1

Effects of Gender and Social Status on

How Filipinos Perceive Political Candidates

MA. ELIZABETH J. MACAPAGAL

Ateneo de Manila University

The study investigated the influences of respondent's gender

and social class, and politician's gender on perceived traits of
political candidates. The survey used a 2 X 3 X 2 factorial design
covering 536 respondents aged 18 to 35 years old. The respondents
rated a hypothetical political candidate who was either male or
female using a Shah scale. Research results show that respondents
perceived political candidates positively. The females and
respondents from the lower social classes showed a more favorable
perception of politicians. Respondents believed that female
politicians are more attractive, emotional, intelligent, and religious
but male politicians are more corrupt. Findings also show that
identity politics seems to be present wherein females tend to
view female politicians more positively. Moreover, the males from
the lower classes tend to prefer male politicians. The implications
of the findings for construal research, voter education,
consciousness-raising, and political campaign strategies are
discussed.

Women comprise half of the world's population but ironically,

they continue to lag behind in positions of decision-making and
political affairs in every country. Women make up only 15% of
national parliaments worldwide as of April 2005. In Asia, 15.2%
of the members of the Lower House and 13.5% of members of the
Upper House are women ("Women in National Parliaments," 2005).

Power and decision-making in the Philippines continue to be

male-dominated. In the Philippine Senate, the highest ratios of .

PHILIPPINE JOURNAL OF PSYCHOLOGY (2006), Vol 39 No 2, pp. 1-30.

2

elected female senators since 1946 has only been 4 out of a total
24 senators in 1992, and 3 out of 12 senators in 1995. Based on
the Commission on Elections (COMELEC) data, the average
participation rate of women in the House of Representative and
local government is roughly 10%. Why is there such an imbalance
of power? Explanations for this sparse representation of women
in leadership roles traditionally focused on the idea that there is
a lack of qualified women or that women display fewer traits and
I

motivations necessary to attain and achieve success in political

I
office (Eagly & Karau, 2002). However, it i, is important to look into
other possible explanations. Could it bJ that women politicians
are still perceived differently from male politicians? Does bias
against women candidates exist? Th is research examines
perceptions of political candidates in terms of character traits in
relation to the gender of both perceiver and candidate. Moreover,
the socioeconomic status of the respondents .is also considered
because public opinion surveys in the 'Philippines report that
attitudes toward politicians and political issues vary according to
social class (Mangahas & Guerrero, 1998).
\

The following questions are addressed in this research:

(a) Does respondent's gender influence perceived traits of political
candidates? (b) Does respondent's social class influence perceived
traits of political candidates? (c) Does politician's gender influence
perceived traits of political candidates? The: study also investigates
possible two-way and three-way interaction effects among the
variables respondent-gender, respondent-social class, and
politician-gender on perceived traits of political candidates.

Gender Differences in Political Attitudes and Behavior

Past researches show that males and females differ in political

information and interest (Bourque & Grossholtz , 1998; Fife-Schaw
& Breakwell, 1990; Kolinsky, 1993). Results of these Western
studies reveal that males consistently know more about politics
whereas females express less interest in the field. Aside from
 3

variations in political interest and information, men and women

were also found to differ in political participation such as voting
and running for public office (Appleton & Mazur, 1993; Bennett
& Bennett, 1993; Pharr, 1981; Tancangco, 1992).

In the Philippines, a study of political values of Filipino women

by Tapales (1992) reported that Filipino women are politically
aware and knowledgeable but refrain from participating in political
discussions, do not have direct contact with government leaders,
and are not members of political organizations. Moreover, their
major political activity is voting in elections and attending rallies.
They regard major political activity as basically for men, preferring
to engage in economic activities.

In sum, the results of previous research on political attitudes

and political participation have shown that women continue to
be less interested and involved, compared to men. The
socialization process continues to encourage political passivity
among women and promotes the idea that politics is more suited
for males. Thus, in this study, I posit that men and women
respondents have varied perceptions of political candidates' traits.
Similarly, men and women have varied attitudes toward male
and female politicians.

Social Class Differences in Elections

Social class is the distinction of wealth and status. In the

Philippines, the existence of social classes is very evident. Public
opinion surveys consider socioeconomic status as a variable to
determine differences in how the different social classes think
(Mangahas & Guerrero, 1998). Surveys classify households into
the following socioeconomic classes according to their type of
dwelling: the rich and the middle Class ABC, the poor Class 0,
and the poorest Class E (Arroyo, 1990). Based on this scheme,
majority of respondents belong to Class 0, roughly 65%, the ABCs,
who amount to about 20%, and the Es, about 15% (Mangahas
& Guerrero, 1998).
4

In politics, the D and E classes, the masses and the very poor
respectively, were the ones responsible 'for Joseph Estrada's victory
in the 1998 presidential election (Marwahas, 1998). It was social
class, rather than region, or religion, or gender, or age that was
critical to that election, because Estrada lost among the ABC
votes-the votes of the rich, the well tb do, and the middle class
(Mangahas, 1998). In the 2004 presidential election, the SWS
survey showed that Gloria Macapagal-Arroyo had a commanding
I
lead in classes ABC versus Fernando Ppe Jr. (FPJ) but FPJ failed
to pull away in classes D and E [Olivares-Cunanan, 2004).

