#6588
#12 - JRL 6587
Young, Educated, Urban -and Anti-American
Recent Survey Data from Russia
Theodore P. Gerber
University of Arizona
Sarah E. Mendelson
CSIS
PONARS Policy Memo No. 267
October 2002
[DJ: The full set of recent PONARS papers can be found at: http://www.csis.org/ruseura/ponars/policymemos/pm_index.htm]
Despite claims by the media that anti-American sentiment is growing rapidly in Russia, recent survey data show most Russians are neutral toward Americans and, by implication, the United States. Those who do express a point of view, however, are about twice as likely to be anti-American as pro-American. Moreover, groups expected to be more pro-American-the young, the highly educated, and residents in the capital cities-are in fact more likely to hold anti-American views. Income and occupation have no direct affect on views toward Americans, but the economic performance and the spread of market institutions in one's region of residence have predictable effects. Ethnicity, religion, and media use also shape Russians' orientations toward Americans. Altogether, the data demonstrate that claims of rampant anti-Americanism are vastly exaggerated. However, policymakers should nonetheless be concerned, because anti-American sentiment appears to be more widespread among the young and among elites. Therefore, its political influence may be greater than its numerical strength implies.
Using Survey Data to Evaluate How Russians View Americans
Some press reports suggest that anti-American sentiment has recently been on the rise in Russia. Incidents such as the harassment of congressional delegations by skinheads on Red Square in Summer 2002 and the occasional rock thrown at the U.S. Embassy in Moscow fuel this perception. But these events, however upsetting, offer no clue as to how widespread particular views of the United States or Americans are. Moreover, they provide little guidance regarding the question of which types of Russians are more likely to hold pro- or anti-American views. Survey data are better suited to address these issues.
To develop a more nuanced picture, we turn to data from two surveys we conducted in Russia last spring.* The survey asked respondents to indicate how they "relate" to five ethnic/national groups: Jews, Americans, Chechens, Gypsies, and Azeris. They were given five response categories:
1. With sympathy and interest
2. With no particular feelings-as to any other group
3. With annoyance, hostility
4. With mistrust, fear
5. Hard to say
The survey also contained a host of variables pertaining to the other characteristics of the respondent, which permit us to determine which factors affect Russians' views toward Americans. In addition, we merged official data on regional characteristics with the survey data, which allowed us to gauge the impact of regional variations in economic performance and the spread of market institutions.
Overall Picture: Most are Neutral
We discarded the 187 of our 4909 respondents who found it "hard to say" how they relate to Americans: this is a small proportion of the sample and we are reluctant to interpret these "don't know" responses in any particular fashion. Of the remaining 4722 respondents aged 18-64, a full 70 percent took the neutral position. Thus, our data indicate that by far, most Russians feel neutral toward Americans. Reports of burgeoning anti-Americanism are misleading, as are views that Russians are particularly pro-American for one reason or another. In our sample, 19 percent (combining those in categories 3 and 4) hold negative views toward Americans. This is almost twice the number of those with especially positive views ("sympathy and interest"), who amount to 11 percent of the sample. Thus, even though anti-American sentiment is hardly widespread-as it is shared by about only one-fifth of the Russian population-it is nonetheless true that it is about twice as common as pro-American sentiment.
Who is Anti? Who is Pro?
To see what factors influence how Russians feel toward Americans, we estimated a multivariate statistical model. The technique we used is called "multinomial logistic regression," the appropriate technique for determining the net effects of independent variables-i.e., how their effects control one another-on a dependent variable consisting of more than two discrete categories. In this model, the dependent variable is a three-category variable measuring attitudes toward Americans (positive, neutral, or negative). The independent variables included age, education, sex, size of locality of residence, dummy variables for Moscow and St. Petersburg, regional wages and unemployment, regional strength of the small private business sector (the percentage of the workforce employed by small businesses), dummy variables for Russian ethnicity and Orthodox faith, and a scale measuring media use.
