Iranian Public Is Not Monolithic: Iranians Divide Over Their Government But Unite on Forgoing Nuclear Weapons
May 18, 2009
Download Full Report (PDF)
By Alvin Richman, David B. Nolle and Elaine El Assal
Findings from an early 2008 World Public Opinion (WPO) survey in Iran1 suggest that the Iranian public is far from monolithic in how it views important political and social issues that will likely play prominent roles affecting the outcome of the upcoming 2009 national elections. These findings also have considerable relevance for U.S. policy toward Iran during a period in which the U.S. government is exploring the possibility and usefulness of greater openness in its dealings with Iran.
Using a statistical clustering technique called Latent Class Analysis, we have identified three separate opinion groups within the Iranian public on the basis of their differing degrees of support of the Iranian government on three measures contained in the WPO survey.2 All three measures, discussed below, tap the theoretical construct of “regime support.” We have found that the three public groups identified by using these “regime support” measures mirror the diversity of views among Iran’s political elites.
We are able to label these groups as Conservatives, Moderates and Reformers by examining the patterns of responses to these measures in each group and by borrowing from the research on factions among Iran’s political elites.3 In WPO’s January 2008 survey, we found that Conservatives comprised 45 percent of the total (N = approx. 319), compared to 24 percent for Moderates (N = approx. 168) and 18 percent for Reformers (N = approx. 124).4 The remaining 13 percent fell into a non-substantive, largely “no opinion” group (N = approx. 99). Demographically, Moderates and Reformers tend to be younger, better educated, have higher incomes, and live in larger urban areas than do Conservatives.
The three measures of “regime support” — The three major substantive groups within Iran’s public are separated and defined by their sharp and consistent divergence on three WPO questions relating to support of the Iranian government (“regime support”): (1) Satisfaction with the Iranian election process, (2) support for a free press versus the government’s right to censor destabilizing news, and (3) trust in Iran’s national government. (See Table 1 in Appendix I for the exact wording of each of these questions.) On all three questions, support for the regime declines sharply and progressively from the highest regime support (Conservatives) to the least regime support (Reformers). For example, nearly all Conservatives (97%) trust the Iranian national government “most of the time,” most Moderates (85%) trust it “some of the time,” and most Reformers (64%) trust it “rarely” or “never.” (See Table 1A in Appendix I)
The three groups also differ in the assessment of the process by which authorities generally are elected in Iran: Most Conservatives (83%) and Moderates (70%) are satisfied with the Iranian election process, while most Reformers (74%) are clearly dissatisfied with it. This is an example of sharp divergence of Reformers from the majority views of Conservatives and Moderates (Table 1B).
The three groups also differ considerably in their preferences on having media freedom versus the government’s right to censor potentially destabilizing news. Conservatives favor the government’s right to censor (by a 67-17% majority); Moderates are closely divided (48% for the government’s right to censor vs. 39% for media freedom); and Reformists favor media freedom by a large 73-18 percent majority – a reversal of the preference shown by Conservatives (Table 1C).
These sharp differences in basic attitudinal orientation among the three groups may partly reflect their differences in age, education and income. Majorities of both Moderates (58%) and Reformers (55%) are between 18-29 years of age, compared to 36 percent of Conservatives who fall into this youngest age group. Also, two-thirds or more of Moderates (66%) and Reformers (71%) have completed high school, compared to half of the Conservatives. (See Table 2)
Additional issues on which these three distinct Iranian opinion groups diverge are discussed below. These include other questions relating to Iran’s domestic institutions (e.g., approval of President Ahmadinejad) and perceptions of the United States and the American people. Two cases of “partial divergence” will also be examined, in which two of the groups, Moderates and Reformers, share fairly similar views on an issue, but these views differ sharply from those of the third group, Conservatives. An example of “partial divergence” is satisfaction with Iran’s economy, in which a majority of Conservatives voice satisfaction while most Moderates and Reformers express dissatisfaction. Then a number of areas are examined in which the views of the three groups converge, including sizeable majority opposition within all groups to Iran developing nuclear weapons and predominant support within all groups to U.S.-Iranian governmental talks on a range of issues, including Iraq.
1 This survey is based on personal interviews with a representative sample of 710 Iranians ages 16 and older, interviewed by a local professional research organization between January 13 – February 9, 2008. Additional details on the survey are available on the World Public Opinion website.
2 Latent Class Analysis (LCA) segments the sample into relatively homogeneous groups based on respondents’ answers to three questions relating to “regime support”. Additional details on our analytical approach are contained in Appendix IIA, Obtaining the Basic LCA Results.
3 The labeling for each group is based on the response distributions within each group and the logic for our labels will become evident in our subsequent analyses in the main text. Our analytical discussion is focused on these three substantively important groups. A small fourth group (approximately 13% of the sample), characterized by scattered substantive responses and disproportionately high “Don’t Know” responses, is ignored in order to concentrate on the main features of the public’s political landscape. Although many researchers have written about the major political subdivisions or factions among the Iranian elites (see, for example, the works of Shahram Chubin, Akbar Ganji, and Sanam Vakil), comparable empirical analyses of the major political orientations in representative national samples of the Iranian public seem to be in short supply. Thus, our work aims to contribute to a better understanding of this aspect of the Iranian public.
4 Our analytical strategy depends on examining patterns in the data rather than on constructing tests of significance using margins of error. Thus, most of the tables in Appendix I contain multiple, consistent measures of the single concept being addressed. Obtaining useful approximations of the margins of error (i.e., sampling errors) for our analyses covering each of the three substantive groups is not a simple process, requiring consideration of the multi-stage nature of the sampling design, as well as the magnitudes of the percentages and numerical size of each group being examined. Nevertheless, some very tentative indications of the possible margins of error can be derived by using computations based on a simple random sample formula and increasing the results by a factor of 1.41 to estimate the effects of the complex sampling design. Applying this admittedly arbitrary approach, the following margins of error are offered as tentative suggestions for the maximum variability at the 95 percent confidence level: 7.7% for Conservatives; 10.7% for Moderates; and 12.4% for Reformers.