Using Product Involvement Segmentation to Improve Advertising Productivity Page 1 of 14

Using Product Involvement Segmentation to Improve Advertising

Productivity

William T. Neese, University of Arkansas at Little Rock

Ronald D. Taylor and Louis M. Capella, Mississippi State University

We used product involvement segmentation to test advertising effectiveness,

producing a three-way interaction effect for involvement segment, verbal strategy,
and the actual brand featured in the ad. Applying involvement segmentation can
enable marketers to design advertising preferred by individuals with stronger
purchase intentions toward the sponsoring brand, consumers who cannot be
identifted through demographics alone.

Introduction

Consumer "involvement" represents importance of or interest in, a product or

service, issue, situation, communication, etc., and is a state or condition which
varies across individuals and circumstances (Antil 1984). Most consumer behavior
researchers have conceptualized and operationalized involvement along cognitive
and affective dimensions. Yet the implications of varying levels of involvement on
conation would be most relevant for marketers (Stone 1984). There is ample
evidence that involvement states do impact actual behavior (Belk 1982; Bunn
1993). According to Pucely, Mizerski and Perrewe (1988), one "would expect an
involvement measure to bear a strong relationship to purchase intent (p. 41)."

The possibility of using product involvement as a segmentation variable has been

mentioned several times in the literature (Beatty, Homer, and Kahle 1988; Bolfing
1988; Flynn and Goldsmith 1993; Gensch and Javalgi 1987; Mittal 1989). Several
authors have established comprehensive involvement models (Beatty, Homer, and
Kahle 1988; Bloch and Richins 1983; Zaichkowsky 1986), and numerous studies
have operationalized involvement as an influence on the communication process
(Celsi and Olson 1988; Heslin and Johnson 1992; Leigh and Menon 1987;
Miniard, et al. 1991; Petty and Cacioppo 1981) as well as the buying decision
process in total (Bunn 1993).

Flynn and Goldsmith (1993) reported the superior ability of involvement to predict
behavior over demographics. In addition, Bowen and Chaffee (1974) hypothesized
and found "that pertinent advertisements will be more effective than non-pertinent
ones. but only when consumer involvement with the product is high (p. 613)."
Finally, Wright (1973) reported that receiver involvement influenced the cognitive
process (e.g., counterarguments), which subsequently mediated attitudinal message
acceptance. Thus, segmenting consumers by involvement states should enable

file://C:\WINDOWS\Desktop\1995\SMA\95swa092.htm 5/21/04
Using Product Involvement Segmentation to Improve Advertising Productivity Page 2 of 14

marketers to more accurately test the effectiveness of advertising strategies for

individuals who are more predisposed toward purchasing the product yet who
cannot be precisely identified by demographics. Our method allows marketers to
identify wording strategies which are preferred by consumers with the greatest
inclination to purchase the specific brand advertised.

Focus for the Study

According to Gardial et al. (1993), the effectiveness of advertising in enhancing

consumer orientations toward a brand depends on (1) the consumer's reaction
during processing plus (2) the consumer's ability to categorize new information
within an existing data set representing competing brands. Two levels of product
involvement are, therefore, important for segmentation purposes: (1) involvement
concerning the brand featured in the advertisements, and (2) pre-processing
involvement with the product type group to which the brand belongs. Sujan and
Dekleva (1987, p. 372) pointed out:

For products, the product type level is likely to constitute the basic level of
categorization. For example, for the category of cars, sports cars and family cars
are likely to be perceived as distinct subcategories, but various brands of sports
cars are likely to be seen as having many shared attributes and few distinct ones.

Our investigation establishes segments based on consumer involvement, then tests

for differences of hierarchy of effects response (including attitude toward the ad)
across segments. These variables are commonly and appropriately used to test
advertising effectiveness (Barry 1993; Wilkie and Farris 1975), Bunn (1993),
Richins and Bloch (1991), and Homer and Kahle (1990) indicate that Attitude-
Toward-the-Ad (A), Brand Attitude (A), and behavioral intention means are
greater for more involved individuals.

Meaningful segments will be significantly different from one another along

important dimensions. The global hypothesis therefore is that:

H1: Multivariate significance of differences in the expanded hierarchy of effects

will exist across involvement segments.

