Content uploaded by Azi Lev-On
Author content
All content in this area was uploaded by Azi Lev-On on Feb 25, 2016
Content may be subject to copyright.
16 International Journal of E-Politics, 5(3), 16-32, July-September 2014
Copyright © 2014, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
ABSTRACT
How do candidates for municipal ofce use the Internet in their campaigns, and are there differences between
candidates according to the character of the constituencies and the races? This is the rst a country-wide
study of website usage in low-visibility political campaigns for municipal ofces. Data were collected during
municipal elections campaigns that took place in Israel in three cycles between November 2007 and February
2009, in 143 different municipalities and involving almost 500 candidates across Israel. 1The paper explores
the characteristics of municipal campaigning, including the scope of website usage, the features available in
candidates’ websites, variables predicting website usage, and the perceptions of candidates regarding web-
sites’ effectiveness. While Websites were used by half of the candidates, they tended to be static and include
very few interactive features. The characteristics of the constituencies and the races were correlated with the
scope of Website usage by contenders.
Campaigning Online, Locally
Azi Lev-On, School of Communication, Ariel University, Ariel, Israel
Keywords: Campaigns, Elections, Internet, Local Campaigns, Municipal Elections, Social Media
INTRODUCTION
Studies of online campaigning have focused
almost exclusively on high-visibility candidates
and races, i.e. for presidential, parliamentary
and gubernatorial offices (Foot & Schneider,
2006; Gibson, 2004). This paper contributes
to the study of online campaigning by putting
low-visibility campaigns under the spotlight.
This paper presents and analyzes website
ownership and usage of candidates who ran for
mayor in the three cycles of municipal elections
that took place in Israel between November
2007 and February 2009.
Municipal administration in Israel is
composed of three levels: cities (in general,
municipalities with over 20,000 residents),
local councils (in general, municipalities with
2,000-20,000 residents), and regional councils
– generally comprised of a number of communi-
ties with less than 2,000 residents each. Since
the amendment to the municipal election law
in Israel was passed in 1975, the elections have
been conducted under a two-ballot system: one
for the mayor of the city or local or regional
council and one for members of the municipal
council. If no mayoral candidate gains at least
40% of the votes in the first round of the elec-
DOI: 10.4018/ijep.2014070102
Copyright © 2014, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
International Journal of E-Politics, 5(3), 16-32, July-September 2014 17
tion, a runoff between the two contenders who
received the largest share of the vote is held
two weeks later.
Israel is a promising arena to study online
campaigning, due to the widespread penetra-
tion of the Internet and its significant political
uses in the country. The June 2008 TIM survey
(conducted by TNS) measuring Internet expo-
sure and usage patterns, shows that broadband
Internet penetration in Israel is extensive: Inter-
net usage is estimated at 69% among the adult
Jewish population, and 56% among the adult
Arab population. Primary uses of the Internet
are information search (96% of users), news
reading (89% of users), as well as activities
such as watching videos (73% of users) and
shopping online (56% of users) (Cohen, 2009).
The Internet has also reached deep into the
political landscape, playing a dominant role in
national election campaigns (Lev-On, 2011),
as well as in the municipal elections in 2007-9
that are the subject of this study. Indeed, Israel
Today, a news daily, declared on its municipal
election-day issue that “the campaigns this
year were characterized by comprehensive
usage of Facebook, blogs, websites, and all the
tools the Internet has to offer.” (Israel Today,
2008). Special attention was granted to Internet
campaigning in the two largest municipalities,
Jerusalem and Tel Aviv-Jaffa. Ronen Madzini,
writing for Ynet (the leading Israeli news web-
site) about the two leading mayoral candidates
in Jerusalem, Nir Barkat and Meir Porush, was
especially impressed with (the winner) Barkat’s
seamless integration of the Internet into his cam-
paign (Madzini, 2008). But no less significant
was the Internet usage of the ultra-Orthodox
mayoral candidate, Member of Knesset Meir
Porush, whom Madzini noted used the Internet
in a “most original way”. His mayoral campaign
was launched on a “virtual press conference,”
and he maintained a constantly updated and in-
teractive website and a blog in the Israeli-based
social network TheMarker Cafe, in which he
answered users’ questions and uploaded photos
and videos with holiday greetings.
The municipal elections in Tel Aviv-Jaffa
were no less wired and attention-grabbing. Gal
Mor, also writing for Ynet, argued that “[t]he
Internet has awakened the electoral campaign
in Tel Aviv. Heated debates on blogs and
‘talkbacks’, videos on YouTube and Facebook,
Google-Bombing and other tactics were all
brought to the battle.” Much of this was due to
the work of the young campaign teams of the
two leading candidates, Dov Hanin and Ron
Huldai (the incumbent and re-elected mayor).
Mor (2008) surveyed the activities of the cam-
paigns and wrote that “both camps were present
in all the significant online arenas, responding
to news items, in Facebook, TheMarker Cafe,
forums, blogs, supporters’ websites, YouTube
and more. They left their candidate’s mark on
each significant discussion and were involved
in endless debates.” But is the picture portrayed
here representative of online municipal cam-
paigning throughout the country? This study
explores the presence and activity of contestants
for mayoral offices throughout Israel in the
municipal elections of 2007-9.
ONLINE POLITICAL
CAMPAIGNING
The merits of the Internet as a marketing and
campaigning medium have been discussed at
length. Notably, the Internet allows campaign-
ers to garner and distribute huge amounts of
information at little to no costs, and in a variety
of formats: to large audiences, or alternatively
to smaller and more distinct groups of potential
constituencies, while overcoming difficulties
Copyright © 2014, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
18 International Journal of E-Politics, 5(3), 16-32, July-September 2014
of distance, weather, and other physical detri-
ments. Moreover, an unprecedented ability to
document, measure and analyze Internet traf-
fic and usage patterns has emerged (Gibson
& Römmele, 2005; Foot & Schneider, 2006;
Howard, 2006).
