Ethnocentrism in the U.S.: an examination of CETSCALE stability from 1994 to 2008.
Lumb, Ruth ; Kuperman, Jerome C.
INTRODUCTION
The consumer ethnocentrism scale (CETSCALE) developed by Shimp and
Sharma (1987) has been used by researchers in marketing strategy ever
since its introduction as a tool for measuring consumer attitudes
regarding the appropriateness of purchasing foreign produced goods and
services. Specifically, strongly ethnocentric consumers feel that buying
imported goods is wrong because it negatively affects the domestic
economy and may be viewed as unpatriotic. There is extensive empirical
work that has been done to support the link between CETSCALE scores and
intended purchase behaviors (e.g., Acharya and Elliott, 2003; Balabamis
and Diamantopoulos, 2004; Evanschitzky, Wangenheim, Woisetschlager, and
Blut, 2008; Watson and Wright, 2002). The findings are particularly
robust in studies done in Western developed economies showing a positive
correlation between ethnocentrism and the likelihood of purchasing
domestically produced products (Klein, Ettenson, and Krishnan, 2006).
Consumers in advanced economies generally take pride in their domestic
products and judge them favorably compared to foreign goods (Balabanis,
Diamantopoulos, Mueller, and Melewar, 2001, Netemeyer, Durvasula, and
Lichtenstein, 1991).
The overwhelming evidence linking CETSCALE scores to purchase
intentions in advanced economies is important and useful information to
strategists thinking about competing in markets with both foreign and
domestic competitors and whether to position their brands based on being
foreign or domestic (Klein, 2002). However, one potential weakness of
the empirical work done so far is that ethnocentrism studies are
normally done at a single point in time. Such studies allow for tests
that show significance of the independent variable at that point of
time; however, there is no way to gauge the behavioral sensitivity of
consumers to attitudinal changes over time. Nielsen and Spence (1997)
attempted to address this question by conducting surveys in the U.S.
over an eight week period during which significant "patriotic
events", as the authors termed them, had occurred (e.g., terrorist
bombings, political campaigning, and the Olympics). Lumb and Geib (2011)
also addressed this question but in a developing country context,
looking at two samples of Chinese consumers surveyed eleven years apart.
More studies need to be done over time to judge the stability of
consumer ethnocentrism (CETSCALE scores). Further, to best judge
stability, we would argue for multiple surveys conducted over the course
of years and not weeks. This paper attempts to fill in this gap in the
literature by analyzing the annual movement of CETSCALE scores in the
U.S. between 1994 and 2010. Also of note is the fact that our sample
includes data before and after the attacks of 9/11--arguably one of the
biggest "patriotic events" in recent U.S. history. So, large
shifts in attitude might be expected around this major event. This
provides an even greater opportunity to test, in the face of a major
event, the sensitivity of consumer ethnocentrism as measured by CETSCALE
scores.
THEORETICAL BACKGROUND
Globalization and Its Likely Effect on CETSCALE Stability
Globalization has been made possible by world-wide foreign direct
investment, production and marketing; advances in telecommunication
technologies and the internet; increases in world travel; the growth of
global media; and technological advances that have made it easier and
quicker to complete international transactions--both trade and financial
flows--and to acquire information about other countries (Ozsomer and
Simonin, 2004; Steenkamp & Hofstede, 2002; Stremersch & Tellis,
2004; Van Everdingen, Aghina, & Fok, 2005).
Globalization is particularly impactful in developing countries
where shifts in lifestyles and wealth accumulation can be very dramatic
compared to already developed countries. For instance, in the obvious
example of China, from 1979 to 2006, China's real gross domestic
product grew at an average annual rate of 9.7%, the size of the economy
increased over eleven-fold, and its real per capita GDP grew over
eight-fold (Congressional Research Service, 2007). Such dramatic changes
in wealth accumulation most certainly affect consumption behaviors and
attitudes over time. Hamzaoui, Essoussi and Merunka (2007) observe that
with this economic growth, there is an accompanying change in the
availability of products in local markets. At first, there is a lack of
local manufacturers and the attraction of growing markets invites
foreign manufacturers to fill the void. This phenomena may in part
explain the finding by Lumb and Geib (2011) that consumer ethnocentrism
of Chinese consumers, as reflected in CETSCALE scores, was lower in 2006
than in 1995. Of course, over time this process becomes more complicated
as some foreign brands begin to lose their appeal to Chinese consumers
as local Chinese manufacturers enter the market and their brands
increase in quality and attractiveness (Zhou and Hui 2003; Li, 2004).
