Spouses' Views of Contraception in the Philippines

Ann Biddlecom John B. Casterline, Ohio State University Aurora E. Perez

First published online:

Abstract / Summary

Data from a 1993 survey in the Philippines indicate that, in the aggregate, men and women have similar views of contraception. For example, 72% of husbands and 77% of wives strongly approved of contraception, and at least half believed that relatives and friends approved (although men were less likely than women to think so). At the couple level, however, men’s perceptions about contraception often differ from those of their wives. A relatively large amount of disagreement exists about the importance of certain contraceptive attributes and the extent to which these attributes apply to specific methods. This disagreement is associated both with lower levels of contraceptive use and with greater conflict over intentions to use contraceptives in the future. For example, when both spouses approve of family planning in general, 81% of couples share the same intentions to practice contraception in the future; but among couples who disagree over approval of contraception, just 43% share intentions about future use.

(International Family Planning Perspectives, 23: 108–115, 1997)

The relative importance of demand and supply factors in determining contraceptive behavior has been a dominant issue in international family planning research over the past three decades. "Demand" refers to the motivation to space or limit births (determined by an array of economic, social and cultural factors), while "supply" refers to the accessibility and quality of family planning services. Few analysts would argue that one or the other exclusively determines contraceptive practice; rather, the debate has centered on the relative weight of each.

In the widely-used Easterlin Synthesis Framework, which acknowledges the role of both factors, contraceptive behavior is jointly determined by the motivation to practice contraception versus the costs of contraception.1 Such costs encompass much more than the accessibility of family planning services; they include all social, psychological and cultural factors that act as barriers to contraceptive practice among men and women motivated to practice contraception.2 This expanded definition of contraceptive costs is consistent with the growing recognition that unmet need for family planning services cannot be attributed solely to inadequate access in many settings.3

Despite the general consensus that contraceptive decision-making entails weighing motivation against costs, to date relatively little empirical research has attempted to operationalize these concepts because of the difficulty of measuring the full array of costs. Assessing social, psychological and cultural barriers to contraceptive use is challenging, especially through conventional social surveys. At a minimum, qualitative interviews or structured survey interviews about individuals’ views of contraception are required to uncover the perceptions of costs and benefits of contraceptive use and of the desirability of features of available methods.

A shift toward greater gender balance in contraceptive research has accompanied this more complete view of the determinants of contraceptive use. Increasingly, interviews with both men and women (sometimes, but not always, matched partners) have become standard research practice in developing countries; thus, there are now far more opportunities to examine male-female perspectives toward reproductive attitudes and behaviors.4 In fact, more than 30 surveys (mostly in Sub-Saharan Africa) conducted as part of the Demographic and Health Survey program collected information from men about fertility and contraception. This development also fits the general policy trend toward encouraging men to share more equitably with women the responsibilities of fertility regulation and parenting.5

However, the analyses of available developing-country data from men and women have been seriously imbalanced in their treatment of demand and supply factors, focusing primarily on fertility preferences—especially on whether men agree with their partners. Among numerous recent publications on men’s fertility preferences,6 virtually none examine men’s views of contraception broadly and systematically. Many studies consider men’s overall approval of family planning, fewer examine men’s attitudes toward specific male methods, such as the condom or vasectomy, and men’s views of the full range of available methods have been largely ignored.7 Moreover, the limited literature on the costs of contraception is based primarily on interviews with women.8

The implicit message is that men figure in reproductive decision-making primarily because of their fertility preferences, so only their degree of agreement with women’s preferences truly matters. Surprisingly, however, numerous studies have shown that in many settings, men as a group are not more pronatalist than women,9 and when the attitudes of spouses are directly compared, husbands are not consistently more pronatalist than their wives.10

Some might argue that since contraception is almost universally regarded as women’s responsibility, men do not inform themselves about contraception, because they do not want to be distracted from their own responsibilities or because to do so would violate gender norms. This argument presumes a gender divide in reproductive behavior so radical we doubt it obtains in most societies. For example, recent research in Zimbabwe showed that a large proportion of men felt they had a major say in contraceptive decision-making, and exposure to a multimedia communication campaign to promote family planning even increased their likelihood of stating that the husband alone should decide.11 The argument that contraception is women’s responsibility also ignores men’s documented involvement in the many societies where male methods and coitus-dependent methods are widely used.12

In short, we see no reason to assume a priori that the costs of contraception are so much smaller for men than for women for men to be ignored in efforts to understand determinants of contraceptive use. Thus, we perceive three major research challenges—to measure the costs of contraception from both male and female per- spectives; to determine to what degree men’s and women’s views are similar; and to assess how these views of costs affect contraceptive behavior.