Further unique advantages involve the
ability to conduct a two-way conversation with
target audiences, encourage public participation
and involvement, and generate a ‘viral’ social
transmission of campaign-related messages
among interested users. Such advantages were
evident in Obama’s victorious U.S. Presidency
campaign leading up to the November 2008
elections (Stallings-Carpenter, 2010; see also
Fernandes et al., 2010). Scholars have inquired
whether candidates and parties make efficient
use of the online tools at their disposal. Two
main approaches can be recognized regarding
the relations between “traditional” and online
campaigns. Margolis and Resnick (2000) argue
that in spite of the promises of new media,
online campaigns mostly recycle traditional
campaigning patterns, and hence bring “more
of the same” and reproduce patterns of “politi-
cal as usual”.
Gibson & Ward (2009), who summarize
the academic literature on online campaigns,
argue that the campaigns are characterized
by some common patterns: Most parties are
risk-averse in their online pursuits and their
websites mainly function as tools of information
dissemination and secondary usage of content
produced for the offline campaign. They claim
that interactive tools are used minimally and are
well-monitored. Moreover, they note that sites
become active during campaign season, and
are latent or even non-existent at other times.
But with the fast penetration of the Internet
to various political spheres and the growth of the
online “tool belt” at the disposal of campaign-
ers, some scholars have updated the theory of
“normalization” and argued that the Internet
can indeed alter the patterns of political com-
munication in campaigning – depending on
the environment and context of the relevant
campaigns (Xenos & Foot, 2005; Foot & Sch-
neider, 2006). Indeed, the Internet is increas-
ingly used to reach distinct audiences, using a
variety of platforms and technologies to tailor
different messages to various constituencies.
It is also applied increasingly for recruiting
party volunteers, members and resources (see
also Gibson & Ward, 2009). It seems that the
widespread interactive usage of the Internet in
the 2008 Obama campaign, as well as in other
campaigns during the 2008 US primaries and
congressional races may serve as milestones in
delineating this trend (see Stallings-Carpenter,
2010).
INTERNET USAGE IN LOW-
VISIBILITY POLITICAL
CAMPAIGNS
As noted above, the Internet can be capitalized
on for a variety of campaign activities, such
as recruitment of activists, division of labor
between them, fundraising, and even promoting
voter turnout on election day. Still, in political
campaigns – especially low-visibility cam-
paigns like the municipal ones in this study – the
literature demonstrates that some factors may
intervene and negatively impact candidates’
decisions regarding the scope and character
of Internet usage. Some candidates may feel
that they are sufficiently familiar to the public
and hence do not need to employ an additional
marketing channel (Herrnson, 2004; Kamarck,
2003; Strachan, 2003). Lack of awareness of
the strengths of the Internet as a marketing
and political tool may also lead candidates
Copyright © 2014, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
International Journal of E-Politics, 5(3), 16-32, July-September 2014 19
to prefer traditional channels of advertising.
Finally, the socio-demographic landscape may
also negatively impact the candidate’s decision
about using campaign websites, especially in
elections where large portions of the constitu-
ency do not use the Internet as a political tool
(Herrnson, Stokes-Brown & Hindman, 2007).
This study focuses on three core research
questions regarding municipal elections:
1. What is the impact of structural factors, i.e.
factors associated with the particular elec-
toral races (incumbency, race competitive-
ness, and size of municipality) on website
usage by candidates? (See hypotheses 1-3
below).
2. What is the impact of socio-economic char-
acteristics of the constituency on website
usage by candidates? (See hypothesis 4
below).
3. What is the scope and character of website
usage, and in particular the availability of
interactive tools? (See hypothesis 5 below).
This is the first attempt to study Internet
usage in low-visibility political campaigns on
a national scope in Israel. A small number of
studies documented and analyzed low-visibility
web campaigning elsewhere. Examples include:
Herrnson, Stokes-Brown and Hindman (2007)
studied Internet usage by US state legislative
candidates; Strachan (2003) interviewed profes-
sionals who consulted candidates in municipal
races (but did not interview the candidates
themselves); Brooks and Peterson (2008) stud-
ied constituency exposure to media campaigns
in local races in Akron, Ohio; and Gaziano and
Liesen (2009) studied Internet usage in low-
visibility campaigns in eight Illinois counties.
Unlike these studies, only Yannas and
Lappas (2004) present a country-wide view of
Internet campaigning (focusing on Greece).
These authors found a significant gap between
Internet usage in the municipal elections and
much broader Internet uses in national elections
that were held at roughly the same time. In
the municipal elections, less than a fifth of the
candidates used the Internet, where usage was
much more prevalent in mayoral elections in
metropolitan areas, such as greater Athens. As
in other studies, the scholar found very scant
usage of interactive tools.
The studies above found that that incum-
bents, who are already familiar to the public and
benefit from appearance on municipal publica-
tions, billboards, the municipal website, and
other venues – use the Internet to reach potential
voters less than newcomers, who are typically
less known, and hence need more channels to
expose themselves to their constituency (see,
e.g., Herrnson, 2004; Herrnson, Stokes-Brown
& Hindman, 2007; Kamarck, 2003, especially
pp. 87-88).
Hence, the first hypothesis is:
H1: Incumbents running for reelection will
make less use of websites than new can-
didates.
The literature also documents greater
use of websites in more competitive races,
due to the superior incentives and increased
importance of improving voter turnout (see
Herrnson, 2004; Herrnson, Stokes-Brown &
Hindman, 2007; Kamarck, 2003). Hence, the
second hypothesis is:
H2: Candidates in competitive races are more
likely to have campaign websites than
candidates in non-competitive races.