Subsequently, interaction effects between wording strategies and involvement

segments would allow us to identify the advertisement producing the most
favorable results among those highly involved consumers more likely to purchase
the brand being sponsored if purchase intentions are significantly greater for one
segment versus the others. Our method would thus produce a superior test of
advertising copy effectiveness.

file://C:\WINDOWS\Desktop\1995\SMA\95swa092.htm 5/21/04
Using Product Involvement Segmentation to Improve Advertising Productivity Page 3 of 14

Methodology

An involvement segmentation plan was established which considers two levels of

product involvement: (1) predispositions toward luxury sedans in general, and (2)
involvement with the sponsoring brand in particular. Lastovicka and Gardner's
(1979) battery of items designed to measure normative beliefs, commitment. and
familiarity was adopted to capture predispositions toward luxury sedans prior to
advertisement exposure. Zaichkowsky's (1985) Personal Involvement Inventory
(PII) was used to capture involvement with the sponsoring brand immediately after
exposure to the test ads. Dependent items were based largely on examples given by
Holmes and Crocker (1987), with brand belief measures being developed
specifically for this study (see Table 2 for item description).

Based on a summation of all 22 items, the Lastovicka-Gardner scale was split into
a high involvement group containing 51.1 percent of the sample and a low
involvement group containing 48.9 percent of the sample. Based on a summation
of all 20 items, Zaichkowsky's PII scale was split into a high involvement group
containing 51.5 percent of the sample and a low involvement group containing
48.5 percent of the sample. These proportions were nearest to the median split
procedure used by Laczniak and Carlson (1989) as possible given the nature of the
data. The four groups produced by our segmentation solution are shown in Table 1
alone with response means and standard deviations germane for each cell.

Sample Size and Composition

Miller (1991) reported that in America the major purchasers of luxury automobiles
come from the middle class; 78.2 percent of our responses (422) were drawn from
a civic organization attracting middle-class persons as members, with the
remainder (118) being commuter students from a nontraditional metropolitan
university. A total of 540 usable questionnaires were obtained, representing both
experienced owners and novices. Fifty-five questionnaires had been rejected for
missing data (all from the civic organization), resulting in an overall response rate
of 90.8 percent. No demand artifacts were detected. The civic group was paid
$2.00 per response; students added one point (100 point scale) to their final course
average for participation.

The non-student test group was intentionally formed from individuals likely to
purchase a luxury sedan (i.e., a judgment sample). They are older managers and
professionals with higher incomes. They tend to be married, college educated
males (many with graduate degrees), and caucasian. Of the non-student group, 12.6
percent reported owning the Cadillac Sedan Deville, 6.9 percent reported owning
the Lincoln Town Car, and 2.6 percent reported owning the Chrysler Imperial. This

file://C:\WINDOWS\Desktop\1995\SMA\95swa092.htm 5/21/04
Using Product Involvement Segmentation to Improve Advertising Productivity Page 4 of 14

totals 22.1 percent for our three brands alone, yet Darnay (1991) reported that all
sedans comprised only 21.2 percent of the U.S. automobile market. Ownership of
our brands by the non-student sample outpaced that for the general U.S.
population. However, only four of the 118 students in our sample (3.4 percent)
reported owning one of the three brands tested. Segmentation is useless on a
homogeneous target market, so our objective was to form a sample comprised of
consumers experienced with the products being advertised as well as novices. This
goal was achieved.

Table 1. Involvement Segment Data

Variable Description Mean Standard Deviation
CELL 1(n = 169): High Product Type Involvement, High Brand Involvement
Noncomparative Brand Beliefs 5.31 .87
Comparison Brand Beliefs 4.62 1.08
Ad Attitude 5.06 1.10
Brand Attitude 4.81 1.18
Purchase Intentions 4.37 1.50
CELL 2 (n = 107): High Product Type Involvement, Low Brand Involvement
Noncomparative Brand Beliefs 5.02 1.01
Comparison Brand Beliefs 4.17 1.29
Ad Attitude 4.33 1.23
Brand Attitude 4.10 1.30
Purchase Intentions 3.34 1.49
CELL 3 (n = 109): Low Product Type Involvement, High Brand Involvement
Noncomparative Brand Beliefs 5.08 1.17
Comparison Brand Beliefs 4.28 1.22
Ad Attitude 4.93 1.15
Brand Attitude 4.42 1.11
Purchase Intentions 4.22 1.30
CELL 4 (n = 155):Low Product Type Involvement, Low Brand Involvement
Noncomparative Brand Beliefs 4.92 .99
Comparison Brand Beliefs 4.00 .98
Ad Attitude 4.41 1.25
Brand Attitude 3.98 1.31
Purchase Intentions 2.87 1.55

file://C:\WINDOWS\Desktop\1995\SMA\95swa092.htm 5/21/04
Using Product Involvement Segmentation to Improve Advertising Productivity Page 5 of 14

