Guide on Gender Analysis of Census Data Full Draft of 6 December 2012 Contents


Table 33: Australia (2006) - Percentage of men and women who spent time providing unpaid child care



Download 4.06 Mb.
Page21/33
Date19.10.2016
Size4.06 Mb.
#4105
1   ...   17   18   19   20   21   22   23   24   ...   33

Table 33: Australia (2006) - Percentage of men and women who spent time providing unpaid child care



Type of unpaid child care provided

Men


Women


Own child(ren)

19.7

22.9

Other child(ren)

5.3

10.0

Both own and other

0.6

1.6

Source: Australian Bureau of Statistics CDATA Online (2006 Census)


In those rare cases where this kind of information is available from the census, it can be used to fine-tune either of the two methods outlined above.
445. Some potential tabulations with a gendered perspective regarding social security include:

a) Participation in the labour market of persons 65 years and over, by sex and age (in five-year increments);

b) Social security or old-age pension receipt of persons 65 years and over, by sex and age (in five-year increments);

c) Health insurance receipt of persons 65 years and over, by sex and age (in five-year increments);

d) Contributions to social security or an old-age pension of persons 15 years and over, by sex and five-year increments of age; and

e) Health insurance status of persons 15 years and over, by sex and five-year increments of age.


446. Additional factors, such as race, ethnicity, marital status or rural/urban status, may also be relevant dependent on the composition of the national population and should be added to the above tabulations when relevant.
5. Indicators
447. Minimum Set of Gender Indicators
Computable with census data
Labour force participation rates for persons aged 15-24 and 15+, by sex

Proportion of employed who are own-account workers, by sex

Proportion of employed who are working as contributing family workers, by sex

Proportion of employed who are employers, by sex

Percentage distribution of employed population by sector, each sex

Informal employment as a percentage of total non-agricultural employment, by sex (??)

Youth unemployment, by sex

Proportion of employed working part-time, by sex (??)

Employment rate of persons aged 25-49 with a child under 3 living in a household and with no children living in the household, by sex
448. Not computable with census data
Average number of hours spent on unpaid domestic work, by sex (separate housework and child care if possible)

Average number of hours spent on paid and unpaid work combined (total work burden), by sex

Percentage of firms owned by women

Proportion of children under age 3 in formal care;

Women’s share of managerial positions;

Percentage of female police officers;



Percentage of female judges.
449. A gendered analysis of the factors related to paid and unpaid work begins with the core topic areas that describe persons’ economic situations, and then considers how these may vary in a systematic manner across men and women in the population. This section outlines several indicators that may be useful to measure and describe persons’ economic situations.


  1. Unemployment rate for women and men. To calculate the unemployment rate, the economically active population is divided into employed and unemployed population, and the unemployment rate is the percentage among the economically active who are not employed (i.e. among those who are registered as such with the government. Promoting gender equality in employment is widely recognized as an essential component of economic and social development. Women’s participation in employment increases their contribution to household income and their control over the allocation of those resources. This leads to greater economic independence and self-determination, which are both important for women’s empowerment. In the vast majority of countries, adult unemployment is higher among women compared to men, with important regional differences [the bigger problem is that many unemployed women are classified as not economically active]. Unemployment is also prevalent among the youth population, especially young women. Northern Africa had the highest gap – seven points – between women’s and men’s employment rate overall in 2007, and had a gap of 12 points across young men and women (United Nations, 2010 a). Finland, on the other hand, reported that its register-based census of 2010 had found that in 2009, for the first time, the employment rate of women exceeded that of men. The economic downturn of 2009 mainly affected export industries and brought men’s employment down more than women’s. The occupational structures among women and men differ from each other, so that the majority of women work in the public or services sector, which are less vulnerable to economic trends (UNECE, 2012 c). Larger numbers of unemployed men compared to women were also found in the 2010 censuses of Belarus, the Republic of Moldova, Russian Federation, Tajikistan, and Ukraine (UNECE, 2012 c).


b) Occupation and industry for women and men. The complete analysis of the distributions of women and men by status in employment, occupation and economic sector of activity reveal gender differences and economic segregation. Cross-tabulations, for example, of sector of activity and status in employment are necessary to answer questions such as how men and women are distributed across employment status and whether they differ in any way from one branch of activity to another. But beyond mere sex-disaggregation, the analysis of the causes and consequences of gender differences should be analysed in depth. Women are usually found in specific (i.e. female) occupations and sectors that also have lower status and lower pay. This is part of the existing occupational segregation and discrimination in employment.


