Female participation in housing activities: some assessment of the socio-economic and cultural impact.

Mujahid-Mukhtar, Eshya ; Noor-ul-Hassan

INTRODUCTION

The economic and social importance of shelter provision is universally well-established. For an individual/family the acquisition of shelter is, on the basis of its financial magnitude, among the most critical economic decisions. This decision is usually permanent and in most cases made once-in-a-lifetime. The importance of housing decision is even more pronounced for women in Pakistan, as a vast majority of them are classified as 'housewives' and thus considered responsible for the up-keep of the house. The quality of housing, therefore, is taken as being reflective of their competence and, hence, social status.

It is in view of the foregoing that one expects an extensive and willing participation of women in housing activities. However, a number of socio-economic and cultural factors constrain this participation. The purpose of this paper is to assess the impact (positive and negative) of socio-economic and cultural factors upon female participation in housing activities in Pakistan. The study is divided into the following sections: Section I highlights the role of women in the housing activities, emphasising the types of participation, the nature of activities in which women generally participate and the various socio-cultural and economic factors which influence this participation. Section II presents the methodology and data. Section III contains the results and the conclusion.

I. WOMEN'S ROLE IN HOUSING ACTIVITIES

Since no comprehensive work on the role of women in housing in Pakistan has been undertaken, therefore no statistical information exists in this context. In this regard, relying upon personal observations and verbal feedbacks, one can only assert that although, at present, women in Pakistan play an active role in the 'housing' process yet there is a potential for more participation at all stages of the process.

Participation of women in housing stretches over a variety of activities: decision-making about the various aspects of house-building and maintenance; financial contribution; supervision of various activities and actual physical construction of the structure.

I. Types Of Participation by Women

Similar to men, women are also capable of participating in housing in the following manner:

(i) Decision-making

This is an important aspect of participation. In more conservative societies, women tend to be ignored in this process, and their input begins only after the major decisions have been taken e.g., the decision to buy or move into a particular house, or the decision to renovate or expand the covered area, etc.

(ii) Financial Contribution

In a society where most women are housewives and do not necessarily possess very many assets, this type of participation is, though present, but limited. However, many women contribute towards repairs, improvements, additions, etc. through 'their' savings, i.e., they 'skim' a portion of the household/kitchen budget and set it aside for the time of need. These funds are readily available for the investment in the quality of housing because the housing quality reflects not only the economic status of the household, but also the social status of women.

(iii) Physical Labour

Women perform two major types of physical activities in housing: supervisory and/or actual physical construction of the house (or repairs, improvements and maintenance). This study has tried to incorporate the various physical activities in which women indulge during the building or maintenance of the house. Details of the nature of activity e.g., construction of walls and roof, wood work, paint, etc. were specifically outlined in the survey. However, because of a better developed labour market in the urban areas and the social stigma attached for the physical participation of women in 'rough and tough' housing activities, incidence of physical participation is expected to be relatively low in urban areas.

2. Housing Activities

Once the type of participation of women has been categorised, it is important to briefly list the activities which form part of the overall 'housing' activity. These include:

(i) Plot Acquisition

The role of women, with reference to the decision and purchase of the plot. With no physical labour involved in this activity, women's role is confined to the decision and finances.

(ii) Purchase or Construction of House

Houses are either purchased or built on acquired plot. In the former case, the question of physical input of the buyer is usually not relevant; therefore, in such cases, only the decision and financial aspects of the women's role are applicable. In the latter case, however, the physical role becomes crucial.

(iii) Repairs/Maintenance

These activities are usually related to only the replacement of 'wear and tear' of the house. Typically, these may include the replacement of water faucets, electrical fittings, etc. It is important to mention that women are capable of contributing in all aspects i.e. decision, finances and physical labour, in such activities.

(iv) Improvements

These activities involve a betterment of the existing structure or facilities. Typically, these include the conversion of katcha house to a pucca one. Installation of netting to windows, etc. are examples of housing improvements. Once again, similar to repairs, these activities may also involve women's decisions, finances and physical contribution.