The perceiver's social class likewiselinteracts with politician's

gender. Whether people would accept women as politicians was
found to depend on the social class of the perceiver. For example,
respondents from all over the Philippines were asked to agree or
disagree with the statement" A woman is as capable as a man
in performing the job of President of the 'iPhilippines." The highest
net agreement with this statement came from Class E.
The lower the socioeconomic status, the more the agreement
with the statement (Mangahas, 1998). I

What accounts for the reported variations in the way the

different social classes think? According to cognitive psychologists,
this could be due to the fact that perceptions are dependent on
the subjective meaning that the actor attaches to a situation
(Ross & Nisbett, 1991). What matters for behavior is often not the
objective situation but the subjective :interpretation of this
situation. Moreover, some research efforts have emphasized that
cognition is always social and context dependent (Forgas, 1981;
Moscovici, 1981; Nye & Brown, 1996). Co~nition is not an intra-
individual process and cannot be reduced to individual cognition;
it involves processes that are uniquely social in character (Forgas,
1981). Our social representations are a :product of a: delicate
interplay between information proces'sing strategies and
sociocultural processes (Forgas, 1981; Moscovici, 1981). Just as
cognition cannot be properly understood without placing it into a
social context, society and culture must be studied as the product
 5

of cognitive efforts of individuals. Social cognition, therefore, must

synthesize the individual and collective spheres in explaining
everyday knowledge.

Krauss (as cited in Forgas, 1981) argued that in social

cognition, the crucial problem is not how to process the
information once given, but how to decide what the information
is in an immensely complex, confusing, and subtle social
environment. The way we categorize, represent, and react to
social information is often unpredictable and impossible to
comprehend without a careful analysis of the surrounding culture.
I hypothesize that perceptions of political candidates' traits
vary according to social class because perceptions are a subset
of social cognition (Oskamp, 1981) and social class is a contextual
variable.

Voters' Evaluation of Political Candidates' Traits

How do voters evaluate candidates? Studies show that voters

use several criteria in determining who to vote for. Pierce (1993)
studied how voters evaluate candidates for president in the US.
Four dimensions were found to be significant, namely, leadership
(commands respect, inspiring, strong leadership), competence
(hard working, intelligent, knowledgeable), integrity (decent,
moral, good example), and empathy (compassionate, kind, cares
about people, sincere).

Eagly, Diekman, Schneider, and Kulesa (2003) considered

three facets of perceived candidate characteristics: political
effectiveness (strong leader, intelligent, respected), agentic
personality (competitive, dominant, aggressive, daring, and
communal personality (sympathetic, sensitive, warm, kindj.:

In the Philippines, Montiel (1986), and Sy (2001) studied the

preferred characteristics in politicians. Montiel's 1986 study of
nontraditional political leadership showed that Filipinos prefer
the following characteristics: democratic, kept promises, held
6

consistently political views, was for the masses, did not lie or
cheat, and was not sexist. Sy (2001) studied political candidate
evaluation in the Philippines. She found that urban poor voters
use several candidate-related variables such as background,
sociopolitical status, promises, popularity, and behavior in
evaluating political candidates. Voters also consider whether the
candidates are godly (maka Diyos) , relates well with people (maka-
tao), is capable (may kakayahan), has principle (may paninindigan) ,
for the country (maka-bayan), keeps one's word (may isang salita),
for the poor (maka-mahirap), approachable (madaling lapitan), and
does not hide true self (di pakitang tao).

In the studies of Pierce (1993), Montiel (1986), Sy (2001), and

Eagly et al. (2003), traits were used to evaluate candidates.
Moreover, gender was not a variable or did not appear as one of
the criteria in evaluating political candidates. In this present
study, perceived traits as well as gender of both respondent and
politician are considered.

Attitude Toward Women and Men Political Candidates

Previous studies have shown that men and women political

candidates are perceived as having different traits (Duke, 1993;
Kaid as cited in Fox & Smith, 1998; Leeper, 1991; Rosenwasser
& Dean, 1989). In 1991, Leeper found that voters infer in women
candidates stereotypical female traits. Voters believe women are
more honest. However, another study in the US conducted by
Kaid (as cited in Fox & Smith, 1998) reported contradictory
findings. Students viewed women candidates as more sophisti-
cated, honest, attractive, aggressive, strong, and active. Duke
(1993) reported that in one study, women candidates were
rated more intelligent and concerned about people, but men
candidates were still perceived as more knowledgeable about
politics and considered strong. Males, regardless of gender role,
were perceived as being more likely to win a presidential elec-
tion.
 7

According to a study conducted by Aguilar in 1990, Filipino

women politicians were perceived as weak in facing problems.
Other perceptions include the following: They cannot sustain
themselves in crisis; their freedom of movement is limited even
in going around and knowing the constituency; they cannot tra,vel
without a companion; they may be soft in making decisions that
will require toughness; they may become pregnant and this may
hamper their performance of political duties. The respondents of
the study who were career women politicians argued that the
advantages of having women politicians include: Women have
more patience and perseverance; they are more sincere; the
women voters may tend to support female politician; constituents'
request will seldom be refused; there will be more fairness in
government and politics; women are more sympathetic and
considerate. They think the best woman politician model must
be one who is competent, tough, dynamic, and energetic, has
well-rounded knowledge, and is strong-willed, decisive, good, and
strong.

Studies also show that a bias by voters against women

candidates, especially in the US still exists (Fox & Smith, 1998).
Fox and Smith (1998) conducted a classroom experiment to explore
how voters evaluate candidate sex. They compared Wyoming and
California college students on their voting preferences for
hypothetical elections with male and female candidates. The
Wyoming sample significantly chose the male candidate, ana
thus exhibited bias against women candidates. However, the more
liberal California students did not exhibit any gender bias.

