Husbands’ and wives’ reports of women's decision-making power in Western Guatemala and their effects on preventive health behaviors

https://doi.org/10.1016/j.socscimed.2005.10.006Get rights and content

Abstract

Surveys have attempted to measure married women's decision-making power by asking women who has a say and/or final say in a number of household decisions. In several studies where the same questions were posed to husbands, considerable discrepancies in reports were found. This paper assesses husband and wife reports of decision-making on four matters (whether or not to buy household items; what to do if a child becomes ill; whether or not to buy medicine for a family member who is ill; what to do if a pregnant women becomes very ill) and the relationship of these reports to three recent health behaviors (having an emergency plan during pregnancy; delivering in a health facility; having a postpartum checkup within 4 weeks). A sample of 1000 women in 53 communities in three departments of western Guatemala was selected using a stratified random sampling approach. A standard household questionnaire was used to identify the respondents as well as to obtain data on household characteristics. Husbands of interviewed women were interviewed in every other household giving information on 546 couples for this analysis. Women and men's questionnaires were similar and were designed to obtain information on the respondent's knowledge, attitudes and behaviors regarding maternal health.

Consistent with other research, results show that relative to their husbands’ report, wives tend to under-report their household decision-making power. In couples with both partners educated and in couples in which women work for pay, both partners were significantly more likely to report that both of them participate in the final decisions than was the case in couples without education or in which the wife did not work for pay. Women's reports of their decision-making power was significantly related to the household having a plan for what to do in case of a maternal emergency, but was not associated with place of childbirth or with having a postpartum checkup, while husband's reports of the wife's decision-making power was negatively associated with the likelihood of having the last birth in a health facility.

Introduction

For many decades it has been known that infant and child death rates are lower among children of educated mothers virtually everywhere (e.g. Hobcraft, McDonald, & Rutstein, 1984). Fertility is similarly lower for educated and working women than for uneducated and non-working women throughout the world (e.g. see Martin, 1995, for results from 26 countries). One could consider that these early studies were using women's education and/or labor force participation as proxies for women's status. In the past two decades, the concept of women's status has been expanded from education and socio-economic status to include women's access to and control over resources and decision-making power within the household (Mason, 1986).

Power has been defined (Weber, 1978, p. 53) as “the probability that the actor within a social relationship will be in a position to carry out his [sic] own will despite resistance, regardless of the basis on which this power rests.” In a married couple and in the absence of other major actors in decision-making, (e.g. where the mother-in-law is not co-resident) power to make decisions in various domains of life may be shared, or reside more with one or the other spouse. There is some debate about the extent to which power for decision-making is zero-sum between actors or not (e.g. Mosedale, 2005). It is true that if both spouses participate in a decision, a better outcome may result than if either member alone takes the decision, simply because it is likely that more options were explored when there is joint decision-making. Conceptually, a woman's education and the socio-economic status of her family of origin are bases of her power in the household (England, 2000; Huston, 1983). Nevertheless, researchers typically have treated both women's education and her decision-making power as covariates together in statistical models predicting fertility and health outcomes. While power is manifested in relationships, empowerment refers to a process (Malhotra, Schuler, & Boender, 2002). Specifically, a recent definition of empowerment is: “The expansion in people's ability to make strategic life choices in a context where this ability was previously denied to them.” (Kabeer, 2001). Malhotra and colleagues have described the various dimensions of women's empowerment including mobility, access to and control over economic resources and domestic decision-making. The latter is the focus of the present study.

While the concept of decision-making power in a couple may be clear, its measurement in cross-sectional surveys is not. There are basically three problems. First, who is reporting, second, how joint decision-making is included and third, what domains of household decision-making are considered. Husbands and wives sometimes do not agree on who has the final say in decisions. With regard to response categories some surveys ask “who takes the final decision” and may or may not have “both husband and wife/together/the couple” as a response category. In this study, with the question “Who takes the final decision” one response category was “the couple.” With regard to domains, the areas of cooking and childcare are often seen as women's domains while large purchases and interactions with outsiders are often considered husbands’ domains so a woman's decision-making power varies with the domain.

Women's relative power as measured by these new variables has been shown in specific study settings to be a key variable for the decline of infant and child mortality (in Egypt by Kishor, 2000), for women's use of prenatal care services (in Indonesia by Beegle, Frankenberg, & Thomas, 2001) for immunization of children (in Egypt by Kishor, 2000), for seeking treatment for ill children (in Mali by Castle, 1993) and for use of modern contraception (in India by Jejeebhoy, 2002). In Guatemala, Glei, Goldman, and Rodríguez (2003) found that married rural women who reported greater household decision-making power used biomedical services during pregnancy more often than those who reported less autonomy. Also non-Spanish speakers were significantly less likely to seek care early in the pregnancy.