(v) Additions

Seldom does a house remain the same after some years. Most houses experience renovation, with many changes such as more rooms, garages, servant quarters. In residential units, the most common of such additions is the construction of the top storey. Additions and/or alterations are fairly common and the role of women in these additions is important for the assessment of women's role in housing.

3. Factors Influencing Participation of Women in Housing

The above stated types of participation and housing activities should be studied with reference to the following socio-economic and cultural factors: household income, family size, family system, education, employment, decision-making and urban/rural and provincial location.

(i) Household Income

As regards the influence of household income upon the female participation in housing activities is concerned, there are two types of opposing phenomena which operate: the "need to pay" and the "ability to pay". With increases in family income, a women may not feel the need to pay, although she may possess the ability to pay. On the other hand, a women of a poorer household may be compelled by the need to contribute. It is, therefore, expected that although the trend in financial participation of women may not necessarily change with an increase in the household income, but physical participation of women in housing activities is bound to decline with household affluence.

(ii) Family Size

The size of family is likely to affect positively the participation of women in housing activities: the greater the family size, there are likely to be more women in the household, which increases the chances of female participation. However, it is quite likely that more women may also imply more elderly women who may or not be able to participate, especially in physical activities. However, in physical labour, the absolute number of male members (above 10 years of age) is likely to adversely affect the participation of women.

(iii) Family Systems

The role, particularly the one related to decision-making, of a woman is greatly influenced by the family system in which she lives. A mother-in-law enjoys considerable authority in an extended family system. The daughter-in-law wields much more power in a nuclear family than in a system with other family members, especially those of the husband's. This study has identified three types of family system in which the households exists:

(a) Nuclear Family: This includes the husband, wife, with or without their children;

(b) Extended Family (with same budget): This includes a nuclear family, with other family members (of either or both spouse), but who have common expenses, especially of the kitchen; and

(c) Extended Family (with different budgets): This includes those defined in (b) but with separate budgets for various individuals or nuclear units.

The women in nuclear families tend not only to wield greater authority in household matters but are also psychologically well-aware of the fact that the fruits of their efforts (physical labour or financial investments) will be totally enjoyed only by their husband and children. Therefore, it is expected that in the physical and financial involvement of women in housing activities, nuclear families are likely to display a higher rate of female participation in relation to non-nuclear families.

(iv) Education of Women

Educatio i considere [beta] vita facto i determinin th adegre io aemancipatio ai awomen<< mor educate woma ai ano aonl ibette atraine fo worldl chores suc a ahousin aactivities ietc. abu sh als earn [beta] degre o relianc fro othe afamil imember t b entruste wit suc matters.

However, the actual physical labour in housing activities is usually considered to be a menial task, which requires a tough physique, and hence "unlady-like" in the educated enclaves of society. It is, therefore, quite likely that education will exert a negative impact upon the physical participation of women in housing activities. One may, however, anticipate that financial contribution need not be negatively correlated with education.

(v) Employment of Women

Similar to education, employment outside the house also equips a woman with an ability not only to participate better in matters related to housing but also to contribute financially in such activities. Employment also provides the woman with a certain degree of confidence which can facilitate her physical and supervisory participation in housing activities. On the other hand, an employed woman may be restricted for time due to other responsibilities.

(vi) Decision-making

One of the most crucial factors which is likely to influence the involvement of women in any financial or physical activity is the degree of her participation in the decision-making process of that activity. It is anticipated that there will be a strong positive relationship between the female decision-making and her subsequent participation in any activity.

(vii) Urban/Rural Location

In a society like Pakistan, where major disparities in income, education, and other socio-economic factors exist between the urban and rural sectors, there is also a wide disparity between the attitudes and endowments of women. In general attitude, an urban women tends to be less inclined towards hard physical labour than the rural women who is capable of performing tasks which require more physical hardship. This is supplemented by a better developed labour market in the urban areas. In view of these qualities, one may expect a relatively higher financial contribution of women in urban areas as compared to their rural counterparts, whereas a higher physical participation of rural women as against urban women.