In the 1970s, college students evaluated hypothetical

candidates roughly equally, but both young men and young women
assumed that the male candidate would win office and that the
women would lose (Sapiro as cited in Rinehart, 1992). Even more
disturbing was Bowman's finding (as cited in Rinehart, 1992)
that a physically attractive woman fared even worse among
student voters in simulated elections-although handsomeness.
worked distinctly to the advantage of the male candidates.
8

Because of their deeply-rooted biases, male and female voters

in the Philippines have a prejudicial outlook against female
candidates such that between a man and a woman candidate
with equal qualifications, they exhibit an irrational preference
for the former (Romero as cited in Tangcangco, 1992). Moreover,
the man of the house can dictate to the women voters whom to
vote for. However, this does not mean that women voters do not
consider women candidates. They still do but they vote for a
woman candidate not because they are of the same gender but
because the women candidates are capable, qualified, and
deserving (Tangcangco, 1992).

In sum, these studies conducted in both developed and

developing nations report that there is indeed a difference in the
way male and female politicians are perceived. In many countries,
a promale bias continues to exist among voters. However, attitude
towards female candidates in terms of character traits have also
been positive, although still following stereotypical perceptions
such as being moral and honest. Thus, my third hypothesis is
respondents have varied perceptions of the traits of women and
men political candidates.

Identity Politics
Would female voters tend to view female politicians more
positively? In other. words, would an interaction exist between
gender of voter and gender of politician? Previous studies suggest
that an interaction does exist (Lewis & Bierly, 1990; Plutzer
& Zipp, 1996). Female university students followed a profemale
sentiment in evaluating female politicians' competence (Lewis
& Bierly, 1990). Plutzer and Zipp (1996) found evidence that sex
of the voter is significantly related to voting for female candidates
in 8 of the 13 states they studied. They suggested that gender
identity exists, that is, female voters identified with female
candidates, especially if they espoused feminist values.
 9

I
Identity politics refers to the formation of political allegiances
on the basis of some demographic similarity, such as sex, race,
and religion (Plutzer & Zipp, 1996). An example would be in 1960
in the US when John F. Kennedy, a Catholic, won the support of
approximately 80% of Catholic voters but he lost enough of the
much larger Protestant vote to produce a net loss nearly coating
him the presidency.

In the 1992 elections in the US, a then-record 14 women ran

for governor or US senator, and many made their gender a salient
aspect of their candidacies. According to Plu tzer and Zipp (1996),
given the increasing numbers of women running for office, and
the higher turnout rates of women voters in the US, the existence
of identity politics represents a threat to the current party system
and may playa pivotal role in deciding electoral outcomes. In
this study, I posit that an interaction exists between gender of
voter and gender of politician because of identity politics.

METHOD

The study employed a precoded survey for data gathering and

analysis. The survey employed a 2 X 3 X 2 factorial experimental
design resulting in 12 groups or cells. The gender (male or female]
and socioeconomic status (ABC, D, E) of the respondent were the
subject variables. The variable manipulated was the gender of
the politician (whether male or female). The dependent variable
was the perceived character traits of the political candidate as
measured by a Shah scale.

Participants

The survey covered 536 respondents aged 18 to 35 years old,

which comprise the largest voting age bloc (44%) according to the
Social Weather Stations. There were 253 males and 283 females.
Their mean age is 24.44 years old. The participants were divided
into three groups according to socioeconomic status, namely,
10

ABC, D, and E. The ABC respondents have a household monthly

income of Php 15,000 and above, the D respondents Php 7,500
to Php 14,999 whereas the E participants' families earn below
Php 7,500 a month. There were 189 respondents from the ABC
Class, 161 from Class D, and 186 from Class E.

Quota sampling was used to choose at least 40 respondents

per group using convenience and purposive techniques. The
participants were randomly assigned to receive either male or
female politician.

MEASURES

Shah Scale. The main instrument used is a scale patterned

after Shah's instrument (1999), which is attitude scale based
upon trait adjectives. The first step was to explore different trait
adjectives relevant to the attitude object, in this case, politician.
In the next step, the evaluative values of the trait adjectives
were determined for the target population. Finally, the mean
likeability scores and standard deviations of the trait adjectives
computed were used to operationalize numerical scale positions,
similar to Thurstone's method.

Construction of the Shah scale. The sources of the adjectives

for the attitude scale included the literature and results of an
open-ended survey on 40 Psychology majors from the Ateneo de
Manila University. The students were asked to think of a
hypothetical male or female politician in the Philippines and to
list negative, positive, and neutral traits associated with him or
her. I was able to generate a list of 66 trait adjectives from the
literature and surveys, which I administered to 54 respondents
of varied socio-economic classes, with a mean age of 29 years
old. They were instructed to think of a hypothetical politician
described by each trait adjective and to rate the adjective
according to how much they would like the politician if she or he
had that trait. They were asked to use a 7-point rating scale:
 11

-3 (very much disliked), - 2 (dislikedj, -1 (dislike a little), 0 (neither

liked nor dislikedi, +1 (liked a little), +2 (liked), and +3 (very milch
liked).