Copyright © 2014, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
20 International Journal of E-Politics, 5(3), 16-32, July-September 2014
Next, the literature demonstrates that can-
didates’ online campaigning correlates with the
size of the constituency as well: as the number
of potential voters per municipality grows,
candidates find it more difficult to reach them
using “conventional” methods, and the appeal
of online campaigning grows. Hence, the third
hypothesis is:
H3: Candidates in larger voting municipalities
are more likely to have websites than can-
didates in smaller municipalities.
As explained above, beyond structural
factors, the socio-economic characteristics of
the constituency can also impact candidates’
website usage. Research has documented a
digital divide in Israel based on age, income,
central (i.e. non-peripheral) geographic location
and education (Mizrahi et al., 2005; Lev-On
& Lissitsa, 2010). Mayoral candidates in dis-
tricts whose constituency is characterized by
low Internet availability and usage may have
fewer incentives to campaign via the Internet,
potentially resulting in less candidate websites
in the given municipality. Hence, the fourth
hypothesis is:
H4: Candidates with a younger, well-off,
educated constituency in a central (non-
peripheral) geographic location in Israel
are more likely to establish a website.
Lastly, the literature demonstrates that
online campaigns tend to be characterized
as conservative, emphasizing “top-down”
information dissemination and providing few
opportunities for interactivity (Margolis &
Resnick, 2000; Gibson & Ward, 2009; see also
Stromer-Galley, 2000). A lack of interactive web
usage has also been observed in other political
contexts in Israel, such as in individual Knesset
Members’ parliamentary sites (Haleva-Amir,
2011), municipal websites (Porian-Lukach,
2011), and political parties’ Knesset campaigns
for the February 2009 general elections (Atmor
& Siani, 2011). Hence, the fifth hypothesis is:
H5: Only a minority of the sites include inter-
active features.
METHODOLOGY
The municipal elections that are held through-
out the country in numerous different districts
enable an examination and analysis of web
campaigning at a higher resolution than that
offered by studying campaigns for the general
Knesset elections, which in Israel are held on a
national basis without delineation into districts.
Moreover, studying municipal campaigns pro-
vides a full picture of the national distribution of
website usage for political purposes, which re-
sults from a large number of “micro-decisions”
made by many candidates – rather than just a
small number of decisions made by the large
parties’ image consultants. These characteris-
tics of municipal elections enable us to discern
differences in the geographic, socio-economic,
and cultural landscape that affect candidates’
decisions to use the Internet.
The current study does not analyze Internet
usage by mayoral candidates in Arab munici-
palities in Israel. According to Kaminski and
Bar-Tal (1996), the relations between Jews and
Arabs in Israel (as in pre-state Palestine) are best
captured by the label “separation,” manifest
(among other aspects) in the near-complete
geographic self-segregation of members of
both groups. This separation into geographi-
cally distinct municipal authorities enables
comparison of Internet usage in political cam-
Copyright © 2014, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
International Journal of E-Politics, 5(3), 16-32, July-September 2014 21
paigns by mayoral candidates with Jewish and
Arab populations.
A separate study (Lev-On, 2013), found
that among the 213 candidates who competed
in the non-Jewish municipalities, only eight
candidates (less than five percent) maintained a
website. The lack of Internet usage in the Arab
sector is an interesting phenomenon that cannot
be accounted for by access differences alone,
and requires a separate and distinct explanation.
Hence, the remainder of the analysis refers to
Internet usage among candidates in Jewish
municipalities only. 2
SELECTION OF WEBSITES
Considering the expansion of Internet cam-
paigning to a variety of platforms (such as
Facebook, YouTube, Twitter, etc.), one could
challenge the decision to focus the study of
online campaigning on websites exclusively.
However, the focus on websites is still justified
in the current research for two main reasons.
First and foremost, Facebook pages and profiles
and other profiles in social media sites were
created for candidates indirectly by supporters,
apparently without their knowledge and con-
sent. Not only it is more difficult to locate such
groups, it is also very problematic to associate
them with a deliberate effort by candidates to
reach potential supporters and impact them.
Second, the study shows that campaign
websites in 2008 still functioned, by and large,
as an informational and organizational hub for
candidates, and users are routed to them from
other platforms in the candidates’ political
websphere (in so far as they exist). Websites
function as a central “post,” in which candidates
have greater control over content and presenta-
tion, interaction with constituencies, and other
issues, than, for example, Facebook pages and
profiles. Hence, profiles on Facebook and other
social media sites were excluded from analysis.
Even after deciding to focus the research
effort exclusively on candidates’ websites, some
methodological concerns remain. For example,
a private entrepreneur created the “2008 Mu-
nicipal Elections Portal” (muni2008.co.il), a
platform that enabled every candidate to create
a web presence dedicated to the elections. The
basic page for each candidate consisted of a
nearly-blank tab with only the candidate’s name
and campaign district. A candidate interested
in a more comprehensive presence on this
platform (beyond the “skeleton page”) could
place additional information for a fee. Since the
candidates who chose to do so were provided
with a fully-functional mini-site, such cases
were treated as equivalent to any other website.
However, the bare “skeleton” pages that were
not further developed were not considered to
be a website and hence excluded.
Also, a variety of political players used the
Internet to promote candidates who seemed to
align with them, but without establishing dedi-
cated websites. Thus, I excluded national party
websites (like the Likud site) that made refer-
ences to municipal candidates or to local and
regional websites that hosted debates between
candidates or featured interviews with them.
DATA COLLECTION:
CANDIDATE WEBSITE SEARCH
The list of candidates was obtained from
the office of the Elections Supervisor at the
Ministry of Interior Affairs, and searches for
candidates’ websites were conducted numerous
times throughout the campaigns from among
the first 500 Google results of the various pos-
sible spellings of candidates’ names. Likewise,
information about candidates’ websites was
Copyright © 2014, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
22 International Journal of E-Politics, 5(3), 16-32, July-September 2014
sought on local forums, where many candidates
post information about themselves, including
links to their websites.