II. METHODOLOGY AND DATA

Given the various types of participation i.e., financial and physical, the impact of various socio-economic and cultural variables was estimated for each type of participation in each of the housing activities i.e., plot acquisition, construction, repairs, improvements and additions. (2)

Financial Contribution

To measure the effect of a series of variables upon the financial contribution of women in various housing activities, the following set of equations was estimated:

FCWi = a0 + a1 FSIZE + a2 INCOMEF + a3 INCOMEW + a4 EMPLW + a5 EDUCW + a6 EXTS + a7 EXTD + a8 DECi + a9 DECBi + a10 P1 + a11P2 + a12 P3 a13 URBAN ... ... ... ... ... ... ... ... ... (1)

Where:

FCWi : Proportion of women in the household contributing financially towards activity i, where i is plot acquisition, construction, improvements or addition.

FSIZE : Family size.

ME: Famil incom (i R pe month).

INCOMEW : Income of the women in the household (in Rs per month).

EMPLW : Dummy variable for the employment status of the woman respondent. 1 = employed; 0 = otherwise.

EDUCW : Dummy variable for the literacy status of the woman respondent. 1 = literate; 0 = otherwise. EXTS : Dummy variable for extended family system with same budget. 1 = extended family with same budget; 0 = otherwise.

EXTD : Dummy variable for extended family system with different budgets. 1 = extended family with different budgets; 0 = otherwise.

DECWi: Dummy variable for women participating in decision-making in activity i. 1 = women participated in decision; 0 = otherwise.

DECBi: Dummy variable for both husband and wife participating in decision-making in activity i. 1 = both participated in decision; 0 = otherwise.

P1 : Dummy variable for the province, Balochistan. 1 = Balochistan; 0 = otherwise.

P2 : Dummy variable for the province, North-West Frontier Province. 1 = NWFP; 0 = otherwise.

P3 : Dummy variable for the province, Punjab. 1 = Punjab; 0 = otherwise.

URBAN : Dummy variable for the urban location. 1 = urban; 0 = otherwise.

Physical Participation

Physical participation of women in housing activities is sub-divided into two types of activities: (i) the actual physical labour in construction, repairs, improvements and additions; and (ii) the supervision of each of the above housing activity.

One must however, include the absolute number of male and females as explanatory variable in the specification for the physical participation of women as it is expected that the presence of more men in the house is likely to reduce the physical responsibilities of women especially with regards to the strenuous housing activities.

The following set of equations were estimated for each of the supervisory as well as actual physical participation of women in household activities:

PCWi = a0 + a1 MALE + a2 FEMALE + a3 INCOMEF + a4EMPLW + a5EDUCW + a6EXTS + a7EXTD + a8DECWi + a9DECBi + a10P1 + a11P2 + a12P3 + a13 URBAN ... ... ... ... ... ... ... ... (2)

and

SCWi = a0 + a1 MALE + a2 FEMALE + a3 INCOMEF + a4EMPLW + a5EDUCW + a6EXTS + a7EXTD + a8DECWi + a9DECBi + a10P1 + a11P2 + a12 P3 + a13 URBAN ... ... ... ... ... (3)

Where:

PCWi : Proportion of women in the household contributing in terms of actual physical labour in activity i, where i is construction of house, improvements or additions.

SCWi : Proportion of women in the household supervising activity i, where i is construction of house, improvements or addition.

MALE : Number of male family members (above 10 years of age).

FEMALE : Number of female family members (above 10 years of age).

The data for this analysis was obtained from the survey undertaken in 1989 for the AERC/World Bank study on the role of women in housing and housing finance. The information was collected on a national level in all the four provinces for both urban as well as rural areas. Although the data obtained were for both owner-occupied as well as tenant households, but this analysis is limited to only the owner households. The total number of households analysed is 829, with 370 belonging to urban areas and 454 to rural areas. The household questionnaire was based upon a male as well as a female module, with the latter being administered by the female interviewers to that woman who knew the most about housing in the family.