The mean likeability and standard deviation scores of each

trait were computed. Adjectives that had no similar mean
likeability scores were included in the final list. In cases were
there was more than one adjective per scale position, I selected
the trait with the lowest standard deviation. In cases where the
adjectives had similar standard deviations, I chose the trait that
was simpler to understand based on the comments of the pretest
sample or was more relevant in describing politicians in the
Philippines. The following 18 traits were chosen (see Table 1).
The sum of the mean likeability scores for the negative traits is
-16.57 whereas the sum for the positive traits is 16.53 with a
difference of .04.

Table 1. Traits in the Final Shah Scale With Their Corresponding Mean Likeability Scores

--+----
Negative traits Score Positive traits Score
+---~_._--

Emotional Mafamdamin) - .04 Rich (Mayaman) .39

Loud (Maingay) - 1.11 Popular (Sikat) .81
Moody (Sumpungin) • 1.54 Attractive (Maganda 0 Guapo) .98
Impatient (Wa/ang pasensya) • 1.75 Religious (Maka-Diyos) 1.81
Fickle-minded (Pabago-bagong
isip) - 2.09 For the poor (Maka-masa) 1.93
Indifferent (Wa/ang pakia/am) - 2.19 Moral (Mabutl) 2.43
Power-hungry (Sakim sa
kapangyarihan) - 2.35 Intelligent (Matalino) 2.57
Turncoat (Balimbing) - 2.61 Competent (May kakayahan) 2.74
Corrupt (Kufakot) - 2.89 Honest (Tapat) 2.87
TOTAL - 16.57 TOTAL 16.53
---+------~

Scoring. To manipulate gender of politician, half of the

respondents were asked to rate a male politician, Mario del
Rosario and the other half were asked their perceptions of a
female politician, Marie del Rosario. The names of the political
candidates, Mario and Marie del Rosario, were fictitious to control
12

for possible biases linked with actual or real personalities in the

political scene.

The respondents were given the list of scaled adjectives arranged

in a alphabetical order. They were instructed to check those adjectives
that in their opinion could be used to describe the qualities of the'
target person. The mean likeability scores of the adjectives that the
respondent checked were added. For example, a respondent
checked emotional (-.04), religious (1.81), and intelligent (2.57). The
respondent's score is +4.34. The highest possible score is +16.53
which means the respondent checked all the positive traits associated
with the target politician. On the other hand, if a respondent checked
all the negative traits, his or her score would be --:16.57.

Reliability and validity. The construction of the Shah

attitude scale and its use to measure the attitudes of people has
been validated in several studies, including a study with
politicians as the target attitude object in Pakistan (Shah, 1999).

Shah established the validity of his scale by correlating the

results with two other scales measuring attitudes of people
towards politicians. An r of .57 and .73 were obtained.

The reliability of the present scale was determined by

split half and the measure of internal consistency. The
Spearman-Brown reliability coefficient was .72 which is
significant at p =.00. The coefficient alpha was .82.

RESULTS

Overall Perceived Traits of Politicians

The data show that in general, respondents perceived political

candidates favorably. Majority 'of the respondents think political
candidates are competent, intelligent, religious, for the poor,
moral, and honest.
 13

The descriptive statistics also revealed overall po srtrve

perceptions of political candidates, in terms of character traits. '
When all the mean likeability scores were added per respondent,
an overall mean of 7.04 and standard deviation of 7.81 were ,
obtained. With a highest possible positive score of 16.53, this
result indicates a positive perception of political candidates, in
general. Majority of the respondents (80.4%) obtained a positive
score indicating overall favorable perceptions of traits of political
candidates.

As revealed in Table 2, the three-way ANOVA for perceived

traits produced four significant results with two main effects,
namely, gender of respondent and SES of respondent, and two
interaction effects-gender of politician X gender of respondent
and gender of politician X SES of respondent X gender of
respondent.

Table 2. Analysis of Variance of Perceived Traits

Source df F p

Gender of respondent 1 6.14·· .01

SES of respondent 2 10.79·· .00
Gender of politician 1 1.19 .27
Gender of politician X SES of respondent 2 1.43 .24
Position of politician X SES of respondent 2 1.43 .24
Politician's gender X gender of respondent 1 5.56· .02
SES of respondent X gender of respondent 2 0.47 .63
Politician's gender X SES X gender of res. 2 3.86· .02

Note ." p < .05.•• p < .01.

Further analysis of the data on perceived traits revealed

significant differences on particular traits as well. Aside from
conducting three-way ANOVA for overall perceived traits, a chi
square test was conducted per trait to determine significant
differences of each trait on the three independent variables.
These differences are discussed under each independent variable.
14

Respondent's Gender

The findings suggest that respondent-gender differences exist

in perceived traits of political candidates. The male respondents
had a mean rating of 6.08 whereas the females gave significantly
higher ratings to politicians with a mean of 7.73 (p = .01). On
specific traits, chi square analysis revealed significant gender
differences, namely, competence (p = .00), emotional (p = .00),
fickle-minded (p = .03), impatient, (p = .03), indifference (p = .02),
intelligent (p = .04), and loud (p = .00).

The male respondents significantly rated politicians as more

emotional, fickle-minded, impatient, indifferent, intelligent, and
loud whereas the female voters gave more ratings of competence
to the politicians.

Social Status of Respondent

Respondents from the different socioeconomic status varied

significantly in their perceptions of traits of politicians (p = .00).
The E class gave the highest ratings (M = 8.98), followed by the
D class (M = 6.22), with the ABC Class giving the lowest ratings
(M = 5.52).