Following a couples’ approach which has been advocated for reproductive health generally (Becker, 1996), two recent studies have compared reports of wives and husbands on aspects of women's empowerment and the relative effects of each spouse's reports on reproductive health outcomes. Jejeebhoy (2002), with data from couples in Tamil Nadu and Uttar Pradesh, India found that husbands and wives quite often had discrepant reports of the woman's level of empowerment as measured by questions on her mobility, her access to economic resources and her economic decision-making power vis-à-vis her husband and other significant actors.1 Specifically, on the wife's involvement in three household decisions (purchase of: (a) food, (b) major household items and (c) jewelry), the spouses gave discrepant reports in 25–50% of couples depending on the specific item. Interestingly, for five of the six comparisons (two states and three decisions) when spouses disagreed, husbands were twice as likely as their wives were to report that the wife was involved in the decision (Table 4b, p. 304). Jejeebhoy then considered four outcome variables: current contraceptive use, interspousal discussion of family planning, unmet need for contraception and childbearing in the past five years. In Tamil Nadu, in logistic regressions for each outcome, when there were significant effects they were always for women's reports of her decision-making power, i.e. the coefficients of husbands’ reports of her decision-making power were not significant. However, in Uttar Pradesh, the coefficients of husbands’ reports of women's power were significant in several instances while the coefficients for the variable for wives’ reports were not. Uttar Pradesh is known to be a more gender conservative context, while Tamil Nadu is more egalitarian (Mason & Smith, 2000), providing a possible explanation of the greater influence of women's reports of their own power in the latter.

Ghuman, Lee, and Smith (2004) used data from the same questionnaire from surveys carried out in selected areas in four other countries—Pakistan, Thailand, Philippines, Malaysia—as well as India. They utilized a model for item response to compare husband and wife reports to examine systematic differences contrasted with random error. They found both systematic and random components to be present and concluded that men and women have different cognitive understanding of the questions. For example, with respect to whether the wife has to ask permission to go to certain places, husbands and wives may have different cognitive thresholds to give the answer yes, so one spouse may only answer yes if permission must be asked each time while the other may answer yes to affirm a general code of expected behavior. The authors also studied the relationship of the husband's and wife's reports of empowerment with experience of child death in the couple. Effects in opposite directions were found—women's reports of their power (on a scale from low to high levels of decision-making power) were negatively associated with mortality but husbands’ reports of their wives’ power were positively associated with mortality and significantly so in India for one indicator of power—whether it is the wife who decides on discipline for the children or not.

Given these results, it can be argued that it is also important to ask husbands questions about women's decision-making power, at least in certain settings and if the interest is in reproductive health outcomes. While this is consistent with a couple-approach to reproductive health, it would entail considerable increases in survey costs.

The present analyses with survey data from Guatemala has two main objectives: first to examine spousal reports of women's decision-making power in a Latin American context and second to determine whether wives’ or husbands’ reports of this decision-making power have a stronger relationship with preventive health behaviors. We also consider determinants of the reported decision-making power.

Section snippets

Methods

Data for this study were collected as part of a household survey in 2003 by the Guatemalan Maternal and Neonatal Health (MNH) Program. The survey covered 53 communities in 19 districts in three departments (Quiché, Sololá and San Marcos) in the western region of Guatemala. The main purpose of the survey was to obtain information to measure the impact of the MNH Program interventions in Guatemala. A sample size of 1000 women who were pregnant or delivered in the last year was deemed sufficient

Results

Table 1 gives selected univariate statistics of the covariates and the outcome variables. Women in the sample had an average of 3.4 children, the mean age was 27 years, 53% had a Mayan mother tongue, 22% worked for pay and 35% had no formal schooling. Regarding outcome variables, for those with a recent birth 38% had developed a plan in case of emergency during pregnancy, delivery and the postpartum period, 27% had delivered in a health facility and 37% had a postpartum checkup within 4 weeks

Discussion/conclusions

The level of women's decision-making power in the household is one indicator of her empowerment. Consistent with other research we found that relative to their husbands’ reports, wives tend to under-report their household decision-making power. Since the truth is obviously not known, this difference could alternatively be due to husbands’ over-reporting of the decision-making power of their wives. Alternatively, as Ghuman et al. (2004) suggest from their analyses of Asian data from couples, it

Acknowledgments

Funds for the program and survey in Guatemala were provided by the United States Agency for International Development. The survey was carried out by GSD Consultores group in Guatemala. Gwendolyn Berger assisted with data processing. Jane Bertrand provided comments on an earlier draft.

References (30)

  • F. Fonseca-Becker et al.

    Case study report: Impact evaluation, community mobilization and behavior change, maternal and neonatal health program, ministry of health and public assistance of Guatemala

    (2004)
  • F. Fonseca-Becker et al.

    Measuring the effects of behavior change and service delivery interventions in Guatemala with population-based survey results

    (2004)
  • Ghuman, S. J., Lee, K. J., & Smith, H. L. (2004). Measurement of women's autonomy according to women and their...
  • Guatemalan Ministry of Health (2003). Lineamientos Estratégicos para Reducir la Mortalidad Materna. Ministerio de Salud...
  • J. Hobcraft et al.

    Socio-economic factors in infant and child mortality: A cross-national comparison

    Population Studies

    (1984)
  • Cited by (0)

    View full text