DATA COLLECTION:
INTERVIEWS
While the study centers around a web sphere
analysis of candidates’ presence, it also ap-
plied post-election phone interviews with the
candidates or their representatives, for three
reasons. First, sometimes it may be difficult
to locate candidates’ websites due to lacking
or improper promotion on the popular search
engines, in which cases it is more likely to
obtain this information via interview. Second,
some characteristics of candidates’ web pres-
ence (such as usage of the site for logistical
needs of the campaign) cannot be deduced
from website analysis alone. Third, the study
made references to candidates’ perceptions
of the place of the Internet in their campaign
strategy; again, this information is unobtainable
via study of the website.
The 296 short phone interviews took place
shortly after the elections, when the informa-
tive details were still fresh in the minds of the
candidates – not before the elections, when such
conversations might influence the contenders’
decisions to set up a website or enrich its con-
tent. For districts that had a second round of
elections, interviews were conducted after the
second round. Note that most of the websites
were taken down shortly after the elections. In
anticipation, sites were saved when they were
located, so they could be analyzed later (the last
backup was made on election day).
The semi-structured interviews and web-
sphere analysis were used to study whether or not
candidates had a website, and what usage was
made of the site: static usage (information about
the candidate, his/her vision, and graphics),
multimedia content (audio and video), interac-
tive options (such as forums or chat with the
candidate), and the use of multiple languages.
Candidates and their representatives were also
asked about their Internet usage immediately
prior to the elections in mobilizing activists
and promoting voter turn-out. Furthermore,
they were asked for their views regarding the
potential benefit of Internet use, and if, look-
ing back, they could have made better use of
the Internet considering the circumstances in
which they operated.
DATA COLLECTION:
STRUCTURAL AND SOCIO-
ECONOMIC VARIABLES
In addition to analyzing the features available in
contenders’ websites, data was collected about
variables that may predict the existence of a
site. Thus, data was gathered about each mu-
nicipality’s district, level of peripherality,3 and
number of eligible voters. In terms of electoral
competition, categorized were candidates as
incumbent and newcomers, and for newcomers,
whether they were running against an incumbent
or against a newcomer (as in cases where the
current mayor was not running for reelection).
Additional measures included the number of
candidates and the degree of competitiveness of
the elections, where the percentage of the votes
gained by the winner, and the gap between the
winner and the runner-up, were used as proxies.
Finally, data was also collected regarding the
socio-economic status, median age, education,
and income of the respective constituencies. 4
For logistical reasons, the municipal elec-
tions are broken into three dates, and take place
every five years. The study reports the findings
of the recent municipal elections that were
Copyright © 2014, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
International Journal of E-Politics, 5(3), 16-32, July-September 2014 23
held in November 2008 (for mayors of cities
and local councils), as well as in November
2007 and January 2009 (for mayors of regional
councils). Altogether, elections were held in
143 authorities, involving 489 candidates. In
addition to website search and analysis, 296
of the 489 (60.5%) candidates (or their repre-
sentatives) were interviewed by phone. Only
in very rare cases did interviewed candidates
or representatives say they had a website that
had not been identified in the online research.
Hence, we may assume that the data presented
below would not be considerably different if
the remaining candidates would have been
interviewed as well.
FINDINGS
General
Data shows that 239 of the 489 candidates
(48.9%) had a website, while 250 candidates
(51.1%) did not. Out of the 489 contestants in
Jewish authorities, 123 were incumbents while
366 were not. Among the non-incumbents, 297
ran against incumbents while 69 contested in
open-seat races. There were 4.6 million eligible
voters in the three municipal election cycles
combined.
Among the 123 incumbents running for
reelection, 100 were reelected (81.3%), while
only 23 (18.7%) were unseated by newcomers.
The average percentage of votes the candidates
received in the first round of elections was the
following: incumbents, 54.9% (as stated, over
80% of incumbents were re-elected); newcom-
ers facing incumbents, 18.7%; newcomers fac-
ing other newcomers in open-seat races, 29%.
A significant difference was found between
the percentage of sites among incumbents
(30.9%), and the percentage of sites among
newcomers facing an incumbent (53.2%), or the
percentage of sites among newcomers compet-
ing in open-seat races (62.3%). This finding sup-
ports H1, regarding differential website usage
by new and incumbent candidates (Z=4.89).
However, the percentage of votes gained
(27.8%) by candidates with a website in the first
round of election was lower than the percent-
age received by candidates without a website
(30.6%), although the difference is not signifi-
cant. This is explained by the high percentage
of votes received by incumbents – more than
two thirds of whom lacked a website – versus
the lower percentage of votes received by
newcomers facing them. Thus, the impact of a
website is dwarfed by the “premium” gained
by incumbency in terms of vote gain.
Table 1. Means of the elections’ structural variables, for candidates with and without websites
Structural Variable Candidates with
websites
Candidates without
websites
Number of Eligible voters 54,390** 22,715**
Number of candidates who competed in the election 4.13 3.86
Vote share of winner (in percentages) 51.7* 54.6*
Gap between winner and runner-up (in percentages) 22.6** 27.4**
* p<0.05, **p<0.01
Copyright © 2014, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
24 International Journal of E-Politics, 5(3), 16-32, July-September 2014
Runoff Elections
A second round of elections was held in 18
municipalities. Recall that runoff elections are
held between the two leading candidates in cases
where no candidate crossed the 40% threshold
needed to win the elections in the first round.
Candidates know that the second round
of elections is “crunchtime” and they need to
muster all of their resources to reach the public.
Still, this situation apparently did not motivate
most candidates for greater website usage, with
only 44.4% of second-round contenders having
a website.
Out of the 36 candidates in the second
round of elections, 12 were incumbents (three
of whom had a site), 12 were newcomers facing
incumbents (seven of whom had a website),
and 12 were newcomers in open-seat races, of
which six had a website. In the second round,
candidates without websites received 49.6% of
the total votes, while candidates with websites
received 50.5%. So in this instance as well, the
existence of a website, or lack thereof, had little
to no effect on the outcome of the elections.