Analytic Issues

The degree to which men are actually involved in decisions about contraceptive use is an empirical question. We hypothesize that men are significantly involved more often than is assumed, and that they have many views—some vague, some highly elaborate—about contraception. This perspective is based in part on our fieldwork in the Philippines, where we conducted in-depth interviews with women and men.*

A number of issues must be addressed when structuring an analysis of men’s views of contraception. Even when the principal aim is to analyze men’s views, women’s views must also be examined, for two reasons: First, women’s views can provide a standard against which men’s approval of certain methods or their fear of side effects can be gauged. Second, the most appropriate theoretical models for contraceptive decision-making are dyadic.13 Both women and men need to be jointly analyzed in the empirical testing of these models, with the core attitudinal and behavioral variables adequately represented. In most settings, the attitudinal variables (including views of contraception) cannot be measured accurately by proxy.14 From this perspective, understanding men’s views of contraception is not an end in itself, but is one step toward filling in the elements of a larger causal structure that explains how partners decide on their contraceptive practice.

Comparisons by gender can be performed at either the aggregate or the micro level. Aggregate-level analyses compare averages and distributions calculated for men and women separately, and from the overall perspective of all men in a given sample, allow conclusions to be drawn about whether programmatic efforts should include distinctive or intensive approaches oward men (for example, in information, education and communication campaigns or in designing service provision). Microlevel analyses make comparisons at the couple level. Arithmetically, the degree of aggregate-level agreement must match or exceed that within couples; conversely, a substantial amount of disagreement between partners may coexist with similar aggregate profiles for men and women. Analyses of gender agreement typically focus on the extent of agreement between partners. When partners disagree, contraceptive decision-making may be more difficult, and making contraceptive choices that do not leave one partner ’s aspirations frustrated becomes increasingly hard.

A focus on spousal agreement leads naturally to questions about what factors determine the degree of agreement and to what extent disagreement affects contraceptive practice. Several studies have shown that the degree of agreement varies according to characteristics of spouses.15 For example, the more unequal spouses are (in terms of age or decision-making authority), the more likely they are to differ in their reports of contraceptive use and fertility preferences.16 Another common finding is that, the more often husbands and wives discuss family planning and fertility preferences, the more they share similar views on those topics. While this association in itself need not affect contraceptive practice, the amount of communication between partners is positively associated with contraceptive use.17 Although the latter may be less decisive in determining contraceptive behavior than the relative power of each spouse in the decision-making process, both couple communication and differential power influence family planning and fertility-related outcomes.18

Furthermore, gender agreement about contraception cannot be assumed to be uniform across contraceptive methods. Men’s and women’s views about contraception may be conditioned by the degree to which obtaining and using a male or female method is intrinsically one partner ’s responsibility, by whether a method is coitus-dependent and requires the cooperation of both partners, and by the degree to which one partner is at greater risk of being directly affected by the method’s attributes.

Because we believe that contraceptive costs are not necessarily smaller for men than for women and that men cannot be ignored in studying the determinants of contraceptive behavior, in this article we measure the costs of contraception from both male and female perspectives. We thus determine the degree to which men’s and women’s views converge, and how these views of costs in turn affect contraceptive practice.

Data and Methods

Our empirical analysis of men’s and women’s views of contraception is taken from a 1993 survey on attitudes toward contraception in the Philippines. This survey contained an exceptionally detailed inquiry into contraceptive use and its perceived attributes. We present data on contraceptive attributes from Filipino men to examine whether their views of contraception are well developed, and to determine the degree to which their views coincide with or differ from their wife’s views.

The Philippines is an appropriate setting to examine men’s perceptions of contraception because a wide range of methods have been available over the past two decades.19 (According to the 1993 survey, 88% of men and 96% of women were aware of five or more methods, and less than 5% knew of fewer than three.)

The survey data were collected from May through September 1993, in a collaborative effort between the Population Institute of the University of the Philippines in Manila, and Brown University, Providence, Rhode Island, USA. Probability samples of currently married women aged 25–44 and their husbands were selected at the barangay level, the smallest political subdivision in the Philippines. The overall nonresponse rate was 32%, with approximately one-third of non-response attributable to respondents’ refusal to be interviewed.

The husbands and wives in 780 matched couples were interviewed independently —480 couples from eight rural barangays in the municipality of Munoz, in Nueva Ecija province (roughly a four-hour drive from the outskirts of Manila) and 300 couples from five urban barangays in Manila. In addition, 26 respondents were interviewed in-depth about their reproductive experiences, attitudes toward contraception and reasons for use or nonuse. The interviews were conducted in Tagalog; for all interviews, respondents and interviewers were matched by sex.

We selected the survey sites to yield a sample of rural and urban respondents whose circumstances (ecological and socioeconomic) would roughly resemble those of a majority of Filipinos. However, the data are not nationally representative in terms of contraceptive prevalence. For example, 69% of married women aged 25–44 in our survey were currently using a method, compared with 44% in the 1993 National Demographic Survey.20 Most of the difference is attributable to higher rates of female sterilization and withdrawal in our study sample, especially in the rural barangays. However, the method mix among both samples was similar, with female sterilization being the most commonly used method, followed by the pill, withdrawal and natural family planning (rhythm).