Test Copy

Each subject processed one of twelve ads, randomly distributed in every monthly
civic group meeting or classroom session to avoid confounding the session
participants with specific treatments. Ads were sponsored by one of three existing
American-made luxury sedans (Cadillac Sedan Deville, Lincoln Town Car, and
Chrysler Imperial). Only the brand sponsoring the advertisement was mentioned by
name in each ad, with comparison statements referring to "other American-made
luxury sedans." Four verbal-only strategies were tested for each of the three brands
(less informative noncomparative copy, noncomparative more informative copy,
less informative but having comparison statements included, and more informative
with comparative claims). See Neese and Taylor (1994) for a description of the test
copy used.

Results

Principal components analysis (varimax rotation) and scale purification procedures

determined analysis variables. Table 2 contains reliability coefficients for the
segmentation as well as response variables, plus a brief description of each item.

Table 2. Variable Description and Reliability Measures

Variable Number Coefficient
(Item Description) Coefficient Alpha

Product Type Ego Involvement, 22 .88

Familiarity, and Commitment

(complete scale - Lastovicka and Gardner 1979)

Brand Involvement 20 .94

(complete scale - Zaichkowsky 1985)

Noncomparison Brand Beliefs 6 .78

(has 4-cam 32-valve 250-hp V8 engine,system, has

advanced electronic suspensionhas fulltime all-wheel

drive, has advancedantilock braking system,

has driver and passenger side airbags, is full-sized)

file://C:\WINDOWS\Desktop\1995\SMA\95swa092.htm 5/21/04
Using Product Involvement Segmentation to Improve Advertising Productivity Page 6 of 14

Comparison Brand Beliefs 4 .78

(more aerodynamic, less effort to operate, superior

quality, priced competitively)

Attitude Toward the Ad 7 .84

(offensive? useful? believable? informative?

clear? likable? convincing?)

Attitude Toward the Brand 2 .67

(this brand is distinctive, my attitude toward

the brand is more positive after reading the ad)

Purchase Intentions 3 .81

(would try this LS*, would seek out this LS,

will likely purchase this LS)
*Luxury Sedan (LS)

Having satisfied all assumptions, Multivariate Analysis of Variance (MANOVA)

was used to explore differences across segments, and ANOVAs were run to
explore differences in demographic characteristics across the cells. Results are
displayed in Table 3. Non-demographic items were measured on 7-point scales
with 1 being least favorable to the brand and 7 being most favorable.

All multivariate tests were significant at <.001 as were four of the five univariate
tests, providing support for our hypothesis. The remaining univariate test for
noncomparison brand beliefs was significant at the .006 level. According to Taylor
(1994, p. 61), "GM wants to aim each of its new cars and trucks at discrete market
segments, as measured by age, income and family size [emphasis added]." The
lack of significant demographic differences (with the possible exception of
educational attainment) across cells Strongly indicates that psychological
segmentation based on involvement measures is a more precise method for
determining purchase intentions in the automobile industry than the use of
demographics alone.

Table 3. Multivariate Significance Tests Across Segments for

file://C:\WINDOWS\Desktop\1995\SMA\95swa092.htm 5/21/04
Using Product Involvement Segmentation to Improve Advertising Productivity Page 7 of 14

Hierarchy of Effects and Aad Variables

(S = 3, M = 1/2, N = 265)

Test Name Value F Hypth.DF Error DF Sig.of F

Pillais .2063 7.888 15.00 1602.00 <.001
Hotellings .2458 8.695 15.00 1592.00 <.001
Wilks .7987 8.308 15.00 1469.02 <.001
Roys .1790

Univariate F- Tests Across Segments with (3,536) D. F.