We would propose that the Chinese example discussed above is
consistent with a pattern of change found throughout the developing
world. In this process, there is extreme market volatility as dramatic
changes in wealth create large changes in market competition and, by
extension, large changes in consumer attitudes about foreign products.
While globalization is impacting competition in markets all over the
globe, the quite dramatic changes that can be seen in a developing
country like China are completely different from those that would be
expected in developed countries. By contrast, wealthier countries such
as the U.S. already have very well established markets with relatively
stable competition as well as educated, wealthy consumers. Changes
brought about by globalization should be incrementally much less
impactful to both real economic activity and consumer attitudes in this
context than in the developing country context.
Patriotism, Ethnocentrism and the Durability of Attitudinal Shifts
Social identity theory in the psychology literature (see review in
Hogg, 2006) focuses on the idea that people psychologically identify
themselves as belonging to groups based on shared attributes that
distinguish them collectively. Consumer ethnocentrism reflects a high
level of social identity with the national collective and, consistent
with a strong national identity, is the feeling of patriotism, i.e., to
feel love for country and an attachment to national values (Skitka,
2005). This psychological link between the constructs of consumer
ethnocentrism and patriotism in the minds of consumers is explained by
Shimp and Sharma (1987: 280) as follows:
"From the perspective of ethnocentric consumers, purchasing
imported products is wrong because, in their minds, it hurts the
domestic economy, causes loss of jobs, and is plainly unpatriotic;
products from other countries (i.e., out groups) are objects of contempt
to highly ethnocentric consumers."
Empirical work done in the marketing literature has largely
supported that there is in fact a strong relationship between consumer
ethnocentrism and both patriotism and Hofstede's (1984)
collectivism dimension. In Shankarmahesh's (2006) review of the
literature that has studied the antecedents to consumer ethnocentrism,
he found four studies that looked at the collectivism variable and six
studies that looked at patriotism. In his summary, he identifies all
four studies that looked at collectivism as having found a positive
relationship between collectivism and consumer ethnocentrism while five
of the six studies that looked at patriotism found a positive
relationship (one study found no relationship).
Given the relationship between patriotism, reflected by a strong
sense of collective national identity, and ethnocentrism, the events of
9/11 can be hypothesized to be a particularly relevant variable
affecting U.S. consumer ethnocentrism. Terror Management Theory (TMT),
as discussed by Thomas (2003: 859), "holds that individuals
initially employ simple, direct defenses to banish death-related
thoughts." One defense is an increased identification with and
loyalty to one's worldview. In the case of 9/11, Thomas (2003)
found heightened levels of patriotism in her survey of mid-life American
women. Thomas' (2003) findings are also supported by other survey
research done immediately following the 9/11 attacks on the World Trade
Centers and the Pentagon that also indicated Americans had an increased
sense of patriotism following the attacks (Davies, Steele and Markus,
2008; Skitka, 2005; Traugott, et. al., 2002). The survey done by
Traugott, et. al. (2002: 514) through the Institute for Social Research
found that changes in attitude tended to "reflect more patriotism
and national pride than an opposition to foreigners" noting the
result as "a kind of patriotism of mutual support rather than a
jingoistic reaction to all foreigners or even immigrants." This
evidence suggests that Americans should have become more ethnocentric in
the year immediately following the attacks in large part because of an
enhanced sense of patriotism and a greater sense of collective
belonging.
Of course, while no event in recent American history compares with
9/11, there are still many events that occur over time that could be
considered as well as 'patriotic' events powerful enough to
affect ethnocentrism levels. There is always the possibility that any of
these other events could also have affected people's sense of
patriotism and ethnocentrism just as we have hypothesized for 9/11. The
study conducted by Nielsen and Spence (1997: 70) identified the summer
of 1992 as having "several events" that could affect
patriotism noting that "given the recent spate of terrorist
bombings, political campaigning, and events such as the Olympics, the
effect of patriotism should not be ignored." The findings from
Nielsen and Spence's (1997) study as they tracked CETSCALE scores
over eight weeks were somewhat inconclusive. They found that scores were
stable for the population but not so for two specific
sub-groups--military versus non-military personnel. Our sample will
include a number of events similar to the events identified by Nielsen
and Spence (1997) including the Oklahoma City bombing, multiple
political campaigns, and the tech stock market bubble.