At the time the research sites were selected, no current data were available on contraceptive use by rural or urban area. We selected Nueva Ecija based on its distance from Manila, its relatively large population and its relatively high proportion of rural residents. Our data also showed that 17% of couples in the sample still had an unmet need for family planning,21 which provides further evidence of the importance of examining the reasons for nonuse, including the degree of agreement between men’s and women’s perceptions of contraception.

The survey instrument asked both spouses how much they and their friends and relatives approved of contraception. Respondents were asked, "In general, do you approve of couples using ways to avoid getting pregnant, or do you disapprove of couples using ways to avoid getting pregnant?" If participants either approved or disapproved, they were then asked how much (strongly or somewhat). Each spouse was also asked to identify those methods that fit any of the following seven negatively phrased attributes—the method was ineffective, the respondent’s spouse disapproved, the method caused side effects, it was painful or unpleasant to use, it was difficult to obtain, it cost too much or its use went against the respondent’s religion. This information was solicited through a large method-by-attribute matrix of items; ratings for these seven attributes were obtained for as many as eight methods per respondent.

The list had been pared down to seven attributes with difficulty, and some of the attributes required more than one item to explore adequately. Filipino couples also have a relatively large number of methods to choose from—at least six. This availability implies a method-by-attribute matrix of questionnaire items consisting of 50 or more cells, which is too large a number to function well in the field.

Our solution was not to ask all respondents explicitly about whether every attribute applied to each method, but to proceed down the list of attributes and ask the respondent to name those methods to which the attribute applied. For example, they were asked whether contraceptive methods caused health side effects and, if so, which methods did so. Respondents could name no method or they could name as many as eight methods. Further questions were then asked about the nature of the side effects.

This design expedited progress through the matrix, and undoubtedly was much less tedious than asking about every attribute for six or more methods. At the same time, the design placed the burden of identifying methods that possess certain attributes on the respondent. In addition, since the attributes themselves are characterized in the negative (bad effects on health, expensive, against religion and so forth), the results may reflect an unduly negative view of contraception, as well as an underenumeration of those methods that a respondent actually perceives as possessing negative features. Finally, respondents were asked to rate the relative importance they attached to each of these attributes in deciding whether to use a contraceptive method.

Because using an identical measurement approach for men and women was a guiding principle in the data collection, the method-by-attribute matrix of items, the overall content, the ordering of items and the wording of questions were virtually identical in the male and female questionnaires. All respondents were also asked about their intentions to practice contraception in the future, and current users were asked whether they intended to continue with their method. (This question was not asked of surgically sterilized respondents or of infecund or menopausal women.)

In discussing the results, we compare the data at the aggregate level (among all men and all women interviewed) and at the couple level (among individual husbands and wives). As mentioned earlier, aggregate-level agreement must match or exceed couple-level agreement. Our research considers the extent to which disagreement within individual couples exceeds that between men and women overall. The comparisons of individual couples also allow us to explore the nature of that disagreement—i.e., whether it is completely symmetrical or whether results from one spouse are more positive or negative than results from the other.

We use two measures of agreement—a crude level of consensus (the proportion of couples in which both spouses share the same view) and the Kappa index,22 which is the degree of agreement net of that expected by chance alone. In analyzing ordinal variables, we employ the weighted Kappa index, in which pairs of responses farther apart on the ordinal scale indicate greater disagreement than responses closer together. Kappa values range from 0.0 to 1.0, with 0.0 indicating agreement no greater than that expected by chance alone and 1.0 indicating perfect agreement.§ Thus, the smaller the Kappa is, the more the views of husband and wife disagree; moreover, a statistically significant Kappa value indicates that spouses agree to a greater extent than would be expected by chance.

Results

We first consider husbands’ and wives’ overall approval of contraceptive use. Individual respondents were asked whether they approved of family planning and whether they thought their relatives and friends approved of it (Table 1). In the aggregate, approval of contraception is high for both partners: Roughly three-quarters of men and women strongly approved of contraception, although men were slightly less approving than women (p ≤ .05). Men were also less likely than women (p ≤ .05) to perceive that their relatives (58% vs. 67%) or their friends (50% vs. 70%) strongly approved.

From a couple-level perspective, however, there appears to be much less agreement between spouses: One-half to two-thirds of couples shared the same view. The Kappa statistics are generally low, although they differ significantly from zero, indicating greater agreement than would be expected by chance alone. Approximately one-third of spouses did not accurately perceive their partner ’s approval (not shown).