The. following traits obtained significant chi square results

among the three SES groups: attractive (p = .00), competent
(p = .03), corrupt (p = .00), intelligent (p = .04), moral (p = .00),
popular (p = .02), power-hungry (p = .00), religious (p = .00), and
rich (p = .01).
Compared to the D and E voters, the ABC respondents
perceived politicians as more corrupt, popular, power-hungry, and
rich. The E voters perceived politicians as more attractive,
competent, for the poo~, honest, moral, and religious. Both ABC
and E respondents, on the other hand, concur that politicians
are intelligent but the D subjects think less so.
 1S

Gender of Politician
Contrary to expectations, there was no significant difference
in the overall perceived traits of male and female politicians
(p = .29) although the mean scores revealed that female
politicians were perceived slightly more positively (M = 7.26)
compared to male politicians (M = 6.55). The chi square analyses
revealed that certain traits are associated more with female
than male politicians. For example, female politicians are
perceived as more attractive (p = .04), emotional (p = .00),
intelligent (p = .00), and religious (p = .02) but male politicians
are perceived as more corrupt (p = .05).

Interaction Effects
Based on the three-way ANOVA of overall perceived traits,
only two interaction effects were found to be statistically
significant-a two-way interaction (the gender of the politician X
the gender of the respondent) and a three-way interaction (gender
of politician X SES of respondent X gender or respondent).

Gender of politician X gender of respondent. The results

reveal that an interaction exists between respondent-gender and
politician-gender (see Figure 1). The female respondents
perceived female politicians (M = 8.86) significantly more
positively compared to male respondents (M = 5.66). However,
male and female voters perceive male politicians in a similar
fashion with almost the same mean ratings (M = 6.50).
16

Figure 1. Male and female respondents' perceived traits of male and female politicians.

10 .,-~--------------------------,

'9

.
iii
c
!.
7

~
to-
5

'i 4
~
CD
I!
:. 3

O-l------------~----------_i
Male Female
RIlspondent

Gender of politician X SES of respondent X gender of

respondent. The perceptions of traits of male and female
politicians were found to depend on both the gender and social
status of the respondents (see Figures 2 and 3). The male
respondents from ABC perceive male politicians as the least
positive in terms of traits (M = 3.30). The male respondents from
the D Class, on the other hand, think quite lowly of female
politicians (M = 3.56). The female respondents from the E Class
(M = 11. 73) think, very highly of female politicians, in terms of
traits. Both male and female respondents from the E Class think
highly of male politicians (M = 8.1).

Summary of Results

In sum, here are the highlights of the findings:

1. In general, respondents perceived political candidates

positively.
 17

Figure 2. Female respondents from ABC, D, and E social classes and their perceived traits
of male and female politicians.

12.,-----~-1
11
F2male politiciau
10
,
;;; 9
.e

!.
II
~
...
~
8

7
-------...._-- ---.....,
~
et 6

I---------------~---~- - -... -.
E o ABC
Female Respondents

Figure 3. Male respondents from ABC, D, and E social classes and their percelved traits of
male and female politicians.

9 -,----------------------------------. + --

8
Male pali tici
7

r!male politician

-l----------~---------,_----- ----t---
E o ABC
Male Respondents
18

2. Respondent's gender influenced perceived traits of political

candidates. The females consistently gave more favorable
ratings to politicians.

3. Respondents' socioeconomic status influenced perceived

traits of political candidates.

The higher the social class, the less favorable the

perception towards politicians.

4. There are no overall differences in traits of male and

female politicians. However, there are gender differences
that are trait-specific. For example, respondents believe
that female politicians are more attractive, emotional,
intelligent, and religious but male politicians are more
corrupt.

5. A respondent's gender X politician's gender interaction

effect was reported.

Female respondents tend to evaluate female politicians

more favorably whereas male respondents tend to think
more highly of male politicians.

6. A respondent's gender X respondent's SES X politician's

gender interaction effect exists. Female respondents from
the E Class think very positively of female politicians
whereas male D respondents think lowly of female
politicians. Males from the ABC think less positively of
male politicians.

DISCUSSION

Perceived Character Traits of Politfca! Candidates

The results showed that respondents think that the

hypothetical political candidate possesses mostly positive
character traits. This may seem counterintuitive as many
 19

Filipinos think that a stereotypical politician is dirty, corrupt,

and power-hungry. However, the respondents think otherwise.
This is an interesting finding indeed. It is still possible that the
respondents answered based on what is more socially desirable
and thus checked mostly positive traits. Likewise, it is also
possible that the respondents were rating on the basis of their
ideal political candidate instead of the actual. But because these
problems were foreseen and steps were taken to avoid giving
socially desirable responses, it is safe to say that Filipinos in
this sample really think positively of candidates for public office.
The political candidates in the scale were fictitious and not
associated with any real politician. Although they were told that
Mario or Marie represents a typical politician, she or he remains
faceless. Nothing was said about Mario or Marie's education,
previous experience with politics, or personality. Thus, the
respondents were free to project whatever image they have of a
faceless person running for public office. And it turned out that
their perceptions of Mario and Marie were generally positive.
This is a very encouraging and optimistic discovery. It tells us
that Filipinos generally think highly of people running for public
office.

The top two traits used to describe the fictitious political

candidate were competent and intelligent. Pierce (1993) considered
these two traits as falling under the competence dimension in
evaluating political candidates. This finding suggests that the
political candidates' capability and competence are salient in the
minds of the voter when evaluating them, more than descriptions
of their integrity and empathy.