III. RESULTS AND CONCLUSION

According to the sample, one clearly observes a uniform pattern across various categories of socio-economic variables, such as income, family system, literacy and employment: in financial contributions, urban women tend to display a higher participation rate than rural females; for physical participation, the reverse pattern holds (Table 1).

The econometric estimates provide quantifiable effects of various continuous and dummy variables, indicating their (in)significance. With reference to the financial contributions (Table 2), the following factors generally tend to exert a positive and significant impact upon the proportion of women contributing financially towards various housing activities: women's own income, their employment, their involvement in the decision-making process for that particular activity, the nuclearity of the household and its urban location.

The negative, although insignificant, coefficient for family size indicates that either larger families tend to have more men, or that most women are not able to afford any financial participation. It is easier to understand the negative, although insignificant, effect of the family income variable as the inclusion of women's own income variable in the regression renders the former as capturing the changes in men's incomes only. Literacy fails to generate a significant effect in any activity.

In physical participation (Table 3), the nuclearity of the household, the involvement of the woman in the decision-making process, her employment and the rural location of the household indicate a positive influence upon the proportion of women contributing to the housing activities. As expected, literacy exerts a negative, generally significant, effect upon physical contributions.

A striking feature, contrary to expectations, in the regressions for supervision and physical participation is the negative and significant effect of the number of adult females. The only explanation one may provide is that the physical contributions to housing activities require a tough physique and a certain degree of skill which all women in the household (particularly the elderly) may not be able to possess and as such their number may increase in absolute terms, but as a proportion the contributing women may decline. (The positive, though mostly insignificant, effect of the adult males on the physical participation of women was also not expected.)

In the estimates for supervision (Table 4) the pattern is somewhat unclear. However, the important variables, influencing positively and significantly the supervisory role of women include woman's employment and her involvement in the decision-making process.

In light of the above analyses, one may conclude that women tend to play a significant role in the activities related to housing and housing finance. However, this role is defined by a series of socio-economic and cultural factors related to location, family system, women's employment and their participation in the decision-making process.

Urban women tend to contribute more towards financial requirements of housing activities, whereas rural women are more active in the physical aspect of such activities. It is also observed that employed women are more involved both financially as well as physically in housing activities than the unemployed women. As regards the family system, women belonging to nuclear households tend to participate more than those residing in joint units. The involvement of women in the decision-making process is a vital determinant of their participation in the various types of housing activities.

In the regression table, employment and literacy variables correspond to the female respondent only.

Comments on "Female Participation in Housing Activities: Some Assessment of the Socio-economic and Cultural Impact"

The paper addresses an important issue which has so far received little attention from researchers and the authors should be commended for taking the initiative in this area. The main objective of the paper is to assess the influence of socio-economic and cultural factors on women's participation in various types of housing activities. The. paper distinguishes between three types of participation-financial contribution, physical labour and participation at the decision-making level. However the article only concentrates on women's financial and physical contribution to housing.

The first few comments relate to the statistical analysis which is the main focus of the paper. The regression equation incorporates a wide range of factors which are expected to influence women's participation in housing activities such as women's income, employment and education, household income etc. However the regression specification does not include variables for age and for the relationship of the respondent to the head of household. In the context of Pakistan these two characteristics of the respondents would be expected to have considerable influence on all aspects of participation in housing activities. For instance the contribution of an elderly mother-in-law is likely to be quite different from that of a young unmarried girl.

Another shortcoming of the regression specification is the use of a dummy variable for women's employment status which has only two possible outcomes that a women is employed or she is unemployed. This totally disregards the considerable evidence for Pakistan as well as for other Third World countries that it is not employment per se which influences women's behaviour but the type of employment. The control over resources or a say in household decision-making are likely to be different for women engaged in subsistence production on their own farm to those of women involved in the market economy and within the market economy between women working in the formal and in the informal sectors. It would have been useful to allow for more detail by including three or four broad categories of employment in the analysis ranging from subsistence production to work in the formal sector. Similarly the dummy variable for literacy does not capture the influence of varying levels of education.