Disagreement at the couple level is not symmetrical: Among the couples who did not share the same view on their friends’ approval of family planning, two-thirds of husbands perceived friends as less ap- proving than did their wives. This tendency is more pronounced the more distant the friendship. However, husbands and wives may simply be referring to different friends, so that spousal disagreement reflects differences in opinions held by members of each spouse’s social network. Thus, in this setting, husbands’ social networks may be characterized by more negative and less supportive views of family planning than wives’ networks.

Table 2 presents the results from the item gauging the importance of specific attributes in choosing a contraceptive method; for 10 specific attributes, respondents were given the choice of describing each as "very important" "somewhat important" or "not important." At the aggregate level, the highest proportion of both men and women (76–77%) considered effectiveness in preventing pregnancy to be "very important," and descending proportions viewed spousal approval, health-related side effects and effect on the marriage as "very important." The approval of friends 111 Volume 23, Number 3, September 1997 and of relatives was labeled "very important" by comparatively few men and women.

In the aggregate, the importance of these attributes is in essentially the same order among men and women, although women tended to view social attributes (spousal approval, the effect of contraceptive use on marriage and the approval of others) and accessibility of supplies and services as very important more often than did men.

Much less agreement is apparent, however, at the couple level. As the far right column of Table 2 shows, no more than two-thirds of couples shared the same view on the importance of a specific attribute. The percentages who agreed in ranking a specific attribute as "very important" range from 31% (for financial cost) to 65% (for effectiveness). The Kappa values indicate that almost all of this apparent spousal agreement can be attributed to chance alone (with the exception of agreement on the importance of religious acceptance of contraception and on the approval of friends). This suggests that the decision-making process about whether and how to practice contraception may differ between husbands and wives in ways that complicate and perhaps even impede their final decision.

Since the matrix of views on the negative attributes of specific methods yielded an enormous amount of data, we calculated a summary score by counting the number of negative attributes that respondents assigned to each method.** Respondents rarely identified any method with more than two of the seven negative attributes listed, so we collapsed the count into a three-level ordinal scale: none, one, or two or more negative attributes.

In the aggregate, husbands and wives held differing views, with women consistently rating each method (with the ex- ception of tubal ligation) more negatively than did men (Table 3, page 112). Both men and women viewed the pill and the IUD more negatively than they perceived withdrawal and rhythm. (The prevalence of these latter two methods is relatively high in the Philippines.23 ) Contrary to expectations, the gender differentials in methods’ negative attributes did not vary by type of method (i.e., male-oriented vs. female-oriented or coitus-dependent vs. coitus-in- dependent).

The gender differences are much larger when spouses were compared. We observed moderate-to-low levels of agreement between husbands and wives on negative attributes, with the proportions in crude agreement ranging from 36% to 51%. Although they are small, the Kappa values shown in Table 3 indicate that for four methods—the pill, the IUD, the condom and rhythm—agreement exceeded what would be expected by chance alone. The analysis of spousal agreement about approval (Table 1) showed that when spouses disagreed (especially about relatives’ and friends’ approval), husbands were less approving; the method-specific analysis of attributes, in contrast, shows that wives were more negative when spouses did not concur.

Moreover, wives were more likely than husbands to identify negative attributes of contraceptives in both the aggregate-level and the couple-level analyses. Why might this be so? One explanation is that women necessarily are more exposed to, and have more experience with, a wider range of methods, and are also more at risk of suffering negative consequences of use. However, wives were more negative than their husbands even about methods that directly involve men. A more likely explanation is that women are better informed about specific methods and thus hold stronger opinions about them (both more negative and more positive).

A final potential explanation is that women in general are more comfortable talking about contraception than are men, which would affect the likelihood that they would voluntarily name one or more methods when asked about a certain attribute. Here both respondents’ and the interviewers’ gender must be considered, since the sex of the interviewer may be more critical than the sex of the respondent. Because respondents and interviewers were matched by sex in this study, we cannot untangle these effects.

We also examined differentials in the degree of agreement according to two characteristics of the couple’s relationship—communication between spouses about family planning and sex, and the husband’s role in household decision-making.†† We expected spousal concurrence about contraception to be greater in marriages where spouses discussed family planning and sex more frequently and were more comfortable with such discussions, and where decision-making was more egalitarian.

On balance, the data do not confirm these hypotheses (Table 4); the degree of agreement at the couple level did not vary markedly according to couples’ characteristics. For example, there was no strong and consistent relationship between the frequency of spousal discussion about family planning and agreement on overall approval of contraception or on method attributes. In fact, one unexpected finding was that the less discussion there was, the more spouses agreed on the negative attributes of tubal ligation, one of the most commonly used modern methods in the Philippines.

In addressing a similar finding of the negative association of frequency of discussion and spousal agreement on the number of desired children, the authors of a Taiwanese study suggested that basic agreement may mean that there is less need to formally discuss an issue, which simply does not have to be verbalized.24 Alternatively, discussions may crystallize differences rather than reconcile them. This confusion suggests that we can gain only limited insight about spousal communication when the frequency of conversation is measured but nothing is known about the content or nature of the discussions (e.g. who initiates them, the degree of reciprocity and so forth).