  1. Informal sector work. Two indicators on informal sector work can be obtained from population census data. The first is the percentage of economically active women and men employed in the informal sector by branch of activity; it shows the differences in how women and men are engaged in the informal sector. The second is the sex composition of informal sector workers by branch of activity, and illustrates the relative importance of women and men within the informal sector. In many countries, the informal sector provides women with the only opportunity for work in a situation of limited access to formal sector employment. However, those working in the informal sector lack protection, rights and representation, and represent a vulnerable population.


d) Average hours worked for women and men. One way in which the work of men and women may be different is that women tend to work part-time more often than men. The average number of hours worked, when asked in the census, can be compared between women and men, and even analysed by occupation, branch of activity or urban and rural areas. This provides a measure of intensity of work involvement. The main utility of this indicator is that it is one of the elements by which income differentials between men and women need to be qualified.
Average hours worked by women in occupation or branch i x 100

Average hours worked by men in occupation or branch i




  1. Average number of hours spent on unpaid domestic work for women and men. This measure can be computed similarly to the average hours worked indicator above to examine the branch of activity, whether unpaid domestic work, across women and men.




  1. Percentage of women and men in part-time work. Part-time work is one way employed women balance paid work with family responsibilities. In many countries, employed women typically assume most of the responsibility for domestic work.




  1. Average income by occupation for women and men. The gender pay gap reflects inequalities that affect mainly women. A simple indicator is the ratio of women’s average earnings to men’s average earnings, expressed per 100 and for the same period of time (monthly, yearly…). From the information collected on income, when included in the census questions and identified by category (salary and other), it is possible to estimate the ratio by occupation and branch of activity using the following method.


Average income of women in occupation i x 100

Average income of men in occupation i


Additionally, it is important to investigate whether the gender wage gap is due to discrimination or not. An engendered perspective would ask: for a similar position with the same education and experience, do men earn more than women? Some gender wage gaps may be the result of women being less educated, occupying lower positions, and spending more time out of the labour market long term. Considering women who are equal to men in the areas of education, occupational rank and experience, what is the average wage by sex? The above calculation for income should therefore be computed with these “educational,” “occupational rank” and “experience” qualifications. Income information should not be here but in chapter 8
h) Average household income by female and male household head. If income data are not gathered at the individual level within households, examining the average household income by female or male household head may be another useful indicator of economic well being. It can be computed similarly to average income of women and men by occupation above. If valid individual income data are available within households, women’s income could be presented as a per cent of men’s income similarly to average income of women to men in an occupation discussed just above.
450. Some useful indicators for gender analyses of social security are:
a) Percentage of female and male beneficiaries, by age appropriate to retired population

In the US, women comprise 57 per cent of all Social Security beneficiaries age 62 and older, and 68 per cent of beneficiaries age 85 and older.

b) Average monthly social security income received by women and men, by appropriate age range

c) Percentage of economically active women and men protected by social security, by age range