With respect to the regression results, some of the findings are counter intuitive and need further explanation, for instance, the positive and significant coefficient for Balochistan as an explanatory variable for women's financial contribution to housing activities is puzzling. It is not clear why Balochi women should be more likely to make a financial contribution to housing as compared to women from other provinces, specially in view of the relatively lower status of women in that area in terms of education, employment etc. Similarly physical participation in housing would be expected to be negatively related to the income status of the household and also to the number of adult women in the house. However, neither of these variables is significant.

The conclusion that women tend to play an important role in activities related to housing and housing finance is not warranted by the analysis presented in the paper and is quite misleading. Infact the paper does not provide answers to the question of what is the extent of women's involvement in housing activities. The study does not present basic information on the proportion of women in the sample who contributed to housing activities either through their physical labour or in the form of financial resources or had a say in the decision-making process. There is no data on the socio-economic characteristics of the respondents such as their level of education, employment status etc. The findings indicate that there were instances of households where major decisions related to the acquisition of a plot and construction of a house were taken by the women themselves without any consultations with their husbands. It would be interesting to know more about such households, their socio-economic class or whether they were mainly female headed households. An analysis of these issues would have added more to an understanding of the subject than the statistical exercise which is the main focus of the paper.

Finally and most important the paper, even within its limited scope, fails to address the key aspect of participation which is participation at the decision-making level. Physical participation in the sense of providing physical labour or supervision and participation in decision-making have very different connotations for women's status. A say in important financial decisions has been singled out as a particularly powerful indicator of women's status in the household.

The important question therefore is what are the factors which determine women's participation in housing activities at the decision-making level? The paper totally sidelines this main issue by assuming that women's input into decision-making is determined by external factors. However a priori it is quite likely that women's control or voice in family decision is likely to be determined by factors such as household status, women's employment, education, age, etc. Data for these variables was available and could have been used to assess their impact on women's participation at the level of decision-making. In fact studies along these lines has been undertaken for various developing countries as well as for a sample of rural women in Pakistan.

Shahnaz Kazi

Pakistan Institute of Development Economics, Islamabad.

(1) Except the AERC/World Bank Study, 1989. This work is based on the data collected for that study.

(2) Ordinary Least Squares (OLS) were used to estimate the impact of various socio-economic and cultural factors on financial and physical participation of women. As OLS and the limited-dependent-variable techniques (logit and probit) yielded almost similar results, results from OLS are quoted in the paper on the ground that they are easier to interpret.

Eshya Mujahid-Mukhtar is associated with the Applied Economics Research Centre, University of Karachi, Karachi.

Table 1
Percentage of Households in which Women Participate
in Housing Activities by Selected Socio-economic Variables

 Financial Physical
 Contribution Contribution
 Rural Urban All Rural Urban All

Family Income
 Low 13.33 52.17 30.19 85.29 75.00 82.00
 Lower-middle 26.67 45.95 31.21 76.42 17.65 69.29
 Upper-middle 17.65 47.22 29.89 80.43 55.56 73.44
 Upper 6.25 38.46 34.93 57.14 30.00 35.14

Family System
 Nuclear 24.24 48.36 37.56 86.41 42.37 70.37
 Same Budget 18.37 36.96 27.37 68.42 33.33 57.14
 Diff. Budget 20.00 25.00 21.88 75.00 42.86 66.67

Literacy
 Illiterate 22.63 45.03 32.55 78.17 41.98 67.63
 Literate 11.11 37.33 30.39 71.43 30.00 47.06

Employment
 Status
 Unemployed 20.00 38.28 29.23 76.73 36.54 63.07
 Employed 41.67 94.12 72.41 87.50 71.43 82.61

Table 2
Regression Estimates for Financial Contribution
of Women in Various Housing Activities

