|  
Analysis of Palestinian Public Opinion on Politics
Popular Trust in Palestinian Islamist Factions
(Published by: JMCC Written by:
Gil Friedman, pp. 46 September 2000)
Contents
Preface
Since 1993, the Jerusalem Media and Communication Center
(JMCC) has been regularly conducting public opinion polls. These surveys
encompass a wide range of subjects that are of interest to the Palestinian
public. It also draws the interest of non-Palestinian parties who are,
directly or indirectly, engaged with the Palestine question.
Furthermore, the JMCC Polling Unit conducts commissioned
surveys for researchers whose research and analysis require an examination
of public opinion.
The unit has participated in joint surveys such as regional
polls on an Arab level and another poll with an Israeli research center.
Throughout the course of the periodic polls, it has been
realized that there is inadequate usage of the accumulating technical data.
This realization has led us to expand the polling unit to include data
analyses that are intended to help government officials, political activists,
researchers, journalists, and any other interested people, comprehend Palestinian
attitudes towards the issues that are tackled by the polls. The JMCC has
previously published four analytical reports. The present study is the
second of three complementary public opinion analyses conducted by the
author to do with popular political trust.
Palestinian public opinions on the peace process and on
the Palestinian leadership who are ingrained within this process, are two
of the most important subjects that JMCC surveys have tracked since the
beginning of this political process and the return of the Palestinian leadership.
The most important trend that is clearly demonstrated
by the polls is the continuous and steady increase within the Palestinian
public in the distrust of all leading Palestinian political figures and
factions.
The analysis presented herein studies the Palestinian
people’s level of trust in Islamic factions. Among its various insights,
this analysis confirms the widely-held belief that Islamist factions comprise
the most popular form of opposition to the PA.
Accordingly, the polls and the analytical studies constitute
a worthy contribution in empowering the trust of the people in themselves
as well as reinforcing accountability within the discourse of democratization
of the Palestinian society.
Ghassan Khatib
Director
Introduction
Jerusalem Media & Communication Center public
opinion polls confirm the widely-held belief that Islamist factions comprise
the most popular form of opposition to the PA and Fatah. Graph #1 shows
that Hamas (represented with the dot-studded curve) is the second most
trusted faction after Fatah (represented with the diamond-studded curve)
in the West Bank, Gaza, and East Jerusalem. In fact, Hamas by itself receives
more support than do the PFLP, DFLP, FIDA, and PPP combined (represented
with the asterisk-studded curve) in all of the 14 polls conducted by the
JMCC from May 1995 to May 1998. Across JMCC surveys dating from January
1994 to May 1998, 11.9% of respondents polled reported trusting Hamas most,
and 6.5% reported trusting the PFLP, DFLP, FIDA, or PPP most. Of course,
as is illustrated in Graph #2, when trust in Hamas and trust in Islamic
Jihad are taken together, the Islamist position within the Palestinian
political spectrum is even more prominent. The average level of combined
trust in Hamas and Islamic Jihad across the JMCC surveys conducted from
January 1994 to May 1998 is 14.1%.
Needless to say, popular trust is a crucial source of
factional political power. Equally certain is that Islamist political orientations
diverge dramatically from those of the PA and Fatah in terms of both internal
Palestinian politics and Palestinian-Israeli relations and for that matter
Palestinian-Western and Palestinian-Middle Eastern relations. Furthermore,
the strength of the Islamic bloc may grow in the face of continued Israeli
intransigence. (See, e.g., Shikaki 1998) Accordingly, examination of the
factors accounting for popular trust in Islamist factions is warranted.
In this spirit, this study analyzes data on public
trust in Palestinian Islamist factions collected from three JMCC public
opinion polls. This study subjects these data to a statistical technique
called logistic regression, in order to estimate the impact of particular
independent variables – including demographic variables and variables tapping
attitudes on the peace process, Arafat and the PA, and political Islam
– on the probability that a person trusts an Islamist faction more than
any other faction. Among the more notable findings are the following. Contrary
to the claim of some observers that political Islam plays a relatively
minor role in drawing supporters to Islamist factions (see, e.g., Budeiri
1995: 93; Usher 1995: 75), pro-political Islamic sentiment plays an important
role in this regard. Though support for armed struggle increases the likelihood
of trust in Islamists, preferences regarding suicide bombings do not appear
to exert an important impact on trust in Islamists. This may suggest that
Islamists are more closely identified with armed struggle than with suicide
operations per se. Skepticism about the prospects for peace, Netanyahu’s
commitment to signed agreements, and American objectivity, contributes
to trust in Islamists. As this skepticism is high among secular opposition
factions as well as Islamist factions, the finding that such skepticism
is positively associated with trust in Islamists points to the withering
of the secular opposition in the Oslo era. Lastly, East Jerusalem residents
appear to be less likely than West Bankers and Gazans to trust Islamists
most.
This analysis is divided into two parts. Part I
discusses the data and methods employed in this study. Literature on Palestinian
Islamists will be invoked to guide the construction of the dependent variables
and to identify factors potentially affecting trust in Islamists. Part
I places in bold the key ideas in order to facilitate a quick understanding
of its key points. Part II summarizes and analyzes the results of six distinct
logistic regression analyses of models of trust in Islamists. As Part II
both focuses on the main empirical findings of the study and is generally
accessible to the non-methodologically inclined reader, this part does
not resort to the strategy of bolding the key ideas. Rather, the reader
is encouraged to read Part II in its entirety. The paper concludes with
suggestions for future survey research on public trust in and support for
Palestinian Islamists.