Dependent Variable F-value Sig. of F
Noncomparison Brand Beliefs 4.245 .006
Comparison Brand Beliefs 8.788 <.001
Ad Attitude 13.111 <.001
Brand Attitude 14.027 <.001
Purchase Intentions 34.747 <.001

Demographic Characieristic Differences Across Segments

Dependent Variable F-value Sig. of F
Marital Status .172 .915
Household Income 1.442 .230
Sex 1.776 .151
Age .200 .896
Education 2.449 .063
Race .080 .971
Occupation H.O.H. .592 .621

Main effects plus two- and three-way interaction effects were identified; detailed
discussion of all observed effects is unnecessary for the purpose of this study,
which will instead concentrate on the three-way interaction observed for brand
involvement, verbal strategy, and market ranking (Table 4). All multivariate tests
were significant at .05.

Table 4. Multivariate Significance Tests for Three-Way

Interaction Effect, Involvement by Verbal Strategy by
Sponsoring Brand (S = 5, M = 0, N = 255)

Test Name Value F Hypth. DF Error DF Sig of F.

file://C:\WINDOWS\Desktop\1995\SMA\95swa092.htm 5/21/04
Using Product Involvement Segmentation to Improve Advertising Productivity Page 8 of 14

Pillais .089 1.554 30.00 2580.00 .028

Hotellings .093 1.586 30.00 2552.00 .023
Wilks .913 1.571 30.00 2050.00 .025
Roys .063

Univariate F- Tests for Three- Way Interaction with (6,516) D. F,

Involvement by Verbal Strategy by Sponsoring Brand
Dependent Variable F-value Sig. of F
Noncomparison Brand Beliefs .629 .707
Comparison Brand Beliefs 4.293 <.001
Ad Attitude 2.422 .026
Brand Attitude 1.622 .139
Purchase Intentions .662 .681

Significant effects related to comparative brand belief and Aad responses. Item-by-
item exploration is thus warranted for these two variables (see Table 5).
Consumers with more favorable purchase intentions responded differently to
verbal strategies depending upon the brand. Certain verbal strategies encouraged
more favorable brand attribute beliefs among consumers most apt to purchase the
brand being advertisied. These persons also found certain copy to be significantly
less offensive, more convincing, more likable, and more informative than other
variants. Marketers are additionally able to evaluate which strategies work best for
competitors with this method.

The Lincoln Town Car was best able to position itself as "more aerodynamically
designed than other domestic luxury sedans" for the high involvement group
through more informative comparative advertising (mean = 5.35 versus 3.85 for
the same copy low involvement group, and 3.30 for the more informative
noncomparative high involvement group). Means for the high involvement, more
informative, comparative advertising group were 4.60 and 4.52 respectively for the
market leading Cadillac Sedan Deville and the market lagging Chrysler Imperial.
On the other hand, the Cadillac Sedan Deville received the greatest benefit from
the less informative comparative copy concerning the "less effort demanded" claim
(e.g., mean = 4.67 versus 3.88 for the more informative comparative copy and 3.83
for the less informative noncomparative version).

For the high involvement group concerning the "superior quality" dimension, the
Cadillac fared best among all verbal strategies using more informative
noncomparison copy (mean = 4.63 versus 3.11 for the less involved group), the
Lincoln using more informative comparison copy (mean = 4.65 versus 3.55 for the

file://C:\WINDOWS\Desktop\1995\SMA\95swa092.htm 5/21/04
Using Product Involvement Segmentation to Improve Advertising Productivity Page 9 of 14

low involvement group), and the Chrysler using less informative noncomparative
advertising (mean = 4.12 versus 3.39 for the less involved group). The less
informative comparative treatment worked best versus the other three verbal
strategies in positioning all three brands as "competitively priced with other
American made luxury sedans" in the minds of the high involvement consumers.
For example, the mean was 5.56 for the Imperial versus 4.60 among less concerned
consumers. Effects were least positive for the Sedan Deville (5.00 versus 4.56 for
less involved persons).