 Proportion of Women in Household
 Contributing Financially Towards

Dependent
Variable Plot Construc- Repairs
 tion
Explanatory Var.
Constant 0.01 -0.05 -0.05
 (0.18) 0.00 (-1.66)
Family Size -0.004 -0.001 -0.001
 (-0.77) (-0.61) (-0.66)
Family Income -0.000002 -0.0000003 -0.000001
 (-1.11) (-0.27) (-1.76)
Women's Income 0.00002 0.00002 0.00002
 (1.78) (2.06) (2.24)
Employment 0.12 0.12 0.17
 (1.74) (3.32) (5.48)
Literacy -0.03 -0.03 0.02
 (-0.74) (-1.44) (-0.98)
Ext. Family -0.06 -0.02 0.02
 (Same Budgets) (-2.04) (-1.31) (1.52)
Ext. Family -0.06 -0.06 0.02
 (Diff. Budgets) (-2.99) (-2.06) (0.56)
Decision 0.20 0.21 0.04
by Women (3.33) (4.44) (2.39)
Decision by 0.02 -0.03 0.01
Husband and Wife (0.37) (0.99) (0.63)
Urban 0.07 0.08 0.04
 (1.93) (4.76) (2.61)
Balochistan 0.20 0.04 0.06
 (3.14) (1.59) (2.54)
NWFP 0.01 0.01 -0.01
 (0.18) (0.36) (-0.43)
Punjab 0.05 0.03 -0.005
 (1.47) (1.53) (-1.66)
No . of Obs. 302 741 298
[R.sup.2] 0.20 0.12 0.22

 Proportion of Women in Household
 Contributing Financially Towards

Dependent
Variable Improve- Additions
 ments
Explanatory Var.
Constant 0.09 0.07
 (2.13) (1.23)
Family Size -0.005 -0.004
 (-1.59) (-1.27)
Family Income -0.0000005 -0.000002
 (-0.34) (-0.92)
Women's Income 0.00006 0.000006
 (4.43) (0.58)
Employment 0.12 0.09
 (2.24) (1.64)
Literacy -0.02 0.002
 (-0.94) (-0.08)
Ext. Family -0.03 -0.02
 (Same Budgets) (-1.29) (-0.61)
Ext. Family -0.005 -0.05
 (Diff. Budgets) (-0.22) (-1.14)
Decision 0.05 0.04
by Women (1.59) (1.25)
Decision by -0.005 0.05
Husband and Wife (0.22) (1.91)
Urban -0.01 0.01
 (-0.59) (-0.42)
Balochistan 0.08 -0.03
 (2.05) (-0.64)
NWFP -0.02 -0.03
 (-0.51) (-0.86)
Punjab 0.03 0.02
 (1.04) (0.64)
No . of Obs. 367 315
[R.sup.2] 0.15 0.06

Note: Terms in parantheses are the t-statistics. In the regression
table, employment and literacy variables correspond to the female
respondent only. The omitted category. for the extended family dummies
is the nuclear family. The omitted category in the decision dummies is
the decision by only the men. The omitted category in the provincial
dummy is Sindh.

Table 3
Regression Estimates for Physical Participation
of Women in Various Housing Activities

 Proportion of Women in
 Household Participating
 Physically Towards
Dependent
Variable Construction Repairs

Explanatory Var.
Constant 0.29 0.67
 (5.36) (8.90)
No. of Adult -0.032 -0.025
 Women (-3.39) (-1.81)
No. of Adult Men 0.007336 0.003118
 (1.07) (0.30)
Family Income 0.0000018 -0.0000013
 (0.97) (-0.60)
Employment 0.00668 0.22529
 (0.12) (2.62)
Literacy -0.02 -0.09
 (-0.71) (-2.05)
Ext. Family -0.06 -0.06
 (Same Budgets) (-2.32) (-1.58)
Ext. Family -0.03 -0.04
 (Diff. Budgets) (-0.65) (-0.48)
Decision 0.08 0.14
by Women (1.38) (2.96)
Decision by -0.05 -0.005
Husband and Wife (-1.21) (-0.10)
Urban -0.07 -0.26
 (-2.44) (-6.35)
Balochistan -0.05 -0.15
 (-1.01) (-2.12)
NWFP -0.03 0.009
 (-0.86) (-1.19)
Punjab 0.01 -0.02
 (0.48) (-0.55)
No. of Obs. 394 311
[R.sup.2] 0.10 0.26