d) Average years of contribution and average years of benefits, by sex

6. Multivariate and further gender analyses
451. Occupational Feminization and Pay in the US. A longitudinal study (Levanon, England, and Allison, 2009) using US decennial census data found that occupations with a greater share of females pay less than those with a lower share, even when controlling for education and skill. Therefore, at each level of education and skill level within an occupation where there were more women than men working, the women were paid less than their male counterparts of the same education and skill level. The authors used census data to test two theories about why women’s work is paid less. The first theory, a gendered labour queue for certain occupations reasons that employers’ preference for men creates the greater propensity of women to be represented in lower paid occupations, while the second theory reasons that it is the proportion of women in the occupation that drives down wages or devalues women’s work. They used fixed-effects regression models, which allow the researcher to control for the stable characteristics of occupations over 50 years. Their findings largely support the devaluation view over the queuing view. Similarly, Blackwell and Glover (2008) use linked census and longitudinal study data to examine women’s participation in science, engineering and technology fields. They find that 80 per cent of women in health-related occupations (e.g. nursing) were mothers, compared with only 40 per cent in science, engineering and technology. These results show a connection between occupation and family life choices for women.
452. Labour Force Participation of Married Women in China. Research (Maurer-Fazio, Connelly, Chen and Tang, 2011) using longitudinal population census data from 1982-2000 in China examined married, urban women’s labour force participation. They find that while having a parent, parent-in-law, or person above 75 years old in the household increases women’s likelihood of being employed, while having a preschool-age child in the household decreases women’s propensity to be employed. These older people in the household may take on some of the unpaid household work that women generally do in China, thus freeing these women of working age up to take paid employment outside the home. When rural-to-urban migrant status is included in the analysis, the negative effect on women’s labour force participation of having young children in the household is substantially larger for married, rural-urban migrants than for their non-migrant urban counterparts. These rural-urban migrant women may not have other supports for child care if they are new to the urban area, so they stay at home with their preschool-aged children. Indeed, the study finds that the positive effect of co-residence with elders is greater for the rural-urban migrant women than for the non-migrants. These rural-urban migrant women are in a position where the kin-based help is likely to live in the same household, whereas non-migrant urban women with elder family members established in the urban area may receive kin-based help from kin who do not live with them.
453. Agricultural holdings in households, fertility patterns and women’s labour force participation as unpaid family workers. In the 2010 census round, several countries, following FAO recommendations, have included a question on whether households serve as agricultural production units, with their own plot of land and/or livestock. This opens up some gender-relevant opportunities for analysis as rural households with their own agricultural holdings are expected to be different from those that do not have such holdings. Women belonging to such households are expected to have higher labour force participation rates, although almost exclusively as unpaid family workers, almost all male heads of households are expected to have a wife to help them in the production, fertility is expected to be higher, and children are expected to have lower school attendance rates: boys because they need to help on the family holding, girls because they must replace their mothers in household duties. However, in order to bear out such relationships, certain statistical controls have to be included. It may be appropriate to control for the socio-economic level of households, for example, by introducing some sort of wealth index based on the quality of the dwelling and the ownership of consumer durables. Moreover, it is probably advisable to control for the presence of other adults, such as grandparents, in the household as these may take over some of the household tasks of spouses.
454. On the basis of the 2010 Aruba Population and Housing Census, two separate analyses were done to look into the occupational status of women and men: a) the type of organization of work and b) the status of employment. For both analyses multinomial logistic regressions were set up. The multinomial logistic regression is the extension of the simple logistic model with a dichotomous dependent variable. The multinomial model allows for a categorical dependent outcome with more than two levels. In the analysis, one of these categories has to be chosen as a residual (or reference) category. In the analysis, all other categories of the dependent variable are then compared to this category. The regression coefficients and odds ratios of the predictors in the multinomial regression are equivalent to those of the simple logistic model, i.e. each category of a given predictor is compared to the residual category of that predictor in terms of their probability of occurring.
455. In the Aruban census the following categories of type of work were used: 1) Limited corporation, 2) One-person business, 3) Foundation, 4) General partnership, 5) Association, 6) Government institution, 7) Government company and 8) Other. In the analysis, categories ‘General partnership’ and ‘Association’ were placed in the category’ others’, as they had very few cases. The reference category for type of work was ‘Limited Corporation’.
456. Table 34 presents the results of the multinomial logistic regression for type of organization worked for. The results show that, after controling for age, education, marital status and country of birth, large differences remain between male and females in terms of the type of organization for which they work. Compared to employment in a limited corporation, women are less likely than men to be economically active in a one-person business (odds ratio = 0.728) or in a government company (odds ratio = 0.391). Their chances are almost equal to those of men to find work in government department (odds ratio = 0.986), but they are much more likely to work for a foundation (odds ratio = 3.441) or the ‘other type’ (odds ratio = 1.306). Differences between males and females are highest for ‘Government Company’ and ‘Foundation’. On Aruba, utilities (water, gas, electricity…) are placed in government companies. More men than women work here. On the other hand, many of the organizations in public service (elderly homes, health organizations) and education are foundations. On Aruba, jobs in these sectors are clearly dominated by women.
457. Table 35 sheds some light on the gender differences in status of employment. Categories for status employment in the analysis are: 1) Employer (3 or more employees), 2) Small independent, 3) Small independent, without employees, 4) Temporary employee deployed by a job agency and 5) Temporary employee, volunteer, non-paid family member. The last category actually consists of three response categories in the census questionnaire. As there were only a small number of cases in these categories, they were brought together.
458. Again the same predictors were chosen. The reference category for status of employment is ‘salary earner’. Compared to this reference category, women are less likely than men to be found in any of the other categories of employment status. The differences between men and women are biggest in those categories that involve independent entrepreneurship, i.e. employer (3 or more employees), small independent, small independent without employees. The odds of being an employer with 3 or more employees are about 2.5 times bigger for males than for females (odds ratio = 0.403). Also, men are about three times more likely than women of being a small independent without employees (odds ratio = 0.339) and twice more likely to be a small independent with one or two employees (odds ratio = 0.512).
Table 34: Aruba (2010) - Multinomial logistic regression of type of organization that women work for, compared to men, by various explanatory variables


 

Reference category =

One-p. business

Foundation

Govt.dept.