We originally included measures of family income and dummy variables measuring occupation and employment status, but these turned out to have no statistically significant effects so we dropped them from the model. The nonsignificance of these variables suggests that material standing in and of itself has no net effect on views toward Americans.
To show what does affect Russians' views-and illustrate how strong these effects are-we present a set of estimates based on our model. These estimates are obtained by holding constant at their sample means all the factors that shape views toward Americans except one factor, varying that one factor, and computing the corresponding expected probabilities of falling in each of the three camps based on our statistical model. These results present a pure view of how a given variable affects views toward Americans, because the effects are purged of spurious components that result from the correlations among the independent variables.
Consider the effects of demographic factors (Table 1). Holding constant all the other variables described at their sample means (that is, estimating the distribution of views for Russians who are exactly average with respect to all other characteristics measured in our model), we find that 13 percent of those 18-29 hold positive views, 64 percent are neutral, and 23 percent hold negative views. Among those aged 30-39 with the same (average) values on all other variables, the corresponding figures are 12 percent, 74 percent, and 14 percent.
TABLE 1: NET EFFECT OF DEMOGRAPHIC TRAITS ON ATTITUDES TOWARD AMERICANS
Views Toward Americans
| Positive | Neutral | Negative | |
| Overall Means | 11 percent | 70 percent | 19 percent |
| Age | |||
| 18-29 | 13 percent | 64 percent | 23 percent |
| 30-39 | 12 percent | 74 percent | 14 percent |
| 40-49 | 10 percent | 71 percent | 19 percent |
| 50-64 | 9 percent | 71 percent | 20 percent |
| Education | |||
| College | 12 percent | 65 percent | 23 percent |
| Secondary Degree | 10 percent | 73 percent | 18 percent |
| Less Than Secondary | 13 percent | 69 percent | 18 percent |
| Sex | |||
| Women | 10 percent | 74 percent | 16 percent |
| Men | 13 percent | 66 percent | 22 percent |
Note: Cell entries are probabilities calculated from preferred multinomial regression models where all other variables are held constant at sample means. See text for details.
What do these figures mean? To begin with, they suggest that generational differences have a modest effect on views toward Americans-at least once other factors are controlled. The effect is also ambiguous: although the youngest cohort is the most likely to hold positive views toward Americans, it is also the most likely to hold negative views. Certainly, there is no evidence that younger Russians are more pro-American.
The effect of education is also relatively muted. But the most striking result here is that college-educated Russians are somewhat more likely to harbor anti-American views, compared to those with lower levels of education. Thus, the most educated segment of Russian society is far from a bastion of support for the United States: if anything, it is more likely to espouse a negative outlook on Americans.
Russian men appear somewhat more likely to hold anti-American views than Russian women. But this surely reflects women's greater tendency toward neutrality in response to surveys, as Russian men are also more likely to express positive views.
Turning now to the effects of place of residence, we are quite surprised to see that anti-American sentiment is especially strong among residents of Moscow and St. Petersburg (Table 2). These residents are nearly three times more likely than residents in other places to be anti-American. This is our most striking finding for two reasons. First, it suggests why we have seen so many reports recently about a surge of anti-American sentiment in Russia: such sentiment does appear to be rather strong in the capital cities, where domestic and international media outlets are concentrated and tend to focus their coverage (although it must be kept in mind that these effects serve as a control for all the other variables). Second, U.S. policymakers should be especially concerned about how residents in these major cities, Moscow in particular, view the United States. It does not bode well that anti-American sentiment is so high in these most urbanized-and most politically influential-areas.