Table 5. Item-Level Tests for Three-Way Interaction Effects,

Involvement by Verbal Strategy by Brand
Item Item Description F Sig. of F
This LS brand is more
aerodynamically designed
BLF8: 3.506 .002
than other American-made
luxury sedans.
This LS brand requires
less effort to operate than
BLF9: 2.117 .050
other American-made
luxury sedans.
This LS brand is of
superior quality compared to
BLF1O: 3.014 .007
other American-made
luxury sedans.
This LS brand is priced
BFL1 1: competitively with other 2.195 .042
American luxury sedans.
Aadl: This advertisement is offensive. 2.991 .007
Aad2: This advertisement is believable. .924 .477
Aad3: This advertisement is useful .229 .967
Aad4: This advertisement is informative. 1.934 .074
Aad5: This advertisement isclear. 1.436 .199
Aad6: This advertisement is likable. 2.054 .057
Aad7: This advertisement is convincing. 2.304 .033

Highly involved respondents were significantly more positive toward (less

offended by) copy containing more information--both with and without
comparison claims. For example, consumers highly involved with the Cadillac had

file://C:\WINDOWS\Desktop\1995\SMA\95swa092.htm 5/21/04
Using Product Involvement Segmentation to Improve Advertising Productivity Page 10 of 14

a mean of 5.77 versus 3.89 for the low involvement group when processing the
noncomparative more informative copy. A lower mean indicates greater offense
concerning the advertisement. The comparative informative version was preferred
by those more involved with the Imperial (5.35 vs. 4.75). However, the less
informative noncomparative copy was ranked least offensive for the Lincoln by the
high involvement group (6.08 vs. 5.17).

The noncomparative more informative copy was most convincing for the high
involvement Cadillac group (5. 10 vs. 3.32 for the less involved group), yet the
comparative less informative treatment was more convincing for consumers
involved with the Cadillac as well (4.81 vs. 3.80). The noncomparative less
informative copy, convinced highly involved consumers most when sponsored by
the Lincoln (5.33 versus 3.72 for individuals less concerned with the Town Car),
and the comparative more informative strategy produced a greater convincing
effect on consumers involved with the Imperial (mean = 5.39) than those not so
involved (mean = 4.05).

Interestingly, consumers highly involved with the market lagging Chrysler

Imperial were more convinced by advertising which contained either comparative
statements supported by less information (4.72 vs. 3.77) or by more information
without the presence of comparison claims (4.48 vs. 3.53). This may be evidence
that consumer groups and the Federal Trade Commission (FTC) have been correct
over the years in labeling comparative advertising as more informative for
consumers under certain circumstances. In this case, the two (comparison
statements on the one hand versus a separate set of noncomparison product claims)
seem to substitute for each other in producing an enhanced convincing effect.
Finally, although not directly tested in our study, the apparent tradeoff between
comparison copy and more detailed noncomparison statements might indicate an
information overload effect. This would present one plausible explanation for why
the copy containing both comparison claims plus greater information detail failed
to produce superior results convincing respondents of the sponsoring brand's merit.

Although marginally significant (p = .057), it appears that a three-way interaction

occurred for the "likability" response variable. Consumers most involved with the
Sedan Deville liked both the noncomparative more informative copy (4.67 vs.
3.63) and the comparative less informative copy, (5.00 vs. 4.24) better than
individuals less inclined to purchase the Cadillac. For consumers highly involved
with the Lincoln, the less informative noncomparative copy was preferred 5.33 to
4.17, yet oddly enough (since these two copy versions are opposite extremes),
those more apt to purchase the Town Car also liked the comparative more
informative copy 4.83 to 3.60 for the less involved group. The most informative
comparison advertisement was clearly liked better by persons most involved with

file://C:\WINDOWS\Desktop\1995\SMA\95swa092.htm 5/21/04
Using Product Involvement Segmentation to Improve Advertising Productivity Page 11 of 14

the Chrysler Imperial compared to consumers less interested (5.09 versus 3.90).

Finally (p =.074), the more informative comparison copy was rated most
informative for the Town Car and the Imperial by those most likely to purchase
that brand (5.78 and 5.26 respectively). The noncomparative more informative
copy was perceived as most informative for the Sedan Deville by the high
involvement group (5.73 vs. 4.00 for less involved consumers).