The Impact of Structural
Factors on Website Usage
Next, I examined the impact of the number of
eligible voters, the number of candidates, the
gap between winner and runner-up, and the
percentage of votes that the winner received
(the latter two factors being approximations of
the competitiveness of the elections). Graph 1
presents the percentage of websites by number
of candidates competing in a given authority.
Although the correlation between number
of candidates and percentage of websites was
insignificant, the results are still indicative.
Table 2. Socio-demographic variables of the constituent population, for candidates with and
without websites
Socio-Demographic Variable Candidates w ith
websites Candidates without websites
% of residents ages 20-29 in the population
(2006 data) 15.2* 15.7*
% of residents ages 30-44 in the population
(2006 data) 19.8 19.3
Socio-demographic cluster 6.4** 5.8**
Peripheriality level (1) 4.8 4.9
Peripheriality cluster (2) 6.0* 5.5*
Age median 27.6* 25.8*
% of students (ages 20-29) in the population 20.2** 17.5**
% of high-school graduates (ages 17-18) in the
population 59.2** 54.4**
Avg. income, in NIS 3,394** 3,078**
% of population earning minimum wage 41.1** 43.6**
% of population earning double the average wage 11.6** 9.4**
* p<0.05, **p<0.01
Copyright © 2014, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
International Journal of E-Politics, 5(3), 16-32, July-September 2014 25
In general, the more candidates compete in a
municipality, the higher the percentage of web-
sites. Next, Table 1 compares the mean values
of structural variables, across candidates with
and without websites.
T-test results demonstrate significant dif-
ferences between candidates with and without
websites in the cross-sections of number of
eligible voters and the competitiveness of the
race. Thus, candidates who had websites com-
peted in municipalities with more than twice
the number of eligible voters than candidates
without websites and in municipalities where
the races were more competitive (relative to
the municipalities where candidates without
websites competed). These findings provide
some support for H2, regarding the correlation
between the competitiveness of the race (as
measured by the gap between winner and runner-
up), and website usage. H3, which addresses the
correlation between number of eligible voters
and website usage, was also supported.
The Impact of Socio-Economic
Factors on Website Usage
Next, candidates with and without websites
were compared on a cross-section of a number
of socio-demographic variables. As Table 2
demonstrates, a large number of significant
(but weak) differences exist between the two
populations (candidates with and without
websites), in a cross-section of a variety of
socio-demographic variables.
As demonstrated in Table 2, candidates
with websites typically run in municipalities
where the percentage of residents of ages 20-29
is slightly lower than the municipalities where
candidates without websites compete and the
median population age is higher (in contrast
to the relevant hypothesis, H4). However, in
line with H4, candidates with websites – as
Table 3. Features in candidates’ websites
Feature % of sites with the feature
General
Information about the candidate 95.2
Option of sending candidate an email 80.5
Archive 62.9
Languages other than Hebrew 28.1
Multimedia
Graphs/ Photos 79.6
Video 37.8
Audio 21.4
Interactive
Forums 23.8
Polls 16.7
Chat 7.6
Logistical usage (based on interviews)
Activist recruitment 17.6
Copyright © 2014, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
26 International Journal of E-Politics, 5(3), 16-32, July-September 2014
compared to those without – typically compete
in municipalities that are less peripheral, where
the income of the population is slightly higher,
where the percentage of students enrolled in
academic studies and graduating from high
school is higher, the socio-demographic status
slightly higher, the percentage of residents earn-
ing minimum wage is lower, and the percentage
of residents earning double the average wage
is higher.
Website Features
So far, general data was presented about website
usage by candidates, and structural and socio-
economic variables were compared across
districts of candidates with and without web-
sites. Next, the various features that existed on
candidates’ websites are presented in decreasing
order of popularity.
Table 3 shows the prevalence of “standard”
features on candidates’ websites. Background
information on the candidates, their vision and
action plans are standard features on contender’s
websites. In about 80% of the sites, users could
subscribe to mailing lists and send emails, and
about two-thirds of the sites offered access to
archived articles. However, less than a third of
the sites contained information in languages
other than Hebrew. Very few sites contained
polls. Nearly 80% of the websites included
visual materials such as a photo gallery, graphs
and cartoons. Significant use of videos was also
found. However, relatively few sites included
audio content (such as speeches). Interactive
features, such as forums and blogs, were simi-
larly lacking. These findings are consistent with
the hypothesis regarding the expected static and
information-oriented character of candidates’
websites (H5). Sites were scantly used for
communication between activists, fundraising
and activist recruitment. Also, only a minority
of candidates with websites declared that they
updated the websites on a daily basis. Most
candidates updated the site once or twice a
week, or only several times during the entire
campaign.
Finally, when candidates were asked
whether they exhausted what they could get
from the Internet, most website owners replied
affirmatively. Many site owners responded that
the advantages of the Internet were rather limited
in local campaigns, whereas traditional forms of
campaigning – like door-to-door canvassing and
parlor meetings – are more effective. As one of
the candidates said, compared to Knesset elec-
tions, in municipal elections ” it is more impor-
tant to be on the ground.” Another candidate said
that “the target audience wants direct contact
... you cannot reach out through the Internet.”
A third said that the Internet is “nice, modern,
but has no impact on the results.” However, a
significant minority replied with answers like
“the Internet has great promises for us, and
we only touched the tip of the iceberg.” It was
apparent that a large number of the candidates
with websites were aware of the developments
in their online campaign, followed the changes
in their site, and monitored incoming traffic.
Among the candidates who did not use
websites, few regretted their decision. A sig-
nificant percentage of incumbents commented
that they had not found it necessary to use the
Internet, in light of their constituencies’ exist-
ing acquaintance with them. Many candidates
argued that due to the characteristics of the
electorate, establishing a campaign website was
pointless. In some cases, candidates insinuated
that “not everyone has the luxury of dealing
with the Internet.”