Table 4 also shows somewhat greater agreement among couples who have the same comfort level in discussing family planning; again, though, the differential is small. Moreover, spousal agreement tends to be slightly lower in those marriages where the husband has all of the say in household decision-making (which is consistent with the converse assumption, that agreement is higher in more egalitarian marriages).

Although empirical evidence demonstrates that men’s fertility preferences— and, in particular, couple disagreement about preferences—affect both the decision to use contraceptives and the meth- od chosen, only limited empirical research examines the relationship between men’s views of contraception (and couples’ disagreement about contraceptive costs) and contraceptive behavior. Most of that research is based on cross-sectional data, which makes it difficult to determine whether views of contraception affect contraceptive behavior or vice versa. Our data from the Philippines, unfortunately, are also cross-sectional in design. However, we briefly consider here the relationship between spousal agreement and the intention to practice contraception (either to start to use a method or continue using one) in the future.

While spouses largely agree on intentions to practice contraception in the fu- ture (Table 5), those who disagree in their views of contraception agree less often about future use. For example, when both spouses approve of contraceptive use in general, 81% of couples concur in their intention to practice contraception in the future. In contrast, when only one spouse approves of contraception, the proportion of couples who concur in their intention to use a method in the future falls to 43%. Among these couples, 36% had opposing intentions and 21% concurred in their intention not to use a method in the future.

The same variation in intention by consensus is apparent for the specific methods examined in the remainder of the table, although the association between views of methods and intentions to use one is statistically significant only for tubal ligation (p ≤ .05) and is marginally significant for the pill and rythmn (p ≤ .10). For example, for tubal ligation, couples who agreed that it had relatively few negative attributes were more likely to agree that they intended to use a contraceptive in the future than were those who disagreed about the method’s negative attributes (84% versus 70%); and those who disagreed on a method’s negative attributes were more likely to disagree on future use (23% vs. 11%).

In a separate analysis of the wife’s current contraceptive use (not shown), we find a similar relationship between spousal agreement on perceived contraceptive costs and on use; this finding suggests, but does not prove, that views of contraception have an impact on contraceptive behavior: When both spouses approve of contraception, 78% are current users. However, when spouses do not share in their approval, only 58% use a method.

Approval of contraception is high in the aggregate in the Philippines, however, and even where there is disagreement among spouses, more than half of all couples practice contraception. In settings where general approval is lower (and especially where modern contraceptive prevalence is relatively low), differing views between spouses are more likely to stymie contraceptive use.

Discussion and Conclusions

The answer to the first overall question addressed in this analysis ("Do men have well-developed views of contraception?") is an unqualified "yes." The data reveal that the Filipino men who were surveyed have extensive perceptions about contraception and the attributes of specific meth- ods, even in a setting where two of the most commonly used methods—tubal ligation and the pill—are female-controlled.

The answer to the second question ("Do men and women share similar perceptions of contraception?") is a qualified "yes." Overall, Filipino men and women hold fairly similar views about the social acceptability of using contraceptives, about the relative importance of an array of negative attributes of contraceptives and about many features of specific methods. There are only a few differences of note. First, men tend to perceive significant others (relatives and friends) as being less approving of contraception than do women. Second, women tend to rate methods more negatively than do men. Thus, there is little evidence that programs aimed at Filipino men need to resort to radically different approaches from those strategies used to target women.

However, at the couple level, there is less similarity in men’s and women’s views of contraception. Crude levels of agreement—the percentage of spouses who agree—are largely below 50%, and analysis using the Kappa index indicates that most of this agreement is due to chance alone. This result might be attributed to a lack of communication between spouses, as few couples discuss family planning often (just 9% of men and 14% of women who provided data on frequency of such discussions). Yet given the relatively high levels of contraceptive use in this setting, one could argue that spousal communication about family planning does not matter so much as each spouse’s openness to modern methods.25

From a policy standpoint, it makes little difference whether agreement occurs by chance alone or is due to some other dynamic (assortative mating or spousal communication); the critical issue is whether or not spouses hold the same views. If not, program workers must be prepared to deal routinely with spouses who do not concur in their attitudes and preferences. In the Philippines, levels of disagreement are too high for spousal disagreement to be ignored, either in research on contraceptive use or in the design of interventions, especially those dealing with problems of meth- od choice and discontinuation of use. Moreover, the high degree of disagreement in perceptions about basic aspects of contraception and in features of specific contraceptive methods may have some bearing on contraceptive behavior (as suggested by its association with intentions to practice contraception in the future).