Conclusion

Wells (1993) identified a need to bring consumer behavior research more in line
with the needs of practicing marketers. We seek to do so by clearly demonstrating
the ability of segmentation based on product involvement to differentiate in
meaningful ways among groups of consumers with homogeneous income levels.
Because varying verbal strategies resulted in differential effects across segments,
marketers will be able to use this methodology to identify the most effective
creative strategies for that group of consumers pre-qualified along income
dimensions who are most positive toward purchasing that brand.

Future Research

Future studies should address the deficiencies here (e.g., generalizability to other
product categories). Reliability measures were mostly acceptable for an
exploratory investigation, but coefficient alphas should be in the upper range for
managerial decision-making (Nunnally 1978). Scale refinement efforts specifically
directed toward segmentation applications are warranted, and tests should be
performed to determine if a downward trend exists in reliability from the high/high
to the low/low segments.

Structural equation modeling within and across segments presents opportunities for
future research. One question left open concerns the nature of the influence
comparative brand beliefs exert during brand attitude and purchase intention
formation, and if (and why) they "crowd out" noncomparison beliefs in the
process. Would this have occurred if a less involving product form had been used
or if the sponsoring brand was a new entrant to the market?

Finally, the strong behavioral orientation of involvement segmentation enables

path analysis for entire promotion mixes as well. As penetration rates for electronic
media expand, marketers might find involvement segmentation procedures to be
most effective for message testing, particularly for interactive forms of persuasion.

References

file://C:\WINDOWS\Desktop\1995\SMA\95swa092.htm 5/21/04
Using Product Involvement Segmentation to Improve Advertising Productivity Page 12 of 14

Antil, John H., Conceptualization and Operationalization of Involvement, in

Advances in Consumer Research, T.C. Kinnear, ed., Association for Consumer
Research, 11 1984, pp. 203-9.

Barry, Thomas E., Comparative Advertising: What Have We Learned In Two

Decades? Journal of Advertising Research, 33 (March/April, 1993): 19-29.

Beatty, Sharon E., Pamela Homer and Lynn R. Kahle, The Involvement-
Commitment Model: Theory and Implications, Journal of Business Research, 16
(2, 1988): 149-67.

Belk, Russell W., Effects of Gift-Giving Involvement on Gift Selection Strategies,

in Advances in Consumer Research, A. Mitchell, ed., Association for Consumer
Research, 9, 1982, pp. 408-12.

Bloch, Peter H. and Marsha L. Richins, A Theoretical Model for the Study of
Product Importance Perceptions, Journal of Marketing. 47 (Summer, 1983): 69-81.

Bolfing, Claire P., Integrating Consumer Involvement and Product Perceptions

With Market Segmentation and Positioning Strategies, Journal of Consumer
Marketing, 5 (2, 1988): 49

Bowen, Lawrence and Steven H. Chaffee, Product Involvement and Pertinent

Advertising Appeals, Journalism Quarterly, 51 (Winter. 1974): 613-21, 644.

Bunn, Michele D., Taxonomy of Buying Decision Approaches, Journal of

Marketing, 57 (1, 1993): 38-56.

Celsi, Richard L. and Jerry C. Olson, The Role of Involvement in Attention and
Comprehension Processes, Journal of Consumer Research. 15 (September, 1988):
210-24.

Darnay, Arsen J., Automotive Performance Market by Vehicle, Market Share

Reporter, Detroit, MI, Gale Research, Inc, 1991.

Flynn, Leisa Reinecke and Ronald E. Goldsmith, Application of the Personal

Involvement Inventor), in Marketing, Psychology & Marketing, 10 (4, 1993): 357-
66.

Gardial, Sarah F., David W. Schumann, Ed Petkus, Jr., and Russell Smith.
Processing and Retrieval of Inferences and Descriptive Advertising Information:
The Effects of Message Elaboration, Journal of Advertising, 22 (1, 1993): 25-34.

file://C:\WINDOWS\Desktop\1995\SMA\95swa092.htm 5/21/04
Using Product Involvement Segmentation to Improve Advertising Productivity Page 13 of 14

Gensch, Dennis H. and Rajshekhar G. Javalgi, The Influence of Involvement on

Disaggregate Attribute Choice Models, Journal of Consumer Research, 14 (June,
1987): 71-8 1.

Heslin, Richard and Blair T. Johnson, Prior Involvement and Incentives to Pay
Attention to Information, Psychology & Marketing, 9 (3, 1992): 209-19.