Copyright © 2014, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
International Journal of E-Politics, 5(3), 16-32, July-September 2014 27
DISCUSSION AND
CONCLUSION
The study examines the scope and character
of website usage during the municipal election
campaigns of 2007-9, involving a large-scale
analysis of the features available on the sites
and analysis of socio-economic and structural
variables that may influence the presence or
absence of campaign websites. Interviews were
used to confirm and “fine-tune” the findings.
The Internet is increasingly perceived to
be an essential tool for campaigning (see Pew
Internet and American Life Project, 2009). But,
as others have found (Brooks & Peterson, 2008;
Gaziano & Liesen, 2009; Yannas & Lappas,
2004), the near-universal use of websites in
high-visibility campaigns does not spill over to
low-visibility ones. The present study found that
slightly less than half the municipal candidates
establish websites, and that websites were used
more frequently by new candidates as a tool to
gain exposure against incumbents (for similar
results see Herrnson, 2004; Kamarck, 2003).
Significant differences were found between
candidates with and without websites in terms of
structural variables (size of the voting popula-
tion and the competitiveness of the elections),
as well as a variety of socio-demographic
variables, including the level of peripherality,
education and income levels of residents in
the municipality—supporting hypotheses 1-4
of the study.
The municipal races studied typically in-
volve a few thousand voters. Due to the limited
number of voters and the possibility of a small
margin between winners and runner-ups, any
tool capable of reaching voters and influenc-
ing them should be perceived as valuable by
contenders, who are intensely motivated to win
over swing voters and ensure voter turnout.
5Yet despite the significant penetration and
access rates to the Internet in Israel, and the
awareness of harnessing the Internet for politi-
cal campaigns, many candidates – especially
incumbents – perceive online campaigning as
inconsequential, lacking a significant contribu-
tion to the campaign.
Many contenders perceived Internet
campaigning as a minor ingredient in their
campaign strategy, and even as they recognized
they could have been more active online, they
do not perceive the Internet as a differentia-
tor between success and defeat at the polls. It
seems like candidates choose their campaign
strategy according to the context in which they
operate, and utilizing the Internet depends on
variables such as alternative mobilization and
recruitment channels and the size and character
of the constituency. Where candidates do not
perceive the Internet as potentially valuable,
they are less likely to use it in their campaign,
in which cases its interactive potentials do not
materialize either.
Furthermore, even among those candi-
dates who had established a website, data
clearly shows that the sites were, for the most
part, not more than upgraded brochures. Sites
emphasized the candidate, his/her vision and
actions, and generally enabled visitors to send
e-mail and find contact details. Most websites
contained various types of visuals and access
to archived articles. But it was rare to find a
multi-lingual site, and it was apparent that
many websites were infrequently updated, if
at all, during the campaign.
Even a strongly marketed campaign can
fail if people are not motivated to go out and
vote on election day. While the majority of
campaign work for local elections still relies
on “footwork,” the Internet can support the
organizational effort required by recruiting
Copyright © 2014, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
28 International Journal of E-Politics, 5(3), 16-32, July-September 2014
extra “feet” for projects such as canvassing
campaigns. But at the municipal level at least,
these capabilities are still not being utilized.
Interaction between candidates and website
visitors was quite rare as well. The interactive
online campaigns that gained some mainstream
media coverage, such as those of the leading
candidates in Israel’s largest cities of Tel Aviv-
Jaffa and Jerusalem (surveyed above) were the
exception, not the rule. Interactive features such
as forums and chat rooms gained very limited
popularity.
What are the reasons for the lack of user
interaction in candidates’ website? Beyond the
lack of awareness among many candidates to
the abilities to decentralize the campaign and
involve the public online, it seems that quite
a few candidates interviewed in this study
expressed concern regarding interaction with
the public (see, in this regard, Stromer-Galley,
2000). One of the candidates said that he es-
tablished a forum but after operating it only
three weeks decided to take it down because it
“became bogged down by cursing and vandal-
ism by the competitor’s activists, instead of a
tool for reaching out to new voters.” It seems
that fear of hostile user-generated content, the
possible effects it may have on visitors and the
need for constant monitoring are factors that
may discourage candidates from greater use
of online social media.
These conclusions are consistent with pre-
vious research findings that highlight the lack
of interactive web usage in a variety of political
contexts in Israel, such as in Knesset Members’
parliamentary sites (Haleva-Amir, 2011) and
in municipal websites (Porian-Lukach, 2011).
Similar patterns were found in the national
election campaigns in Israel, including the
round of elections in February 2009 that took
place only a month after the last cycle of the
municipal elections discussed above (Atmor
& Siani, 2011). Note also that the majority
of websites were taken down shortly after the
elections, which illustrates their functioning as
a goal-oriented tool, not as a means to generate
log-lasting conversation with constituencies.
The findings shed light on the possibilities
of interactive political usage of the Internet and
comprehensive public engagement online. It
is during election campaigns that candidates
have clear incentives – more so than after the
elections – not only to pass on their message to
the public, but also to hear the public’s message
and possibly adjust their platform accordingly,
thus yielding a better fit between the public
agenda and the agenda of the campaign (see
Nadeau et al., 2008). The fact that this process
does not occur even when technology allows
for it demonstrates the imperfect realization
of participatory and bi-directional democracy
through advanced technologies.
Another obstacle towards realizing the vi-
sion of online democracy is the digital divide,
which is manifest in the negligible percentage
of websites among Arab non-Jewish candidates,
compared with almost fifty percent usage in the
Jewish sector. The usage gap between Jewish
and Arab candidates certainly demonstrates that
online interactive exchange between representa-
tives and constituencies might not be practiced
simultaneously among all population segments
(see Lev-On, 2013).