Respondent-Gender Differences in Perceived Traits

This study has shown that males and females differ in their
perceptions of the traits of political candidates. The females in
the present study consistently showed a more favorable perception
of politicians. This finding coincides with a study in the United
20

Kingdom by Fife-Schaw and Breakwell (1990) that males were

more likely to be critical and cynical about the handling of political
crises. In the present study, the male respondents significantly
rated politicians as more emotional, fickle-minded, impatient,
indifferent, and loud whereas the female voters gave more ratings
of competence to the politicians.

Why do women perceive traits and effectiveness of political

candidates in a more positive light, compared to men? It is possible
that women are more attentive than men to candidates'
personality characteristics and thus rated them more liberally
(Eagly et al., 2003).

The fact that women think more highly of politicians ,may be

related to, the welfare state dismantlement hypothesis (Erie &
Rein as cited in Gidengil, Blais, Nadeau, & Nevitte, 2002) in the
US. According to this proposition, women should be more
supportive than men of the government's efforts in providing
employment and social welfare. It would thus be reasonable to
, assume that women would support politicians in the Philippines
more than men.

Another explanation for this gender difference focuses on

differences in women's and men's values and priorities that have
their origin in childhood socialization (Gidengil et al., 2002).
According to Gilligan's work (as cited in Gidengil et al.) on gender
differences in moral reasoning, females are less individualistic
than males. Applied to the world of politics, it is possible that
women are more willing to endorse government solutions that
translate to their perceptions of politicians in general on behalf
of the needy.
Moreover, men are more critical of political candidates perhaps
because politicians are generally male and thus with males
perceiving males, there might be a tendency to be more
competitively critical due to their macho image.
 21

Social Class Differences

The present study confirmed the second hypothesis that social

class influences the perceived traits of political candidates. In
particular, the highest social class consistently perceived political
candidates in a more negative manner. The lower the socio-
economic status, the more favorable the perception of candidates.
Compared to the D and E voters, the ABC respondents perceived
politicians as more corrupt, popular, power-hungry, and rich. The
E voters perceived politicians as more attractive, competent, for
the poor, honest, moral, and religious, which are all positive
traits.
This finding suggests that the more educated ABC class may
be more critical and have higher standards and expectations
from personalities running for public office, compared to the less
educated and politicized citizens. As Oskamp (1981) said, the
less educated citizens often focus on personalities of political
candidates.

The results are also consistent with the findings on gender

of respondent. It appears that the lower status groups, the females
and poor people, tend to be more hopeful about political candidates.
Both low power groups, the females and E class perceive politicians
more positively. Because of their lower status in society, these
groups may conceivably look up to politicians as people who will
help and serve them. For an aspiring political candidate, this
implies that the Filipino females and poor people seem to be
easier to satisfy, compared to the males and higher status voters.

Results of conformity and persuasion studies have shown

that women are slightly more influenceable than men (Eagly
& Wood as cited in Myers, 2004). Wood (as cited in Myers, 2004)
claimed that this gender difference is more a function of social
roles. In reality, women are in lower status roles in society and
thus, experience accepting more influence. Perhaps it is the
same case with respondents from the lower social class and with
attitudes toward politicians.
22

These findings support the assertion that cognition is social

and context dependent (Forgas, 1981; Moscovici, 1981; Nye & Brown,
1996). All the respondents in the study share an identical social
environment yet have varied perceptions about an identical target
person, that of a political candidate. This finding indicates that
different groups of people can subjectively view the same
contextual stimulus in varied ways. In the case of this study,
men and women, and high and low status groups differ in opinions,
insights, and observations of male and female political candidates.

One of the more significant pursuits in the study of social

and political cognitions has been the effort to view cognition as
context dependent. Social cognition and social context are
interdependent (Nye & Brown, 1996). Our social representations
are a product of a delicate interplay between information-
processing strategies and sociocultural processes (Forgas, 1981;
Moscovici, 1981). Just as cognition cannot be properly understood
without placing it into a social context, society and culture must
be studied as the product of cognitive efforts of individuals. Thus,
it is important to recognize the different cultures surrounding
high and low status groups in the Philippines.

The impact of any objective situation depends upon the

subjective meaning that the actor attaches to that situation
(Ross & Nisbett, 1991). According to Ross and Nisbett (1991), the
construal process has typically illuminated that what matters for
behavior is often not the objective situation (i.e., "how things
are") but the potentially variable, subjective interpretation of this
situation (i.e., "how a person sees things"). Thus, in the case of
politicians, perception may be very crucial; how voters see things
may be more important than objective reality. And as this study
has shown, different groups vary in how they see politicians.

Gender of Politician

Although no overall significant differences were reported in

the survey, the current research has shown some evidence that
 23

men and women political candidates are perceived as having

different traits.
Respondents have not decided that male politicians are better
overall than female politicians or vice versa. This finding implies
that Filipinos may be beginning to think in egalitarian ways,
that women and men politicians are perceived as equal.

However, respondents associate different traits with men and

women political candidates. Respondents of the survey reported
that women political candidates are more attractive, emotional,
intelligent, and religious but male politicians are perceived as
more corrupt. The interviews support this observation and add
that female politicians are also more fickle-minded, impatient,
and moral. Male politicians, aside from being corrupt, are also
competent, popular, and power-hungry. These results may be
explained in terms of the social context when the survey was
taken which was after People Power II when President Joseph
Estrada was toppled due to corruption charges.