 Proportion of Women in
 Household Participating
 Physically Towards
Dependent
Variable Improvements Additions

Explanatory Var.
Constant 0.47 0.47
 (6.80) (6.60)
No. of Adult -0.037 -0.033
 Women (-3.15) (-2.89)
No. of Adult Men 0.010870 -0.002790
 (1.20) (-0.17)
Family Income -0.0000017 0.0000000
 (-0.79) (-0.17)
Employment 0.13690 0.25233
 (1.78) (3.24)
Literacy -0.06 -0.05
 (-1.54) (-1.40)
Ext. Family -0.05 -0.03
 (Same Budgets) (-1.49) (-0.97)
Ext. Family 0.02 -0.04
 (Diff. Budgets) (-0.36) (-0.72)
Decision 0.09 0.04
by Women (1.82) (0.82)
Decision by 0.05 0.04
Husband and Wife (1.39) (1.06)
Urban -0.19 -0.17
 (5.49) (4.98)
Balochistan -0.03 -0.01
 (0.50) (0.20)
NWFP 0.01 -0.005
 (0.26) (0.10)
Punjab 0.09 0.07
 (2.37) (1.86)
No. of Obs. 372 324
[R.sup.2] 0.20 0.19

Note: Terms in parantheses are the t-statistics.

The committed categories of the dummies remain
the same as in Table 2.

Table 4
Regression Estimates for Supervision by Women
in Various Housing Activities

 Proportion of Women in
 Household Participating
 Physically Towards
Dependent
Variable Construction Repairs

Explanatory Var.

Constant 0.12 0.13
 (2.29) (1.88)
No. of Women -0.02 -0.01
 (-1.93) (-1.05)
No. of Men 0.015 -0.003
 (2.10) (-0.31)
Family Income -0.000002 -0.000004
 (-1.23) (-2.07)
Employment 0.08 0.28
 (1.37) (3.60)
Literacy -0.01 -0.02
 (-0.33) (-0.53)
Extended Family -0.06 0.01
(Same Budget) (-2.18) (0.26)
Extended Family 0.01 0.02
(Diff. Budget) (0.11) (0.32)
Decision by 0.11 0.12
Women (1.90) (2.52)
Decision by -0.02 0.04
Husband and Wife (-0.56) (0.91)
Urban -0.01 0.02
 (-0.34) (0.55)
Balochistan -0.01 -0.04
 (-0.30) (-0.55)
NWFP 0.001 0.02
 (0.04) (0.34)
Punjab 0.07 0.05
 (2.26) (1.17)
No. of Obs. 389 304
[R.sup.2] 0.09 0.11

 Proportion of Women in
 Household Participating
 Physically Towards
Dependent
Variable Improvements Additions

Explanatory Var.

Constant 0.22 0.27
 (3.24) (3.81)
No. of Women -0.03 -0.03
 (-2.86) (-3.01)
No. of Men -0.006 -0.003
 (-0.68) (-0.33)
Family Income -0.000001 -0.000001
 (-0.75) (-0.75)
Employment 0.15 -0.03
 (1.94) (-0.38)
Literacy 0.01 -0.006
 (0.27) (-0.16)
Extended Family -0.03 -0.05
(Same Budget) (-0.82) (-1.30)
Extended Family 0.06 0.003
(Diff. Budget) (1.03) (0.05)
Decision by 0.05 0.05
Women (1.12) (0.97)
Decision by 0.07 0.04
Husband and Wife (1.87) (1.05)
Urban -0.01 -0.03
 (-0.39) (-0.75)
Balochistan 0.02 -0.02
 (0.30) (-0.33)
NWFP -0.04 -0.05
 (-0.78) (-1.10)
Punjab 0.10 0.08
 (2.41) (1.97)
No. of Obs. 371 320
[R.sup.2] 0.12 0.11

Note: Terms in parantheses are the t-statistics.

The omitted categories of the dummies remain
the same as in Table 2.