Conclusion
As this study accounted for popular trust in Islamists
with a research design that employed three different representative public
opinion polls, two different coding schemes of the dependent variable,
and a widely-respected method of model estimation, i.e., logistic regression,
we can have a fair degree of confidence in the validity of the study’s
findings. At the same time, this study did not attend to some considerations
that may enhance our understanding of trust in Islamists. Accordingly,
this paper concludes by pointing out some suggestions for future research
that may contribute to our understanding of trust in and support for Palestinian
factions.
As concerns the independent variables, perhaps
the most obvious point is that any given survey intended to be used to
explain trust in Islamists, and trust in other Palestinian factions, for
that matter, should aim to collect data on all of the factors thought to
be relevant in this regard. Judging by the findings of this study, surveys
intended to explain factional support must include factors on preferences
regarding Israel, attitudes on political Islam, and evaluations of the
PA and Arafat, or whomever follows Arafat. On the topic of including all
relevant variables, additionally, it may be desirable to collect data tapping
feelings of relative deprivation and/or anomie. One might view the JMCC
question on “optimism about the future in general” to tap the attribute
of anomie, but responses to this question were not found to exert a statistically
significant impact on trust for Islamists in preliminary models analyzed
by this author, and, regardless, it is doubtful that this question adequately
captures either of the sui generis attributes of relative deprivation and
anomie. It might also be worthwhile to solicit data capturing levels of
trauma experienced either during the Intifada, or at the hands of Israelis
more generally.
With respect to the collection of data on preferences
regarding Israel, it might be worthwhile to devise questions that discriminate
between different forms of armed struggle. Though the limited performance
of the question on suicide bombings in this study points to the possibility
that the reliability of data on highly sensitive preferences may be suspect,
explicitly differentiating between support for violence against settlers
and/or military personnel, on the one side, and support for violence against
civilians within the Green Line, on the other side, may be useful for differentiating
between supporters of Islamists, on the one side, and supporters of secular
opposition factions and elements of Fatah, on the other side.
Regarding preferences pertaining to political Islam,
this study has suggested that survey questions demanding that the respondent
prioritize political Islam in relation to other values may be more adept
in tapping respondents’ effective commitment to political Islam than are
general questions that solicit respondents’ attitudes toward political
Islam in vacuo. In this regard, both the question soliciting preferences
over Shari’a and secular law and the question soliciting data on the most
important issue facing Palestinian society are effective. Survey research
on trust in Islamists, on this note, may benefit from asking respondents
to rank the relative importance of the various issues consistently included
in close-ended “most important issue” questions. The reason for this is
simply that rank-ordering of fundamental issues provides a measure, if
only a crude one, of the extent to which religion (as well as other individual
issues) is or is not a crucial issue to the respondent. For that matter,
it may be useful for “most important issue” questions to include a follow-up
question asking the respondent to select from a group of actors – i.e.,
PA, Israel, America, Palestinian secular opposition, and Islamist opposition
– the actor or actors (s)he holds most responsible for failures regarding
the issue(s) (s)he indicated as most important. The reason that this follow-up
question is relevant is simply that Palestinians’ sentiments toward particular
factions may in part depend on who they blame for the most important problem(s)
in Palestinian society.
With respect to demographic factors, suffice it
to say that attention to the possibility of curvilinear relationships and
interactive effects may clarify and indeed make more pronounced the effects
of such variables as age, level of education, and income. Thus, for example,
though this study did not find level of income to consistently exert a
statistically significant impact on trust in Islamists, analysis sensitive
to the possibility that lower levels of income exert a stronger impact
than higher levels of income may reveal more robust findings regarding
this variable.
Finally, one limitation of the dependent variables
analyzed in the present study is that they were dichotomous, or binary,
and that, moreover, the majority of “0” responses referred to Fatah supporters.
These dependent variables were constructed by converting responses to an
open-ended survey question asking “which faction do you trust the most?”
into one of two responses, i.e., trust Islamists most and do not trust
Islamists most. While soliciting open-ended data on respondents’ most preferred
faction has various benefits, such data may not be ideal for modeling the
factors that account for trust in particular factions. For one, given the
withering of the secular opposition in the Oslo era, the open-ended trust-most
survey question lends itself to only two categories sufficiently large
to subject to statistical analyses – Islamists and non-Islamists,
with the latter category comprised primarily of people trusting Fatah most.
In analyses of the probability of trusting Islamists most with such a dependent
variable, we thus can estimate the factors that differentiate Islamist
supporters from Fatah supporters, but cannot estimate the factors that
differentiate between support for Islamists and support for Leftist factions.
In contrast, soliciting ordinal-level sentiments
on each faction enables analysis of the extent to which the types of factors
emphasized in this study account for trust in Islamists, Fatah, and Leftist
factions. Additionally, collection of ordinal-level data on attitudes toward
particular factions enables the specification and estimation of models
of trust in or support for particular factions that attend to the possibility
of both indirect and reciprocal relationships, among the independent variables,
on the one side, and factional trust or support, on the other side.
|
|