Except for intelligence, all the traits associated with female

politicians are stereotypical of a woman, namely, being attractive,
emotional, fickle-minded, and religious. These have something
to do with one's physical attributes and personality. In Roces'
1998 study, she asserted that women politicians in the Philippines
have to use feminine practices in order to maximize their
exercise of power. For example, being a female candidate allowed
them to appeal to the audiences' emotions during campaigns.
They could also draw more crowds because of their physical
attractiveness.
The perception that female politicians were more intelligent
than males coincides with the findings of Duke (1993) who
also reported that women candidates were rated as more
in te lligen t.
24

Respondent's Gender and Politician's Gender

Interaction Effect

Would female voters tend to view female politicians more

positively? The answer is a resounding yes. The female
respondents perceived female politicians significantly more
positively compared to male respondents, when it comes to
character traits. Male respondents perceived male politicians
more positively, but to a lesser degree compared to female
respondents. Male and female respondents perceived male
politicians in a similar fashion with almost the same mean
ratings.

These findings imply that identity politics seems to be

working, especially for female respondents. Perhaps, female
respondents were able to identify with the politician of the same
sex and thus, perceived them in more positive terms. As
mentioned in the introduction, identity politics refers to the
formation of political allegiances on the basis of some
demographic similarity, such as sex, race, and religion (Plutzer
& Zipp, 1996). In this case, it was sex or gender that mattered.

Previous studies suggest that an interaction does exist

between respondent's gender and politicians' gender (Eagly
et al., 2003; Lewis & Bierly, 1990; Plutzer & Zipp, 1996). Female
university students followed a profemale sentiment in evaluating
female politicians' competence (Lewis & Bierly, 1990). Plutzer
and Zipp (1996) found evidence that sex of the voter is significantly
related to voting for female candidates in 8 of the 13 states they
studied.
Eagly et al. (2003), predicted a similar result but in relation
to voting. Their experiments supported that attitudinal gender-
congeniality effect in which participants of each gender reported
greater likelihood of voting for the candidate who endorsed
positions typically favored more by their own sex than the other
sex. Women were also found to favor the interests of their gender.
Women consistently favored the female-congenial candidate.
 25

Eagly and Karau (2002) provides an explanation for this

interaction effect. Because men have a more masculine
perception of leadership roles, they are thus expected to show a
stronger tendency to view women as less qualified. They also
stated that the tendency for men to view women as less qualified
for leadership roles derives from men's greater social power,
which can enhance their tendency to use gender-stereotypical
information rather than available individuating information.

The findings of the present study suggest that women adhere

more to the interests of the female gender than men adhere to
the interests of the male gender. This is a logical conclusion as
women's movements are very active in the Philippines whereas
there is no comparable social movement organized to promote
men's interest. Thus, it is expected that women be more conscious
I
in perceiving more positively candidates who are also women.

However, just because women think favorably of women

candidates does not immediately follow that women would vote for
women candidates. Remember in this study, all other factors were
controlled for but in reality, many other variables come into play
in the decision to vote. But certainly, for women to have more
favorable perceptions of women candidates is a crucial step but it
remains to be seen whether identity politics will translate to voting.

Respondent's Gender X Respondent's SES X Politician's

Gender Interaction Effect

A three-way interaction effect between respondent's gender,

politician's gender and respondent's social class was reported for
perceived traits of political candidates. According to the results,
females from the ABC, 0, and E social classes think more
positively of female candidates. The male respondents, on the
other hand, have varied perceptions, depending on their social
status. The ABC males prefer female candidates. While the m~les
from the 0 and E social classes believe that male politicians! are
better, character-wise.
26

The D and E classes comprise majority of the voting population

in the Philippines. It is thus crucial to look into their perceptions.
Despite being in the same economic class, the males and females
from the D and E classes disagree on their perceptions of political
candidates. The females consistently think highly of female
candidates and the males consistently think highly of male
candidates. Thus, identity politics seems to be the strongest at
the D and E classes.

This implies that gender consciousness efforts should focus

on males from the D and E classes.

Implications

In terms of practical implication, this research can be a

vehicle for raising consciousness about the underrepresentation
of women in politics, gender differences, and gender bias in
perceiving and evaluating political candidates.

Many lessons are learned which can be of practical use to

politicians, in particular.

Male politicians need to improve their image of integrity because

they are perceived as corrupt. Women politicians now have an
advantage with female voters who seem to identify with them.

When facing an audience, politician's appeal also matters. It

seems easier to satisfy females and the E Class. But when facing
a crowd of males and from the ABC Class, politicians need to. be
more prepared because these groups are more critical.

In raising consciousness about gender equality, one way is to

change stereotypes about men and women, but another way,
which may be easier, is to change perceptions of leadership
roles. Providing examples of success stories of women politicians
will help in making the politician's role as less macho. The
D and E class males would need more consciousness-raising on
gender equality in politics.
 27

REFERENCES

Aguilar, C. (1990). Women in politics in the Philippines. Philippine

Political Science Journal, 31, 39-74.
Appleton, A., & Mazur, A. (1993). Transformation or modernization:
The rhetoric and reality of gender and party politics in France.
In J. Lovenduski, & P. Norris (Eds.), Gender and party politics
(pp. 86-112). London: Sage Publications.
Arroyo, D. M. (1990, June). The usefulness of the ABCDE market
research system: A means to check social welfare and class'
attributes. Social Weather Bulletin, 90, 11-12.
Bennett, L., & Bennett, S. (1993). Changing views about gender
equality in politics: Gradual change and lingering doubts. In
L. Duke (Ed.), Women in politics: Outsiders or insiders? A collection
of readings (pp. 46-56). New Jersey: Prentice Hall.
Bourque, S., & Grossholtz, J. (1998). Politics an unnatural practice:
Political science looks at female participation. In A. Phillips
[Ed.], Feminism and politics. Oxford: Oxford University Press. I