TABLE 2: NET EFFECTS OF PLACE OF RESIDENCE ON ATTITUDES TOWARD AMERICANS
Views Toward Americans
| Positive | Neutral | Negative | |
| Overall Means | 11 percent | 70 percent | 19 percent |
| Locality of Residence | |||
| Moscow | 5 percent | 37 percent | 57 percent |
| St. Petersburg | 13 percent | 35 percent | 52 percent |
| Other city over 1 mil | 10 percent | 70 percent | 19 percent |
| Small city (400,000) | 11 percent | 72 percent | 18 percent |
| Large town (80,000) | 11 percent | 73 percent | 16 percent |
| Small town (40,000) | 11 percent | 73 percent | 16 percent |
| Rural village (10,000) | 12 percent | 74 percent | 15 percent |
| Small suburb (10,000) | 12 percent | 74 percent | 15 percent |
Note: Cell entries are probabilities calculated from preferred multinomial regression models where all other variables are held constant at sample means. See text for details.
Although individual-level economic well-being has no net effect on how Russians view Americans, regional economic performance has predictable effects: residence in an Oblast that has higher than average wages and/or lower than average unemployment is associated with less negative (though not especially more positive) views toward Americans, while the opposite holds for residence in a region with lower than average wages and/or higher than average unemployment. The spread of market institutions at the Oblast level also appears to matter: anti-American sentiment is more common in regions with smaller than average proportions of the labor force employed in small private businesses, and less common in those regions where small business is more widespread than average.
In addition, ethnicity, religion, and media use all affect views toward Americans (Table 3). Ethnic Russians are more likely to be neutral toward Americans-they are both less likely to hold negative views and less likely to hold positive views. The opposite holds for those who consider themselves to be Orthodox Christians: they are less neutral, more positive, and more negative. Finally, Russians who make more use of the media-and therefore might be viewed as more politically engaged-tend to be more positive and less negative in their views of Americans, while the opposite holds for those who use the media less frequently than average.
TABLE 3: NET EFFECTS OF ETHNICITY, RELIGION, AND MEDIA USE ON ATTITUDES
TOWARD AMERICANS
| Positive | Neutral | Negative | |
| Overall Means | 11 percent | 70 percent | 19 percent |
| Ethnic Russian | |||
| Yes | 11 percent | 71 percent | 18 percent |
| No | 13 percent | 66 percent | 21 percent |
| Orthodox | |||
| Yes | 12 percent | 68 percent | 20 percent |
| No | 10 percent | 73 percent | 16 percent |
| Media Use | |||
| Low | 10 percent | 69 percent | 21 percent |
| Average | 11 percent | 70 percent | 19 percent |
| High | 13 percent | 71 percent | 16 percent |
Note: Cell entries are probabilities calculated from preferred multinomial regression models where all other variables are held constant at sample means. See text for details.
Conclusions
U.S. policymakers are increasingly concerned about how populations around the world view Americans and the United States, because widespread anti-Americanism poses a security threat. Domestic or international opponents of the United States could conceivably mobilize individuals or groups within any country who dislike America. Because Russia is a key partner in the struggle against terrorism, it is especially important to understand how Russians view Americans. To assess these views, we must not rely on impressionistic reports of journalists and other observers to gauge the extent of both negative and positive views toward the United States within the Russian public. Survey data provide a systematic measure of the extent of support or antipathy toward Americans, and also tell us what types of Russians are anti- or pro-American.
Our survey demonstrates that claims of rampant anti-Americanism are blown far out of proportion: 70 percent of Russians are neutral toward Americans and an additional 11 percent have positive views. That said, some concern is justified: young, well-educated residents of Moscow and St. Petersburg are significantly more anti-American than other Russians. The relative strength of anti-American sentiment within these influential groups may account for the false impression that anti-American views are common throughout Russian society.
The greater tendency of younger Russians to harbor hostile views toward Americans is particularly disturbing because it may reflect a broader rejection of values that are intrinsic to liberal democracy. In other research we have found that, contrary to conventional wisdom, Russians under 30 are no more likely to support civil liberties than their grandparents. Support for rights of the person (freedom from arbitrary arrest, torture, and slavery) is notably weak within this age group. Those concerned with the development of democracy in Russia ought to think more about policies and programs specifically targeted at those under 30.
Back to the Top Next Issue
- Back to the Top -