordinal scale. c. interval scale. d. ratio scale. e. Likert scale.

c. interval scale.

A researcher wishes to compare the differences in consumer feelings about the perceived reliability of a set of products so as to know the relative strength of feelings about each product's reliability. The lowest measurement scale the researcher could use to measure this is a. ratio. b. ordinal. c. interval. d. nominal. e. conscious.

c. interval.

Regarding measurement basics, it is possible to measure a. objects. b. quantities of objects. c. quantities of attributes of objects. d. variance among objects. e. variance among attributes.

c. quantities of attributes of objects.

The ordinal scale represents a higher level of measurement than the nominal scale in that a. the assigned numerals serve to identify the objects. b. the magnitude of the differences in the objects is shown. c. the assigned numerals represent the order as well as identifying the object. d. it has a natural zero. e. it has an arbitrary zero.

c. the assigned numerals represent the order as well as identifying the object.

For a scale to have ordinal properties, which of the following must be established? a. If "a" is greater than "b", then "b" is not greater than "a". b. If "a" is greater than "b", and "b" is greater than "c", then "a" is greater than "c". c. If "a" is equal to "b", and "b" is equal to "c", then "a" is equal to "c". d. All of these are correct. e. None of these are correct.

d. All of these are correct.

Which of the following is (are) permissible measure(s) of central tendency with ratio scales? a. Arithmetic mean, median, and mode b. Geometric mean c. Median and mode d. Both geometric mean as well as arithmetic mean, median, and mode e. Both geometric mean as well as median and mode

d. Both geometric mean as well as arithmetic mean, median, and mode

What type of scale is the following? "Cars manufactured by South Korean companies are of (place on "X" on the point of this spectrum that represents your view) Poor Quality/High Quality" a. Stapel scale b. Likert scale c. Temperature scale d. Graphic-rating scale e. Itemized rating scale

d. Graphic-rating scale

What type of scale is the following? "Indicate your opinion about services Delta Airlines provides, by placing an "X" at the appropriate position on the lines to the right of the services below. Services—Very Bad/Very Good: Frequent flyer program/Continental magazine/On flight food service" a. Stapel scale b. Likert scale c. Temperature scale d. Graphic-rating scale e. Itemized rating scale

d. Graphic-rating scale

The magnitude of the differences in objects is shown when which types of scales are used? a. Interval or ordinal b. Ratio or ordinal c. Ordinal or nominal d. Interval or ratio e. Nominal or interval

d. Interval or ratio

What characteristic must a scale possess in order for the researcher to compute the ratio of the magnitude of the numbers? a. It must have an arbitrary zero point. b. The range of the absolute scale intervals must be divisible by two. c. It must use only integers. d. It must have an absolute zero point. e. The modal ratio must be greater than two.

d. It must have an absolute zero point.

____ scales possess an absolute zero. a. Nominal b. Ordinal c. Interval d. Ratio e. Continuous

d. Ratio

What type of scale is the following? "What was your annual gross income (before taxes) last year? $____." a. Nominal scale b. Ordinal scale c. Interval scale d. Ratio scale e. Constant-sum scale

d. Ratio scale

When it comes to measurements, which of the following are TRUE? a. Market researchers measure objects, not their attributes. b. The definition of "measurements" is narrow, specifying how numbers will be assigned. c. We could measure a person's income, age, gender, or attitude toward something using a scale of 1-5. d. The qualities of attributes being measured dictate the rules for how numbers are assigned. e. All of these are true.

d. The qualities of attributes being measured dictate the rules for how numbers are assigned.

Validity is a. concerned with the relationships of random error with systematic error. b. equal to the true score plus the systematic error. c. equal to the true score plus systematic error minus random error. d. concerned with the extent to which differences in scores reflect true differences in the characteristic. e. concerned with the extent to which differences in scores reflect instability in the measurements.

d. concerned with the extent to which differences in scores reflect true differences in the characteristic.

You should use the highest level of measurement possible when developing a measure for some at-tribute because a. higher levels have more properties. b. there are more kinds of analysis available than at lower levels of measurement. c. it is easier to use higher levels of measurement. d. higher levels have more properties and there are more kinds of analysis available at lower levels of measurement. e. None of these are correct; you should use the lowest level of measurement possible when developing a measure for some attribute.

d. higher levels have more properties and there are more kinds of analysis available at lower levels of measurement.

