Electoral Systems, Contextual Factors and Women's Opportunity for Election to Parliament in Twenty-Three Democracies

• Author: Wilma Rule
• Published date: 01 September 1987
• Date: 01 September 1987
• DOI: http://doi.org/10.1177/106591298704000307
• Subject Matter: Articles

ELECTORAL

SYSTEMS,

CONTEXTUAL

FACTORS

AND

WOMEN’S

OPPORTUNITY

FOR

ELECTION

TO

PARLIAMENT

IN

TWENTY-THREE

DEMOCRACIES

WILMA

RULE

Markleeville,

California

HE

PURPOSE

of

this

article

is

to

ascertain

which

electoral

arrange-

ments,

parties,

and

socioeconomic

conditions

offer

the

most

-JL

opportunity

for

women’s

election

to

parliament.

Another

objective

is

to

explicate

the

relationship

between

electoral

systems

and

other

contex-

tual

variables

as

they

relate

to

women’s

parliamentary

recruitment

in

var-

ious

countries.

Thirdly,

we

seek

to

determine

which

countries

among

twenty-three

democracies

in

our

sample

allow

the

greatest

to

the

least

po-

litical

opportunity

for

women.

PART

I.

ANALYSIS

OF

SIGNFICANT

PREDICTORS

The

Variables

Studied

and

the

Hypothesized

Relationships

Numerous

researchers

have

observed

a

positive

relationship

between

proportional

representation

(PR)

and

women’s

opportunity

for

election

to

parliament

(Duverger

1955;

Currell 1974;

Bogdanor

1984).

However,

with

few

exceptions

(Rule

1981, 1984,

and

Norris

1985)

there

has

been

no

sys-

tematic

empirical

verification

of

this

hypothesis,

nor

has

there

been

an

at-

tempt

to

control

for

other

political

and

socio-contextual

factors.

In

addition,

except

for

Norris

(1985)

there

has

been

no

systematic

study

of

the

effect

of

different

PR,

majoritarian

and

plurality

systems

on

women’s

election

to

parliament.

Moreover,

the

impact

of

district

magnitude

(M) -

the

number

of

representatives

per

district

-

on

women’s

recruitment

to

national legisla-

tures

has

not

been

investigated,

although

it

would

appear

to

be

a

fruitful

area

for

inquiry.

Previous

research

on

multimember

districts

and

women’s

s

election

to

U.S.

state

legislatures

show

a

positive

relationship’

(Clark

et

al.

1984;

Carroll

1985;

Rule

1986,

and

Darcy,

Welch

and

Clark

1987).

Received:

April

2,

1986

Revision

Received:

August

8,

1986

Accepted

for

Publication:

August

19,

1986

NOTE:

This

is

a

revised

version

of

a

paper

presented

at

the

World

Congress

of

the

Interna-

tional

Political

Science

Association,

Paris,

July

15-20,

1985.

I

wish

to

thank

the

anony-

mous

reviewers,

the

editor

and

Irving

Krauss

for

their

helpful

comments,

and

also

Robert

Ciscewski

for

research

assistance.

1

For

example,

of

the

twenty

U.S.

states

with

some

or

all

multimember

assembly

and

state

senate

districts

in

1984,

the

Pearsonian

correlation

with

the

percentages

of

women

in

both

houses

was

.29

significent

at

the

.02

level

(Rule

1986).

All

multimember

district

states

have

non-party

list

systems.

478

In

addition,

recent

work

on

district

magnitude

indicates

that

the

number

of

representatives

per

district

(M)

has

a

direct

effect

on

the

amount

of

turn-

over

in

parliament,

which

varies

from

18

percent

to

54

percent

in

thirteen

Western

countries

(Katz

1986,

Table

4.2).

The

smaller

the

M,

the

smaller

the

turnover

and

vice

versa

(Katz

1986:

97-101).

Significantly,

low

turno-

ver

has

been

pinpointed

as

a

major

reason

for

women’s

low

representation

in

the

U.S.

House

of

Representatives

(Anderson

and

Thorson

1984)

and

Britain’s

House

of

Commons

(Welch

1985).

Thus

women’s

underrepresen-

tation

in

both

these

countries’

national

legislatures

in

part

may

be a

func-

tion

of

the

single-member

district

(SMD)

majority/plurality

electoral

system.

The

bi-variate

relationship

between

district

magnitude

in

twenty-two

coun-

tries

and

the

percent

of

women

elected

to

parliament

in

1980-82

is

displayed

below

in

Figure

1.

(West

Germany

is

omitted

from

the

figure

because

of

its

hybrid

electoral

system.)

A

linear

relationship

in

Figure

1

is

shown

along

with

several

outliers

-

namely

New

Zealand,

Italy,

Austria,

Israel

and

The

Netherlands.

The

first

hypothesis

to

be

tested

statistically

is

that

the

greater

the

dis-

trict

magnitude,

the

greater

the

percent

of

women

in

parliament.

As

the

number

of

representatives

in

constituencies

increases,

the

percentage

of

the

vote

needed

for

election

diminishes

(the

general

formula

is

100/M

+ 1).