But the prospects of online and interactive
web campaigning do not seem glum. First,
the interest generated by early adopters that
used the Internet in a more participatory and
bi-directional manner – notably the leading
candidates in Tel Aviv-Jaffa and Jerusalem, as
discussed above – may pave the way for others
who acknowledge the potential benefits of the
Copyright © 2014, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
International Journal of E-Politics, 5(3), 16-32, July-September 2014 29
Internet in campaigning, and possibly even after
the campaigns conclude.
Second, one must not confuse the lack of
Internet usage by candidates with lack of interest
by the public in online political information. A
nice example comes from the ultra-Orthodox
public, one of the sectors with the lowest rates
of Internet connectivity in Israel. The municipal
elections, and especially the Jerusalem mayoral
elections, raised a lot of interest among the ultra-
Orthodox public, and in the day following the
elections when the unofficial results were first
published, record-high amounts of traffic and
posts were recorded in Behadrei Haredim – the
leading news website for the ultra-Orthodox
public. In a single day, 1.5 million page-views
were recorded, as well as a 30% increase in the
number of unique visitors compared to the daily
average. Moreover, 50% more traffic than the
daily average was recorded on the sites’ forums
(Global Networks, 2008).
Apparently, Internet campaigning is cer-
tainly picking up, as does the recognition of
the benefits that can be derived from it under
certain conditions. The Israeli public certainly
expresses an interest in obtaining relevant po-
litical information online. Further research is
necessary to determine the circumstances under
which the “supply” by candidates would meet
the “demand” by constituencies and incentiv-
ize the public not only to listen, but also to
participate and to be heard.
REFERENCES
Atmor, N., & Siani, A. (2011). Party websites in the
2009 elections: A comparative view. In Lev-On, A.
& Cohen, E. (Eds.), Connected: Politics, Technology
and Society in Israel, pp. 177-210. Tel-Aviv: The
Israeli Political Science Association Press. (Hebrew).
Brooks, S., & Peterson, J. (2008). All politics is lo-
cal: Campaign communication in urban elections.
Paper presented at the annual meeting of the Midwest
Political Science Association, Chicago.
Cohen, M. (2009). 250,000 New Internet users in
2008. Haaretz, Jan 20th. http://www.haaretz.com/
hasite/spages/1055634.html (Hebrew).
Fernandes, J., Giurcanu, M., Bowers, K. W., & Neely,
J. C. (2010). The writing on the wall: A content
analysis of college students’ Facebook groups for
the 2008 presidential election. Mass Communica-
tion & Society, 13(5), 653–675. doi:10.1080/1520
5436.2010.516865
Foot, K. A., & Schneider, S. M. (2006). Web Cam-
paigning. Cambridge, MA: MIT Press.
Gaziano, J., & Liesen, L. (2009). Use of campaign
websites by candidates running for state and local
office. Paper presented at the annual meeting of the
Midwest Political Science Association, Chicago.
Gibson, R. K. (2004). Web campaigning from a global
perspective. Asia-Pacific Review, 11(1), 95–126. do
i:10.1080/13439000410001687779
Gibson, R. K., & Römmele, A. (2005). ‘Down
periscope’: The search for high-tech campaigning at
the local level in the 2002 German federal election.
Journal of E-Government, 2(3), 85–109. doi:10.1300/
J399v02n03_05
Gibson, R. K., & Ward, S. (2009). Parties in the
digital age: A review article. Representation, 45(1),
87–100. doi:10.1080/00344890802710888
Gibson, R. K., & Ward, S. J. (2000). A proposed
methodology for studying the function and ef-
fectiveness of party and candidate web sites.
Social Science Computer Review, 18(3), 301–319.
doi:10.1177/089443930001800306
Copyright © 2014, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
30 International Journal of E-Politics, 5(3), 16-32, July-September 2014
Haleva-Amir, S. (2011). Present-absentees: MK’s
usage of personal Internet tools. In Lev-On, A., &
Cohen, E. (Eds.), Connected: Politics, Technology
and Society in Israel, pp. 211-261. Tel-Aviv: The
Israeli Political Science Association Press. (Hebrew).
Herrnson, P. S. (2004). Congressional Elections:
Campaigning at Home and in Washington. Wash-
ington, DC: CQ Press.
Herrnson, P. S., Stokes-Brown, A. K., & Hindman,
M. (2007). Campaign politics and the digital divide:
Constituency characteristics, strategic consider-
ations, and candidate Internet use in state legislative
elections. Political Research Quarterly, 60(1), 31–42.
doi:10.1177/1065912906298527
Howard, P. N. (2006). New Media Campaigns
and the Managed Citizen. Cambridge: Cambridge
University Press.
Kamarck, E. C. (2003). Political campaigning on
the Internet: Business as usual? In E. C. Kamarck,
& J. S. Nye (Eds.), Governance.com? Democracy
in the Information Age (pp. 81–103). Washington,
DC: Brookings Institution Press.
Kaminski, M., & Bar-Tal, D. (1996). Stereotypical
perceptions of various labels of Israeli-Arabs as
functions of age and religiosity. [Hebrew]. Iyunim
Bechinuch, 1-2, 121–157.
Lev-On, A. (2011). Campaigning online: Use of the
Internet by parties, candidates and voters in national
and local election campaigns in Israel. Policy and
Internet, 3(1), article 6. Online: http://www.psocom-
mons.org/policyandinternet/vol3/iss1/art6
Lev-On, A. (2013). Another flew over the digital
divide: Internet usage in the Arab-Palestinian sector
in Israel during the municipal election campaigns,
2008. Israel Affairs, 19(1), 154–169. doi:10.1080/
13537121.2013.748292
Lev-On, A., & Lissitsa, S. (2010). Digital divide,
Israel 2008. Proceedings of MCIS 2010, Paper 54.
http://aisel.aisnet.org/mcis2010/54
Madzini, R. (2008). Mayoral candidates in Jerusalem
storm the net. Ynet, November 9th. http://www.ynet.
co.il/articles/0,7340,L-3620037,00.html. (Hebrew).