Duke, L. (1993). Women in politics: Outsiders or insiders?:

A collection of readings. New Jersey: Prentice-Hall.
Eagly, A. H., Diekman, A. B., Schneider, M. C., & Kulesa, P.
(2003). Experimental tests of an attitudinal theory of the
gender gap in voting. Personality and Social Psychology Bulletin,
29, 1245-1258.
Eagly, A., & Karau, S. (2002). Role congruity theory of prejudice
toward female leaders. Psychological Review, 109, 573-598.
Fife-Schaw, C., & Breakwell, G. (1990). Predicting the in tention
not to vote in late teenage: A U.K. study of 17- and 18- year-
olds. Political Psychology, 11, 739-755.
Forgas, J. (Ed.). (1981). Social cognition: Perspectives on everyday
understanding. London: Academic Press.
Fox, R, & Smith, E. (1998). The role of candidate sex in voter
decision-making. Political Psychology, 19, 405-419.
28

Gidengil, E., Blais, A, Nadeau, R, & Nevitte, N. (20021. Women

to the left? Gender differences in political beliefs and policy
preferences. In M. Tramblay & L. Trimble [Eds.], Women and
electoral representation in Canada (pp. 140-1591. Don Mills:
Oxford University Press.
Kolinsky, E. (1993). Party change and women's representation in
unified Germany. In J.Lovenduski, & P. Norris [Eds.], Gender
and party politics (pp 113-146). London: Sage Publications.
Leeper, M. S. (1991). The impact of prejudice on female candidates:
An experimental look at voter inference. American Politics
Quarterly, 129,248-261.
Lewis, K., & Bierly, M. (1990). Toward a profile of the female
voter: Sex differences in perceived physical attractiveness
and competence of political candidates. Sex Roles, 22, 1-12.
Mangahas, M. (1998, May 22). Characteristics of the presidential
vote. The social climate. Retrieved August 5, 2002, from
http://www.sws.org.ph.
Mangahas, M. (1998, February 4). Filipinos don't mind having
another woman president. SWS poll on election issues. Retrieved
August 5,2002, from http:www.sws.org.ph.
Mangahas, M. (2001, May 15). No surprises in exit poll. The social
climate. Retrieved August 5, 2002, from http://www.sws.org.ph.
Mangahas, M., & Guerrero, L. (1998, August). Social weather reporting
in the Philippines. Paper presented at the 14th World Congress of
Sociology, Working Group 6/ Social Indicators at Montreal Canada.
Retrieved August 5, 2002, from http:www.sws.org.ph.
Montiel, C. J. (1986). Non-traditional Filipino political leadership in
the Aquino Cabinet. Philippine Sociological Review, 34, 102-109.
Moscovici, S. (1981). On social representations. In J. Forgas (Ed.),
Social cognition: Perspectives on everyday understanding
(pp. 181- 209). London: Academic Press.
Myers, D. (2004). Social psychology (8th ed.). USA:. McGraw Hill,
Inc.
 29

Nye, J., & Brown, A. (1996). What's social about social c~gnition?:
Research on socially shared cognition in small groups. Thousand
Oaks: Sage Publications.
Olivares-Cunanan, B. (2004, May 3). Macapagal hold on E class
spells difference. Retrieved July 25, 2005 from http: / /
www.inq7.net/opi/2004/may/03/tex.t/opLbocunanan-~-p.htm.

Oskamp, S. (1981). Attitudes and opinion. New Jersey: Prentice-

hall.
Pharr, S. (1981). Political women in Japan. Los Angeles: University
of California Press.
Pierce, P. A. (1993). Political sophistication and the Use of
candidate traits in candidate evaluation. Political Psychology
14,21-35.
Plutzer, E., & Zipp, J. (1996). Identity politics, partisanship, and voting
for women candidates. Public Opinion quarterly, 60, 30-57.
Rinehart, S. (1992). Gender consciousness and politics. New York:
Routledge.
Roces, M. (1998). Women, power, and kinship politics. Westport:
Praeger.
Rosenwasser, S., & Dean, N. (1989). Gender role and political
office: Effects of perceived masculinity/femininity of candidate
and political office. Psychology of Women Quarterly, 13, 77-85.
Ross, L., & Nisbett, R. E. (1991). The person and the situation. New
York: McGraw-Hill.
Shah, A. (1999). Measuring attitudes in test alien-cultures: When
illiteracy and unfamiliarity do not matter. Asian Journal of
Social Psychology, 2, 245-264.
Sy, B. (2001). Choosing political candidates in the Philippines: A
look at how urban poor voters evaluate political candidates.
Unpublished Master's thesis, Ateneo de Manila University.
Tancangco, L. (1992). Women's political participation in the
Philippines: The cultural dimension. In P. Tapales (Ed.), Filipino
30

women and public policy (pp. 59-108). Diliman, Quezon City:

University Center for Women's Studies.
Tapales, P. (1992). Filipino women and public policy. Diliman,
Quezon City: University Center for Women's Studies.
Women in National Parliaments (2005). Retrieved July 13,
2005, from httpr/ ywww.ipu.org.