The durability of attitudinal changes following traumatic events is
a complex variable to assess. Some events will tend to produce more
permanent reactions compared to other events. There is extensive work
done in psychology to deal with stress disorders following traumatic
events and issues of recovery In their review of the associated
literature, Aldwin, Sutton, and Lachman (1996) note that "much of
the literature supports the notion that the effect of stressors on
mental and physical health is relatively short lived (p. 841)."
However, they also observe following their own experimental study that
"under highly stressful circumstances, deviation amplification
processes may occur that result in long-term changes in personality (p.
866)." Basically, coping mechanisms lead to no lingering effects in
many cases; but, more devastating events require internal coping
processes that have more long-lasting personality impacts. Any of the
events that we have identified as potential patriotic events, including
the 9/11 attacks, were not personalized events with important impacts on
the daily lives of most Americans. Comparatively, such events are of
minimal impact when contrasted with things like being assaulted or
facing the unexpected death of a loved one and, as noted in Aldwin, et.
al., (1996), such effects are often very short-term in nature. We would
therefore suggest that any observable attitudinal changes in terms of
ethnocentrism should similarly be only short-lived; immediate reactions
that would not persist into the long-term.
RESEARCH QUESTIONS
Given the preceding, we present the following research questions:
Research Question 1: Do CETSCALE scores in the U.S. remain stable
with little variation over time?
Research Question 2 Consistent with heightened levels of patriotism
felt by Americans following the 9/11 attacks, did CETSCALE scores in the
U.S. increase immediately following the 9/11 attacks?
Research Question 3: Did CETSCALE scores following the 9/11 attacks
return towards pre-attack levels?
RESEARCH DESIGN AND SAMPLE CHARACTERISTICS
Data Collection
Survey research techniques were utilized to obtain the data for the
study from a convenience sample. The total sample consisted of a total
of 949 respondents, the majority of who were college students, who
participated in the study. The 1994 and 2005 samples were from a larger
study and not limited to college students. The questionnaire contained
demographic information questions and the Consumer Ethnocentrism Scale,
which is a measure of respondents' ethnocentric tendencies.
Respondents completed the 17-item CETSCALE on a scale of 1 (strongly
disagree) to 7 (strongly agree). The sum of the responses on each of the
17 items indicates the respondent's ethnocentric tendency with a
higher score indicating a higher ethnocentric tendency. In addition to
the yearly data, samples were found for both the Fall and Spring term
classes in 2001 and 2002, allowing for a comparison of CETSCALE scores
immediately before and after 9/11.
Respondent Characteristics
Age
For all the samples the largest percentage of respondents placed
themselves in the 17-25 year old age category: 95.2% of the total number
of respondents is between the ages of 17-25 years. As this was mainly a
convenience sample of college students, this age category was expected
to be the largest.
Gender
The number of respondents who disclosed their gender was 489 males
and 377 females. To determine if there was significant difference in the
number of male and female respondents a Chi-Square statistic was
computed ([chi square] = 18.80713, df = 18, prob. =.4038), indicating
that there was not a significant differences in the number of males and
female respondents.
Student Sample
The results of this study should be viewed in light of the
convenience sample of students. Some researchers have found that student
samples may suffer from a lack of external validity (e.g., Sears, 1996)
although external validity concerns have been found to require more
attention in theoretical research studies (Winer, 1999; Lynch, 1999). It
is possible that different samples of the population may have higher or
lower CETSCALE Scale scores than the students sampled. However, the use
of student samples in this study can be plausibly rationalized since the
9/11 event was also experienced by the students.
ANALYSIS
Research Question 1: Do CETSCALE scores in the U.S. remain stable
with little variation over time?
In order to answer this question, the mean CETSCALE score for each
group (i.e., year) was calculated. Figure 1 graphically shows the mean
CETSCALE score by year. Table 1 presents the specific sample data for
each of the year's in the study.
[FIGURE 1 OMITTED]
In order to determine if CETSCALE scores remain stable with little
variation over time, a t test between the mean CETSCALE score for
successive years was conducted. The results of the t tests indicate that
there are statistically significant differences between many of the
successive years. A statistically significant difference for the
CETSCALE scores was not found for the following samples: 1994-1995;
1998-1999; 2001 Fall -2002 Spring; 2003-2004; 2005-2006; 2008-2009;
2009-2010.