Margolis, M., & Resnick, D. (2000). Politics as
Usual: The cyberspace Revolution. Thousand Oaks,
CA: Sage.
Mizrahi, Y., Bar, N., Hezronov, I., & Oron, N.
(2005). Digital Readiness Survey and Digital Gaps
in Israel. Jerusalem: Ministry of Finance. http://
www.maor.gov.il/Maor/Pages/HE/DigitalSurvey.
aspx?P=MyComm2007. (Hebrew).
Mor, G. (2008). Local network: The munici-
pal elections in Tel-Aviv are bubbling on-
line. Ynet, November 7th. http://www.ynet.co.il/
articles/0,7340,L-3619129,00.html. (Hebrew).
Nadeau, R., Nevitte, N., Gidengil, E., & Blais,
A. (2008). Election campaigns as information
campaigns: Who learns what and does it mat-
ter? Political Communication, 25(3), 229–248.
doi:10.1080/10584600802197269
Networks, G. (2008). http://www.prweb.co.il/view.
php?pressId=T1Rneg==
Pew Internet and American Life Project. (2009).
The Internet’s Role in Campaign 2008. http://www.
pewInternet.org/Reports/2009/6--The-Internets-
Role-in-Campaign-2008.aspx?r=1
Purian-Lukach, R. (2011). Municipal e-government:
Comparative index, Success factors and models
for management and evaluation. In Lev-On, A., &
Cohen, E. (Eds.), Connected: Politics, Technology
and Society in Israel, pp. 137-173. Tel-Aviv: The
Israeli Political Science Association Press. (Hebrew).
Copyright © 2014, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
International Journal of E-Politics, 5(3), 16-32, July-September 2014 31
Stallings-Carpenter, C. A. (2010). The Obam-
achine: Techno-politics 2.0. Journal of Informa-
tion Technology & Politics, 7(2-3), 216–225.
doi:10.1080/19331681003765887
Strachan, J. C. (2003). High-Tech Grass Roots: The
professionalization of Local Elections. Lanham:
Rowman & Littlefield Publishers.
Stromer-Galley, J. (2000). Online interaction and
why candidates avoid It. Journal of Communica-
tion, 50(4), 111–132. doi:10.1111/j.1460-2466.2000.
tb02865.x
Today, I. (2008). November 11th, p. 7 (unsigned).
ENDNOTES
1 Many thanks to Chen Sabag, Mark Steinitz,
Odelia Adler and Sarit Mazouz for their as-
sistance. Special thanks to Eliya David for
dedicatedly supervising the project.
2 Note that three Arab candidates competed in
three Jewish-majority municipalities with an
Arab minority population: Haifa, Acre, and
Tel Aviv-Jaffa. Furthermore, a small number
of Jewish regional councils contain some
Arab villages, which in all cases comprise
minorities in their respective municipalities.
One Arab candidate competed in the election
for mayor of the Gilboa Regional Council.
These Arab mayoral candidates competing
in Jewish-majority districts were included in
the present study.
3 The Peripherality Index is constructed by the
Israeli Central Bureau of Statistics (CBS)
and takes into account the distance of the
municipality from Tel-Aviv district, which is
the economic and business center of Israel; and
the potential accessibility index – a function
of the distance between the municipality and
other municipalities, as well as the size of
population – which functions as a proxy to a
variety of economic parameters. See http://
www.cbs.gov.il/hodaot2008n/24_08_160b.
pdf.
4 Data about socio-economic status, median
age, education (percentage of students among
residents in the age group 20-29, percentage
of high school graduates among residents of
ages 17-18) and income (average income,
percentage of residents earning minimum
wage, percentage of residents earning twice
or more than the average wage) are based on
Israeli Central Bureau of Statistics (CBS) data
published in March 2008 on the website of the
Ministry of Interior Affairs, http://www.moin.
gov.il/Apps/PubWebSite/publications.nsf/Al
l/5FCD19CFBECFBC18C2257427002FF5
DC/$FILE/Publications.pdf?OpenElement
As well as http://www.moin.gov.il/Apps/
PubWebSite/publications.nsf/All/795AF
BDE9DEB7E79C22574270030E5D8/$F
ILE/Publications.pdf?OpenElement . The
district and the percentages of residents ages
20-29, 30-44 in the population were taken
from CBS website, www1.cbs.gov.il/pub-
lications/local_authorities2006/excel/p_li-
bud.xls. Data about peripherality are also
taken from the CBS website, www.cbs.gov.il/
hodaot2008n/24_08_160b.pdf . The numbers
of eligible voters were taken from messages
posted by the Elections Supervisor in the
Israeli Ministry of Interior Affairs. Details
of incumbents were downloaded from the
website of the Ministry of Interior Affairs and
from the site of the Israel Union of Local Au-
thorities, www.iula.org.il. Unfortunately, data
are unavailable on other potentially relevant
parameters, such as Internet connectivity rates
Copyright © 2014, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
32 International Journal of E-Politics, 5(3), 16-32, July-September 2014
and population religiosity in different locales
across the country.
5 The four narrowest margins between winners
and runner-ups were 10, 33, 37 and 43 votes.
In the city of Metula, where a second round of
elections took place, a single vote separated
between the candidate who came up second in
the first round (and ultimately won the second
round) and the candidate who finished third
in the initial elections.
Azi Lev-On is the chair of the school of communication in Ariel University in Israel. His studies explore
behaviors and collective action in computer-mediated environments, employing a variety of methods such
as content and link analysis, surveys and laboratory experiments. Recent research analyzes interactions
in communities of practice of the Israeli Ministry of Social Services, in municipal forums, and by Ultra-
Orthodox women who participate in closed forums online.