Men’s views of contraception deserve much more attention than they have received heretofore. Rarely have men been asked in detail about their views on this subject, and hence there is little empirical basis for determining whether men have well-developed views about contraception and what effects those views have on contraceptive practice. Except where contraception is unknown or not practiced, it should be assumed that men are aware of it and hold attitudes about it. The Filipino data presented here confirm the validity of this assertion. Thus, men should be asked their views about contraception—whether they think that modern method use is immoral or socially unacceptable, whether they fear side effects for themselves or their wives and whether they are concerned that contraceptive use might detract from their sexual pleasure, among other views.

One result of the 1994 International Conference on Population and Development was widespread agreement that men should assume more responsibility for family planning and family support.26 This stance, combined with increased programmatic attention to spousal communication, makes further inquiry into the nature of men’s and women’s reproductive attitudes a high priority. Indeed, policy and programmatic actions may serve to strengthen the contribution of men’s views to the determination of reproductive outcomes.

The fundamental reason for interest in men’s and women’s views about the costs of contraception is the likelihood that these views play a significant role in decisions about contraception and related reproductive behavior. In our qualitative research in the Philippines, a variety of decision-making styles prevailed, so it would be an oversimplification to speak about one style among Filipino couples. Furthermore, Filipino men cannot be assumed to usually have the upper hand in the decision to use contraceptives, as might be the case in strongly patriarchal societies. A further complication is that merely discussing family planning must be distinguished from taking action and assuming responsibility to practice contraception.

Excerpts from in-depth interviews for this study in the Philippines reveal a subtle and multifaceted decision-making process.27 Although discussion and communication on these subjects take place, they are partial and limited. Filipino wives assume responsibility for obtaining contraceptive methods, although the process can differ sharply across couples; sometimes the wife chooses the contraceptive on her own, and at other times the husband appears to dominate the decision-making (at least in his version of events). The complexity of spousal decision-making styles in the Philippines makes an overall assessment of the relative significance of husbands’ and wives’ views of contraception a daunting task.

In evaluating the impact of men’s views of contraception on contraceptive practice, focusing on unmet need is valuable. Not only is unmet need a priority for policies and programs in many developing countries, but cases where women’s preferences conflict with their behavior might be especially revealing of the determining power of men’s views. An analysis of the factors underlying unmet need using the same Philippine survey data set revealed that nearly 20% of that need can be attributed to husbands’ negative perceptions of contraception and its attributes.28

Thus, we can conclude that men’s views need to be taken seriously in the development of interventions to reduce unmet need in the Philippines. If family planning and reproductive health programs are to go beyond the goal of satisfying unmet need and are to nurture a decision-making process that involves and respects the interests of both partners, these programs must be informed by more accurate and complete measurement of men’s (as well as women’s) views of contraception than has been the standard in fertility and family planning research during the past two decades.

Our distinction between aggregate-level and couple-level perspectives has important implications for research design and the development of effective programs. Research on an "aggregate-level" perspective can approach men and women as separate, although obviously related, target populations. Alternatively, men and women can be approached as dyadic pairs, in which the basic and most meaningful comparisons are between matched spouses. The distinction is important, because power imbalances (usually favoring men) characterize most marriages, and this tends to exaggerate the ultimate impact of spousal disagreement. The husband’s opposition to contraception may be sufficient to block contraceptive use in many cases, but the converse will occur much less often. This asymmetry means that when spouses disagree, women’s family planning aspirations will more often be frustrated than men’s.

Although this specific situation may typify a relatively small minority of women, unmet need for contraception ranges from 20% to 40% of women in most societies. The disagreement over the costs of contraception that characterizes a small proportion of couples overall can translate into a substantially larger fraction of the couples who are a priority interest for reproductive health and family planning policy and programs. As we learn more about men’s perceptions of contraception, we will simultaneously learn more about the nature of reproductive decision-making among both men and women.

Footnotes

*For excerpts from the unstructured in-depth interviews, see: A. E. Biddlecom, J. B. Casterline and A. E. Perez, Men’s and Women’s Views of Contraception , Research Division Working Paper No. 92, Population Council, New York, 1996.

Separate interviews with husbands and wives were generally conducted simultaneously (90% overall). In rural Nueva Ecija, nearly all couples were interviewed simultaneously, because both husband and wife returned home from work during the hot midday hours. In metropolitan Manila, however, simultaneous interviews were more difficult to achieve, because men’s and women’s work schedules differed. Overall, for 74% of interviews, spouses were physically removed from each other ’s presence (interviewers recorded who was present at different points of the interviews with women). The separate interview rate was 62% in Metropolitan Manila, where smaller dwellings and less associated outdoor space made separate interviews more difficult to conduct.

While a contraceptive prevalence rate of 69% among 25–44-year-olds in union is high by Philippine standards, it is typical of that in other Southeast Asian populations, such as Thailand. (See: Levels and Trends of Contraceptive Use as Assessed in 1994 , United Nations, New York, 1996.)

§Although Kappa values smaller than 0.0 (indicating less agreement than expected by chance alone) are arithmetically possible, Kappas are rarely negative and are almost always close to zero.