It seems important at this point to remember that statistical
significance is not the same as practical significance. Strategists
recognize that attitudinal adjustments are always occurring within the
population; but, they also recognize that not all adjustments are
significant shifts. A visual examination of Figure 1 indicates that
although t tests indicate some statistically significant differences
between the CETSCALE scores for a number of successive years, there are
only two dramatic spikes during the time period of the current study.
Otherwise, scores appear to the naked eye to be actually quite stable
outside of those two spikes. One spike can be, as we anticipated,
attributed to the 9/11 terrorist attack. The other spike, similar in
scale to that found following the attacks on 9/11, occurs in 1996.
Statistically speaking, it may be that this one data point is simply a
random outlier. But, the other possibility is that some other
'patriotic' event(s) in 1996 may have occurred that
contributed to the spike. While no event in recent American history
compares with 9/11, it is interesting that 1996 did have two acts of
terrorism that affected U.S. civilians. On June 27th of that year a
truck bomb exploded outside the U.S. military barracks in Saudi Arabia
killing 19. The Atlanta Olympic Games began July 20th and on July 27th a
bomb exploded at Atlanta Olympic Park killing 1 and injuring 110.
It is impossible to know if either of these events contributed to
the 1996 spike, but it is certainly very reasonable to consider that
they might have. However, we cannot know for sure which types of events
or what specific aspects of the events might contribute to spikes. While
1996 did witness two terrorist attacks which were prominently covered by
the news media, it was not the only year where terrorist attacks
happened (e.g., 1995 Oklahoma City car bombing outside a federal office
building; attack of the U.S. Navy destroyer USS Cole in 2002; attacks on
Western targets in Saudi Arabia in 2003 and 2004). It is not clear what
is exactly unique about the events in 1996 that would have made them
more like 9/11 but still different from any of the other events. Some
have argued that because of the constant television coverage and the
opportunity for people to continually reminded of the horrific
experience by the media, "this event [9/11 attacks] was unique in
the number of people exposed and the degree to which it was perceived as
personal (Meisenhelder, 2002:771)." Maybe the extensive television
coverage associated with the Olympics is the common thread.
Research Question 2: Consistent with heightened levels of
patriotism felt by Americans following the 9/11 attacks, did CETSCALE
scores in the U.S. increase immediately following the 9/11 attacks?
In order to answer this research question, two statistical
procedures were conducted. First, to determine if there are
statistically significant differences between respondents' CETSCALE
scores in the years before and after 9/11, the sample was divided into
two groups, pre and post 9/11. The data were analyzed using Analysis of
Variance (ANOVA). The results of the ANOVA (F=15.5743, df = 1,902, prob.
=.0001) indicate that there is a statistically significance difference
in the CETSCALE scores between the pre 9/11 and post 9/11 groups.
Second, to determine if the CETSCALE scores increased immediately
following the 9/11 attacks, Spring 2001 and Fall 2001 samples were
subjected to a Student's T-Test for a comparison of the means. The
results, which are presented in Table 2, indicate that CETSCALE scores
in the U.S. did increase immediately following the 9/11 attacks. This
suggests that Americans became more ethnocentric in the year immediately
following the attacks and, as previously stated, this may be in large
part because of an enhanced sense of patriotism and a greater sense of
collective belonging.
Research Question 3: If they changed following 9/11, did CETSCALE
scores following the 9/11 attacks return to pre-attack levels?
In order to answer this research question, t tests between
successive yearly means were performed. The results of the t tests
indicate a significant difference between the means for the 2002 Spring
and 2002 Fall samples (t = 14.53, df = 77.65, prob =. 000). As indicated
in Table 1, the Mean CETSCALE score of 93.11 for the 2002 Spring sample
is significantly higher than the Mean CETSCALE score of 47.21 for the
2002 Fall sample. It is possible that in the spring of 2002 people were
still recovering from 9/11 and by fall the recovery process had started.