Govt. company

Other

 

Limited corporation

B

exp(B)

B

exp(B)

B

exp(B)

B

exp(B)

B

exp(B)

Intercept




-2.124

 

-4.308

 

-2.065

 

-2.779

 

-3.636

 

Age




0.008

1.008

0.026

1.027

0.007

1.007

-0.005

0.995

0.028

1.029

Sex

Male

0.000

.

0.000

.

0.000

.

0.000

.

0.000

.

 

Female

-0.318

0.728

1.236

3.441

-0.014

0.986

-0.940

0.391

0.181

1.306

Marital status

Never married

0.000

.

0.000

.

0.000

.

0.000

.

0.000

.

 

Married

0.152

1.164

0.046

1.047

0.166

1.180

0.332

1.393

-0.871

0.460

 

Divorced/Legally sep.

0.074

1.077

-0.161

0.851

0.145

1.156

-0.013

0.987

-0.637

0.604

 

Widowed

0.432

1.540

-0.338

0.713

0.034

1.035

0.351

1.421

-0.513

0.777

Educ. attainment

None

0.000

.

0.000

.

0.000

.

0.000

.

0.000

.

 

Primary

0.038

1.039

-0.204

0.816

-0.104

0.901

0.126

1.135

-0.151

1.001

 

Lower vocational

-0.100

0.905

0.219

1.245

0.276

1.318

0.770

2.160

-0.397

0.813

 

High school (4 - 6 yrs)

-0.169

0.845

-0.079

0.924

0.437

1.548

0.501

1.650

-0.546

0.682

 

Higher vocational

-0.256

0.774

1.150

3.157

0.904

2.469

1.373

3.949

-0.612

0.665

 

Higher (BA - MA - PhD)

-0.354

0.702

2.289

9.862

1.468

4.339

1.076

2.934

-0.540

0.714

Country of birth

Aruba

0.000

.

0.000

.

0.000

.

0.000

.

0.000

.

 

Colombia

0.458

1.580

-2.169

0.114

-2.845

0.058

-3.225

0.040

1.633

5.738

 

USA

0.415

1.515

-1.474

0.229

-1.646

0.193

-1.576

0.207

0.919

4.479

 

Dominican Republic

0.282

1.326

-1.505

0.222

-2.482

0.084

-2.123

0.120

0.967

3.102

 

Venezuela

0.390

1.478

-1.824

0.161

-2.569

0.077

-3.302

0.037

1.095

3.629

 

Curaçao

-0.169

0.844

-0.461

0.631

-0.752

0.471

-0.859

0.424

0.221

1.750

 

Netherlands

-0.014

0.986

-0.181

0.835

-0.561

0.571

-1.709

0.181

0.354

1.868

 

Other

0.124

1.132

-1.172

0.310

-2.240

0.106

-2.272

0.103

1.118

3.452



Table 35: Aruba (2010) - Multinomial logistic regression of women’s status in employment, compared to men, by various explanatory variables



 

Reference category =

Employer (3 or more employees)

Small independent

Small independent, without employees

Temporary employee deployed by a job agency

Temporary employee, volunteer, unpaid family member

 

Salary earner

B

exp(B)

B

exp(B)

B

exp(B)

B

exp(B)

B

exp(B)

Intercept




-5.982

 

-5.148

 

-4.143

 

-2.080

 

-1.008

 

Age




0.032

1.033

0.025

1.025

0.029

1.030

-0.021

0.980

-0.019

0.981

Sex

Male

0.000

.

0.000

.

0.000

.

0.000

.

0.000

.

 

Female

-0.908

0.403

-0.670

0.512

-1.080

0.339

-0.117

0.890

-0.137

0.872

Marital status

Never married

0.000

.

0.000

.

0.000

.

0.000

.

0.000

.

 

Married

0.720

2.054

0.776

2.172

0.177

1.194

-0.256

0.774

-0.341

0.711

 

Divorced/Legally sep.

0.386

1.471

0.596

1.815

0.230

1.259

-0.132

0.877

-0.167

0.846

 

Widowed

0.863

2.370

0.867

2.379

0.447

1.564

-0.085

0.918

0.190

1.209

Educ. attainment

None

0.000

.

0.000

.

0.000

.

0.000

.

0.000

.