**Although the aggregate profiles were relatively similar by sex, women were more likely to assign a negative attribute to a method than were men, with the exception of the accessibility of supplies and services. In the couple-level analysis, method-specific agreement on attributes was high, especially for efficacy and side effects. Typically, 70% or more of couples agreed that a method suffered from a specific negative feature; this level of agreement was higher for modern methods (i.e., the pill and the IUD) than for traditional methods (i.e., withdrawal). Analysis using the Kappa index, however, indicated that most of this agreement at the couple level could be attributed to chance.

††Because our emphasis in this article is on men’s views, we use the husband’s reported frequency of discussion about family planning and his reported role in household decision-making. There were no dramatically different patterns, however, when wife’s views were examined. We constructed a summary index of the husband’s influence in family and household decision-making for six major domains: purchase of major appliances; division of household chores; choice of place of residence; disciplining of the children; choice of workplace for the wife (within or outside of the home); and changes in the husband’s job. Response categories were no influence, some input, equal input with wife, primary influence and total control. The index is a simple summation of the responses on the six items, relatively evenly distributed over four categories: Scores of 5–17 were considered to indicate that the husband had less influence than the wife; a score of 18 meant that he had equal influence to that of his wife (many respondents reported “equal input” in all six domains of decision-making); a score of 19–21 indicated that the husband had more influence than his wife; and a score of 22 or higher was interpreted as meaning that the husband had a dominant influence.

References

1. R. A. Easterlin and E. M. Crimmins, The Fertility Revolution: A Supply-Demand Analysis, University of Chicago Press, Chicago, Ill., USA, 1985.

2. R. A. Bulatao and R. D. Lee, eds., Determinants of Fertility in Developing Countries: Fertility Regulation and Institutional Influences, Vol. 2, Academic Press, New York, 1983; and M. Nag “Some Cultural Factors Affecting Costs of Fertility Regulation,” Population Bulletin, No. 17, United Nations, New York, 1984.

3. J. Bongaarts and J. Bruce, “The Causes of Unmet Need for Contraception and the Social Content of Services,” Studies in Family Planning, 26: 57–75, 1995.

4. S. Becker, “Couples and Reproductive Health: A Review of Couple Studies,” Studies in Family Planning, 27: 291–306, 1996.

5. J. Bongaarts and J. Bruce, 1995, op. cit. (see reference 3); and United Nations, “Program of Action of the 1994 International Conference on Population and Development (Chapters I–VIII),” Population and Development Review, 21: 187–213, 1995.

6. A. Bankole, ”Desired Fertility and Fertility Behaviour Among the Yoruba of Nigeria: A Study of Couple Preferences and Subsequent Fertility,” Population Studies, 49: 317–328, 1995; F. N.-A. Dodoo and A. Seal, “Explaining Spousal Differences in Reproductive Preferences: A Gender Inequality Approach,” Population and Environment, 15: 379–394, 1994; A. C. Ezeh, M. Seroussi and H. Raggers, Men’s Fertility, Contraceptive Use, and Reproductive Preferences, Demographic and Health Surveys (DHS) Comparative Studies No. 18, Macro International, Calverton, Md., USA, 1996; S. J. Jejeebhoy and S. Kulkarni, “Reproductive Motivation: A Comparison of Wives and Husbands in Maharashtra, India,” Studies in Family Planning, 20: 264–272, 1989; and K. O. Mason and A. M. Taj, “Differences Between Women’s and Men’s Reproductive Goals in Developing Countries,” Population and Development Review, 13: 611–638, 1987.

7. D. J. Adamchak, M. T. Mbizvo, and M. Tawanda, “Male Knowledge of and Attitudes Towards AIDS in Zimbabwe,” AIDS, 4: 245–250, 1990; J. T. Bertrand et al., “Attitudes Toward Voluntary Surgical Contraception in Four Districts in Kenya,” Studies in Family Planning, 20: 281–288, 1989; A. C. Ezeh, “The Inf fluence of Spouses Over Each Other’s Contraceptive Attitudes in Ghana,” Studies in Family Planning, 24: 163–174, 1993; M. Khalifa, “Attitudes of Urban Sudanese Men Towards Family Planning,” Studies in Family Planning, 19: 236–243, 1988; T. A. McGinn, A. Bamba and M. Balma, “Male Knowledge, Use and Attitudes Regarding Family Planning in Burkina Faso,” International Family Planning Perspectives, 15: 84–87, 1989; and K. Ringheim, “Factors That Determine Prevalence of Use of Contraceptive Methods for Men,” Studies in Family Planning, 24: 87–99, 1993.

8. M. Nag, 1984, op. cit. (see reference 2).

9. A. C. Ezeh, M. Seroussi and H. Raggers, 1996, op. cit. (see reference 6); and K. O. Mason and A. M. Taj, 1987, op. cit. (see reference 6).