The t test between the 2002 Fall and 2003 samples (t = -3.01, df =
57.39, prob = .004) indicates a significant difference between the
CETSCALE score for the two years. For the years between 2003--2005,
there is not a significant difference in the CETSCALE score between the
20032004 samples (t = 0.39, df = 57.05, prob = .695) while and between
the 2004 and 2005 samples there is again a statistically significant
difference between the CETSCALE scores (t = -3.68, df = 58.69, prob. =
.001). It appears that CETSCALE scores following the 9/11 attacks did
return to pre-attack levels. As indicated in Figure 1, the CETSCALE
scores then increased for a few years, although they did not come close
to reaching the spikes present in 1996 or 2001.
CONCLUSION
The larger implications of this study for business strategy is to
reinforce in the context of a large Western economy, the U.S., that even
though ethnocentrism levels do fluctuate year to year, they are
basically stable within a range over longer periods of time. People did,
as expected, react strongly to the attacks on 9/11; still, the reactions
were very short-lived suggesting that fundamental underlying attitudes
were probably unaffected by the event and people needed just a short
amount of time to cope and adjust. While the 1996 spike is harder to
evaluate in terms of its causes, it also had a very short-term effect
that was quickly dissipated. We cannot know without further
investigation if fundamental attitudes in other Western developed
economies are similarly stable, but it certainly is a reasonable
hypothesis to think so. Foreign companies that sell their products in
these markets should obviously be aware of the ethnocentrism variable
and plan for it. But, the encouraging news for business that comes from
this study is that they can count on the stability of attitudes over
time as they compete in these markets over the long-term.
REFERENCES
Acharya, C. & G. Elliott (2003). Consumer ethnocentrism,
perceived product quality and choice--an empirical investigation.
Journal of International Consumer Marketing, 15(4), 87-115.
Aldwin, C.M., K.J. Sutton & M. Lachman (1996). The development
of coping resources in adulthood. Journal of Personality, 64(4),
837-871.
Balabanis, G. & A. Diamantopoulos (2004). Domestic country
bias, country-of-origin effects, and consumer ethnocentrism: A
multidimensional unfolding approach. Academy of Marketing Science
Journal, 32(1), 80-95.
Balabanis, G., A. Diamantopoulos, R.D. Mueller & T.C. Melewar
(2001). The impact of nationalism, patriotism and internationalism on
consumer ethnocentric tendencies. Journal of International Business
Studies, 32(1), 157-175.
Chan, K., T. Chan & L. Leung (2010). How consumer ethnocentrism
and animosity impair the economic recovery of emerging markets. Journal
of Global Marketing, 23, 208-225.
Davies, P., C. Steele & H. Markus (2008). A nation challenged:
The impact of foreign threat on America's tolerance for diversity.
Journal of Personality and Social Psychology, 95(2), 308-318.
Evanschitzky, H., F. Wangenheim, D. Woisetschlager & M. Blut
(2008). Consumer ethnocentrism in the German market. International
Marketing Review, 25(1), 7-32.
Hamzaoui, L., & D. Merunka (2006). The impact of country of
design and country of manufacture on consumer perception of bi-national
products' quality: An empirical model based on the concept of fit.
Journal of Consumer Marketing, 23(3), 145-155.
Herche, J. (1992). A note on the predictive validity of the
Cetscale. Journal of the Academy of Marketing Science, 20(3), 261-264.
Herche, J. (1994). Ethnocentric tendencies, marketing strategy and
import purchase behavior. International Marketing Review, 11(3), 4-16.
Hofstede, G.H. (1984). Culture's consequences, international
differences in work related values. Beverly Hills, CA.: Sage
Publications.
Hogg, M.A. (2006). Social identity theory. In P.J. Burke (Ed.),
Contemporary Social Psychological Theories (pp. 111-136). Stanford, CA:
Stanford University Press.
Klein, J.G. (2002). Us versus them, or us versus everyone?
Delineating consumer aversion to foreign goods. Journal of International
Business Studies, 33(2), 345-363.
Klein, J.G., R. Ettenson & B. Krishnan (2006). Extending the
construct of consumer ethnocentrism: When foreign products are
preferred. International Marketing Review, 23(3), 304-321.
Lynch, J. (1999). Theory and external validity. Journal of Academy
of Marketing Science, 27(3), 367-376.
Lumb, R. & P. Geib. (2011). A decade comparison: Chinese
ethnocentric tendencies. Working Paper.
Netemeyer, R., S. Durvasula & D. Lichtenstein (1991). A
cross-national assessment of the reliability and validity of the
CETSCALE. Journal of Marketing Research, 28 (August), 320-327.