 

Primary

0.179

1.196

0.377

1.458

0.056

1.057

-0.392

0.676

-0.262

0.769

 

Lower vocational

-0.075

0.928

0.221

1.248

0.045

1.046

-0.711

0.491

-0.531

0.588

 

High school (4 - 6 yrs.)

0.729

2.073

0.466

1.594

-0.049

0.952

-0.538

0.584

-0.503

0.605

 

Higher vocational

0.619

1.858

0.378

1.460

-0.128

0.880

-0.801

0.449

-0.905

0.404

 

Higher (BA - MA - PhD)

1.243

3.466

0.581

1.787

-0.281

0.755

-1.174

0.309

-0.792

0.453

Country of birth

Aruba

0.000

.

0.000

.

0.000

.

0.000

.

0.000

.

 

Colombia

-0.024

0.976

0.208

1.232

0.493

1.638

0.393

1.481

0.512

1.668

 

USA

1.291

3.638

1.055

2.873

0.670

1.954

0.157

1.170

1.059

2.885

 

Dominican Republic

0.217

1.242

0.080

1.083

0.426

1.531

0.488

1.629

0.625

1.869

 

Venezuela

0.568

1.764

0.743

2.102

0.512

1.669

0.320

1.377

0.733

2.082

 

Curaçao

0.367

1.444

0.382

1.465

0.099

1.104

-0.394

0.674

0.212

1.237

 

Netherlands

0.892

2.440

0.624

1.865

0.333

1.396

0.203

1.225

0.448

1.565

 

Other

0.812

2.253

0.508

1.663

0.268

1.308

0.125

1.134

0.412

1.510

Source: Population and Housing Census Aruba, 2010


7. Interpretation, Policy and Advocacy
459. Compared to their male counterparts, women participate in the labour market at a lower rate and are represented in higher numbers in less lucrative occupations and sectors of the economy. This is not a coincidence, but a pattern all over the world, which reflects discriminatory practices in the labour force (i.e. education, selection, promotion, etc.). And, women still earn less than men even after controlling for hours worked, education and skills over 15 years after the Beijing Declaration (United Nations, 1995 b) affirming women’s right to employment and productive resources, and the Millennium Declaration’s further commitment to full, productive employment for both women and men.
460. Unpaid work mainly carried out by women need to measured, valued and accounted in the national accounting systems. In this regard, data should be collecetd in such a way that household and caretaking work, predominantly done by women across countries, is not misclassified and underestimated. Related to this, policies should be enacted that pay women for doing domestic work (e.g. Canada pays an allowance for unpaid and caretaking work) and provide access to daycare to help families manage work and household responsibilities.

461. Advocates should inform policymakers and the general public about the importance of this unpaid work done by women, when in turn allows men to do paid work. Also, advocates can alleviate women’s domestic burden by sensitising the general public to inequalities in the amount of domestic work by sex among those couples where both women and men work in the labour market. Finally, the women’s domestic burden may be lightened by providing basic infrastructure (e.g. clean, running water) and labour-saving equipment and technologies (e.g. cooking, grinding and cleaning appliances) accessible to all.

462. The Institute for Women’s Policy Research (IWPR, 2011) has used US Census Bureau data to examine women’s disadvantaged economic position with respect to social security benefits and retirement. The institute finds that older American women are more likely to face poverty than older men, especially unmarried or widowed women. Women’s median annual social security benefits reach just 70 per cent of that of men. Further, these social security benefits mean the difference between living in poverty or not for over two-thirds of unmarried women living alone. With Medicare health insurance covering individuals at age 65, few women or men lack health insurance. The IWPR combines analysis with policy activism in order to represent the interests of women. Their research on social security pensions and Medicare health insurance raises awareness of the income inequalities within the US pension system.
463. The World’s Women report (United Nations, 2010 a) found that one-half of the countries worldwide meet the new international standard for minimum duration of maternity leave and that 40 per cent meet the minimum standard for cash benefits, but there remains a gap between statutory law and what is practiced. Many women, in particular those who do not work in the formal or public sectors, are not covered by the legislation. Oun and Trujillo (2005) make the case that where the maternity benefit funds come from is the reason for this inequality towards women. They suggest that payment with public funds or social insurance could reduce this inequality and gap between law and practice. Employers no longer bear the direct costs of maternity. Currently, about one in four countries, especially in Africa, Asia and the Arab States, continue to provide payment during maternity leave through the employer with no public or social security assistance.


Download 4.06 Mb.

Share with your friends:
1   ...   17   18   19   20   21   22   23   24   ...   33




The database is protected by copyright ©ininet.org 2024
send message

    Main page