10. A. Bankole, 1995, op. cit. (see reference 6); and S. J. Jejeebhoy and S. Kulkarni, 1989, op. cit. (see reference 6).

11. M. T. Mbizvo and D. J. Adamchak, “Family Planning Knowledge, Attitudes, and Practices of Men in Zimbabwe,” Studies in Family Planning, 22: 31–38, 1991; and P. T. Piotrow et al., “Changing Men’s Attitudes and Be- havior: The Zimbabwe Male Motivation Project,” Studies in Family Planning, 23: 365–375, 1992.

12. S. L. Curtis and K. Neitzel, Contraceptive Knowledge, Use, and Sources, DHS Comparative Studies No. 19, Macro International, Calverton, Md., USA, 1996.

13. L. J. Beckman et al., “A Theoretical Analysis of Antecedents of Young Couples’ Fertility Decisions and Outcomes,” Demography, 20: 519–533, 1983; and E. Thomson, “Dyadic Models of Contraceptive Choice, 1957 and 1975,” in D. Brinberg and J. J. Jaccard, eds., Dyadic Decision Making, Springer-Verlag, New York, 1989, pp. 268–285.

14. S. Becker, 1996, op. cit. (see reference 4).

15. L. J. Beckman et al., 1983, op. cit. (see reference 13); and P. Hollerbach, “Fertility Decision-Making Processes: A Critical Essay,” in R. A. Bulatao and R. D. Lee, eds., Determinants of Fertility in Developing Countries, Academic Press, New York, 1983, pp. 797–828.

16. F. N.-A. Dodoo and A. Seal, 1994, op. cit. (see reference 6); and M. A. Koenig, G. B. Simmons and B. D. Misras, “Husband-Wife Inconsistencies in Contraceptive Use Responses,” Population Studies, 38: 281–298, 1984.

17. G. A. Oni and J. McCarthy, “Family Planning Knowledge, Attitudes and Practices of Males in Ilorin, Nigeria,” International Family Planning Perspectives, 17: 50–54 & 64, 1991; and S. Salway, “How Attitudes Toward Family Planning and Discussion Between Wives and Husbands Affect Contraceptive Use in Ghana,” International Family Planning Perspectives, 20: 44–47 & 74, 1994.

18. L. J. Beckman et al., 1983, op. cit. (see reference 13); F. N.-A. Dodoo and A. Seal, 1994, op. cit. (see reference 6); M. A. Koenig, G. B. Simmons and B. D. Misras, 1984, op. cit. (see reference 16); G. A. Oni and J. McCarthy, 1991, op. cit. (see reference 17); and S. Salway, 1994, op. cit. (see reference 17).

19. National Demographic Survey 1993, National Statistics Off fice and Macro International, Manila, Philippines, and Calverton, Md., USA, 1994.

20. Ibid.

21. J. B. Casterline, A. E. Perez, and A. E. Biddlecom, Factors Underlying Unmet Need for Family Planning in the Philippines, Research Division Working Paper No. 84, Population Council, New York, 1996.

22. J. Cohen, “A Coeff ficient of Agreement for Nominal Scales,” Educational and Psychological Measurement, 20: 37–46, 1960; and J. Cohen, “Weighted Kappa: Nominal Scale Agreement with Provision for Scaled Disagreement or Partial Credit,” Psychological Bulletin, 70: 213–220, 1968.

23. National Demographic Survey 1993, 1994, op. cit. (see reference 19).

24. L. C. Coombs and D. Fernandez, “Husband-Wife Agreement About Reproductive Goals,” Demography, 15: 57–74, 1978.

25. M. M. B. Asis, Husband-Wife Discrepancies in Fertility-Related Attitudes and Perceptions: Levels, Correlates and Their Influence on Contraceptive Practice, University of the Philippines, Manila, 1984.

26. United Nations, 1995, op. cit. (see reference 5).

27. J. B. Casterline, A. E. Perez, and A. E. Biddlecom, 1996, op. cit. (see reference 21).

28. Ibid.

Author's Affiliations

Ann E. Biddlecom is research associate and John B. Casterline is senior associate in the Policy Research Division, Population Council, New York. Aurora E. Perez is professor at the University of the Philippines, Manila. An earlier version of this article was presented at the annual meeting of the Population Association of America, New Orleans, La., USA, May 9-11, 1996, and appeared in revised form as Working Paper No. 92 of the Research Division, Population Council. Financial support for the research on which this article was based was provided by the EVALUATION Project, University of North Carolina; the Ford Foundation (Manila); the Mellon Foundation; The Rockefeller Foundation; and the United Nations Population Fund. Helpful comments on earlier versions were provided by David Hotchkiss, Cynthia Lloyd and John Bongaarts.

Disclaimer

The views expressed in this publication do not necessarily reflect those of the Guttmacher Institute.