Nielsen, J.A. & M.T. Spence (1997). A test of the stability of
the Cetscale, a measure of consumers' ethnocentric tendencies.
Journal of Marketing Theory and Practice, 5(4), 68-76.
Ozsomer, A. & B.L. Simonin (2004). Marketing program
standardization: A cross-country exploration. International Journal of
Research in Marketing, 21(4), 397-414.
Sears, D. (1986). College sophomores in the laboratory: Influences
of a narrow database on social psychology's view of human nature,
Journal of Personality and Social Psychology. 51(3), 515-530.
Shankarmahesh, M.N., (2006). Consumer ethnocentrism: An integrative
review of its antecedents and consequences. International Marketing
Review, 23(2), 146-172.
Shimp, T. & S. Sharma (1987). Consumer ethnocentrism:
Construction and validation of the CETSCALE. Journal of Marketing
Research, 24 (August), 280-289.
Skitka, L.J. (2005). Patriotism or nationalism? Understanding
post-September 11, 2001, flag-display behavior. Journal of Applied
Social Psychology, 35(10), 1995-2011.
Steenkamp, J. & T. Hofstede (2002). International market
segmentation: Issues and perspectives. International Journal of Research
in Marketing, 19(3), 185-213.
Stremersch, S., & G.J. Tellis, (2004). Understanding and
managing international growth of new products. International Journal of
Research in Marketing, 21(4), 421-438.
Thomas, S.P. (2003). None of us will ever be the same again:
Reactions of American midlife women to 9/11. Health Care for Women
International, 24, 853-867.
Traugott, M., T. Brader, D. Coral, R. Curtin, D. Featherman, R.
Groves, M. Hill, J. Jackson, T. Juster, R. Kahn, C. Kennedy, D. Kinder,
B.E. Pennell, M. Shapiro, M. Tessler, D. Weir & R. Willis (2002).
How Americans responded: A study of public reactions to 9/11/01. PS:
Political Science and Politics, 35(3), 511-516.
Van Everdingen, Y.M., W.B. Aghina & D. Fok (2005). Forecasting
cross-population innovation diffusion: A Byesian approach. International
Journal of Research in Marketing, 22(3), 293-308.
Watson, J.J. & K.Wright (2000). Consumer ethnocentrism and
attitudes toward domestic and foreign products. European Journal of
Marketing, 34(9/10), 1149-66.
Winer, R. (1999). Experimentation in the 21st century: The
importance of external validity. Journal of Academy of Marketing
Science, 27(3), 349-358.
Zhou, L.X. & M.K. Hui (2003). Symbolic value of foreign
products in the People's Republic of China. Journal of
International Marketing, 11(2), 36-43.
Ruth Lumb, Minnesota State University Moorhead
Jerome C. Kuperman, Minnesota State University Moorhead
Table 1: Overall CETSCALE Score
Year Mean Std Dev Cases
1994 56.043 16.7461 93
1995 50.1212 14.1989 33
1996 91.025 13.9734 40
1997 47.6316 13.6232 38
1998 59.9792 12.5774 48
1999 55.9574 13.2697 47
2000 47.3793 11.9719 58
2001 Spring 56.2692 13.6599 52
2001 Fall 98 10.3971 21
2002 Spring 93.1136 16.8788 44
2002 Fall 47.2195 11.976 41
2003 56.871 14.5275 31
2004 55.641 15.0618 78
2005 69.775 21.7792 40
2006 70.4568 23.2943 81
2007 62 24.0596 52
2008 46.322 17.1056 59
2009 46.7647 21.839 17
2010 57.7097 14.3369 31
Total Cases = 949
Missing Cases = 45 (4.7%)
Table 2: T-test of 2001 data samples pre and post 9/11
for Total CETSCALE Score
Number of Mean
Cases
Group 1 52 56.2692
Group 2 21 98.0000
Pooled Variance Estimate
F Value 2-Tail T Value Degrees of 2 Tail
Prob Freedom Prob
1.73 182 -12.58 71 . 000
Std Deviation Std Error
Group 1 13.660 1.894
Group 2 10.397 2.269
Separate Variance Estimate
F Value T Value Degrees of 2 Tail Prob
Freedom
1.73 -14.12 48.38 .000