Holmes, John H. and Kenneth E. Crocker, Predispositions and the Comparative

Effectiveness of Rational, Emotional and Discrepant Appeals for Both High
Involvement and Low Involvement Products, Journal of the Academy of Marketing
Science, 15 (1, 1987): 27-35.

Homer, Pamela M. and Lynn R. Kahle, Source Expertise, Time of Source

Identification, and Involvement in Persuasion: An Elaborative Processing
Perspective, Journal of Advertising, 19 (1, 1990): 30-9.

Laczniak, Russell N. and Les Carlson, Examining the Influence of Attitude-

Toward-the-Ad on Brand Attitudes, Journal of Business Research, 19 (1989): 303-
311.

Lastovicka, John L. and David M. Gardner, Components of Involvement, in

Attitude Research Plays for High Stakes, J.C . Maloney and B. Silverman, eds.,
Chicago: American Marketing Association, 1979, pp. 53-73.

Leigh, James H. and Anil Menon, Audience Involvement Effects on the

Information Processing of Umbrella Print Advertisements, Journal of Advertising,
16 (3, 1987): 3- 12.

Miller, Cyndee, Author Says Middle Class Buys Most Luxury Goods, Marketing
News, (February 18, 1991): p. 6.

Miniard, Paul W., Sunil Bhatla, Kenneth R. Lord, Peter R. Dickson, and H. Rao
Unnava, Picture-based Persuasion Processes and the Moderating Role of
Involvement, Journal of Consumer Research, 18 (June, 1991): 92-107.

Mittal, Banwari, Measuring Purchase-Decision Involvement, Psychology &

Marketing, 6 (2, 1989): 147-62.

Neese, William T. and Ronald Taylor, Verbal Strategies for Indirect Comparative
Advertising, Journal of Advertising Research, 34 (2, 1994): 56-69.

Nunnally, Jum C., Psychometric Theory, Second Edition, New York: McGraw-Hill

file://C:\WINDOWS\Desktop\1995\SMA\95swa092.htm 5/21/04
Using Product Involvement Segmentation to Improve Advertising Productivity Page 14 of 14

Book Company, 1978.

Petty, Richard E. and John T. Cacioppo, Issue Involvement as a Moderator of the

Effect on Attitude of Advertising Content and Context, in Advances in Consumer
Research, K.B. Monroe,ed., Association for Consumer Research, 8,1981, pp. 20-4.

Pucely, Marya J., Richard Mizerski and Pamela Perrewe, A Comparison of

Involvement Measures for the Purchase and Consumption of Pre-Recorded Music,
in Advances in Consumer Research, M.J. Houston, ed., Association for Consumer
Research, 15, 1988, pp. 37-42.

Richins, Marsha L. and Peter H. Bloch, Post-Purchase Product Satisfaction:

Incorporating the Effects of Involvement and Time, Journal of Business Research,
23 (1991): 145-58.

Stone, Robert N., The Marketing Characteristics of Involvement, in Advances in

Consumer Research, T.C. Kinnear, ed., Association for Consumer Research, 11,
1984, pp. 210-15.

Sujan, Mita and Christine Dekleva, Product Categorization and Inference Making:
Some Implications for Comparative Advertising, Journal of Consumer Research,
14 (December, 1987): 372-8.

Taylor, Alex III, The New Golden Age of Autos, FORTUNE, (April 4, 1994): 51-
66.

Wells, William D., Discovery-Oriented Consumer Research, Journal of Consumer

Research, 19 (March, 1993): 489-504.

Wilkie, William L. and Paul W. Farris, Comparison Advertising: Problems and

Potential, Journal of Marketing, 39 (October, 1975): 7-15.

Wright, Peter L., The Cognitive Processes Mediating Acceptance of Advertising,

Journal of Marketing Research, 10 (1, 1973): 53- 62.

Zaichkowsky, Judith L., Measuring the Involvement Construct, Journal of

Consumer Research, 12 (December, 1985): 341-52.

Zaichkowsky, Judith L., Conceptualizing Involvement, Journal of Advertising, 15

(2, 1986): 4-14, 34.

file://C:\WINDOWS\Desktop\1995\SMA\95swa092.htm 5/21/04