A research effort requires the researcher to use numbers to identify or categorize particular objects. The type of scale the researcher will use is a. ratio. b. ordinal. c. interval. d. nominal. e. conscious.

d. nominal.

It is reasonable to state that one object has twice as much of the attribute property when it has a score of 60, and the other object has a score of 30 when measurement is on a(n) ____ scale. a. nominal b. ordinal c. interval d. ratio e. interval or ratio

d. ratio

The first step in developing a measure of marketing constructs is to a. generate sample items. b. purify the measure. c. assess validity. d. specify the domain of the construct. e. collect data.

d. specify the domain of the construct.

The most widely used attitude-scaling technique in marketing research is a. the itemized-ratings scale. b. the semantic-differential scale. c. the graphic-ratings scale. d. the summated-ratings (Likert) scale. e. the comparative-ratings scale.

d. the summated-ratings (Likert) scale.

In a recent survey, those who brushed their teeth often under-reported the activity, while those who brushed their teeth less frequently over-reported their dental hygiene behavior. The possible cause for differences in scores could best be attributed to a. a true difference in the characteristic being measured. b. differences due to situational factors. c. differences due to temporary personal factors. d. true differences in other relatively stable characteristics of the individuals. e. differences due to variations in administration.

d. true differences in other relatively stable characteristics of the individuals.

Of the many sources of differences of scores, the researcher is especially interested in isolating the differences due to a. situational factors. b. variations in administration. c. mechanical factors. d. true differences in the characteristic. e. the sampling of items.

d. true differences in the characteristic.

Which of the following statements is FALSE with regard to ratio scales? a. A ratio scale exhibits the property of identity. b. With a ratio scale, the comparison of the absolute magnitude of the numbers is legitimate. c. A ratio scale possesses a natural zero point. d. The geometric mean is a meaningful measure of average with a ratio scale. e. A ratio scale allows the comparison of absolute magnitudes but not the differences in magnitudes.

e. A ratio scale allows the comparison of absolute magnitudes but not the differences in magnitudes.

Which of the following can be measured? a. Weight b. Height c. Attitude d. None of these are correct. e. All of these are correct.

e. All of these are correct.

Which of the following statements about the ordinal scale is TRUE? a. There is a consistent order in whatever numbers are assigned to the scale. b. It is possible to calculate the median as a measure of central tendency. c. Two individuals with the same rank order response can have very different feelings about what they ranked. d. It is impossible to say how much any individual respondent preferred one object to another. e. All of these are correct.

e. All of these are correct.

An advertiser has developed a television advertisement designed to create a "feeling of warmth" on the part of viewers because she believes that this will lead to positive attitudes toward the product and ultimately to increased sales. In order to determine if the ad does produce the desired feelings in viewers, she has devised a three-item scale to measure the warmth construct. After viewing the commercial (independently), Joe had a score of 15 on the warmth scale, and Perry had a score of 9. Which of the following is NOT a potential source of difference between these scores? a. True differences with regard to the feeling of warmth created by the ad. b. Perry had a rotten day at work, while Joe had a good day. c. The sampling of items chosen to measure warmth could have contributed to the difference in scores. d. Perry and Joe interpreted one of the questions differently. e. All of these are potential sources of differences between the scores.

e. All of these are potential sources of differences between the scores.

Which attitude rating scale is good for eliminating the halo effect that is common in scaling? a. Itemized-ratings scale b. Semantic-differential scale c. Graphic-ratings scale d. Summated-ratings (Likert) scale e. Comparative-ratings scale

e. Comparative-ratings scale

What type of scale is the following? "Indicate your overall opinion about Dell computers by circling one of the following categories: Very Bad/Bad/Good/Very Good: 1 2 3 4" a. Stapel scale b. Likert scale c. Temperature scale d. Graphic-rating scale e. Itemized rating scale

e. Itemized rating scale

The zero-scale value found when measuring marketing variables typically represents a. the ordinal nature of the variable. b. an absolute lack of an attribute. c. the logical existence of a natural zero point. d. a point of absolute zero magnitude. e. a point of indifference to which the value zero is arbitrarily attached.

e. a point of indifference to which the value zero is arbitrarily attached.

In marketing research, all of the following are levels of measurement EXCEPT a. ratio. b. ordinal. c. interval. d. nominal. e. conscious.

e. conscious.

The notion that equal differences among scores represent equal differences in the amount of the at-tribute possessed by the object applies to ____ scales. a. nominal b. ordinal c. interval d. ratio e. interval and ratio

e. interval and ratio