Culture and Fertility: The Case of Thailand 9789814377508

Citation preview

FOREWORD It is generally recognized that ethnic diversity is characteristic of Southeast Asian countries. However, very little is known about the relationships between this diversity and other social and behavioural dimensions. In most countries in the region, such relationships have yet to be fully explored, partly because of the sensitivity of the issue and partly because of the dearth of essential data on ethnic behavfour and differentials. In the case of population behaviour, although several aspects, such as fertility, mortality and a variety of attitudinal dimensions, have been measured successfully in most Southeast Asian countries, they are normally presented as national aggregates, and are often only broken down by rural-urban residence, education, income, household size and ethnic group. Thus, while there is some information available on the relationship between ethnic identity and fertility, tile relationship has not been systematically examined in most of the Southeast Asian countries. The lack of this kind of basic data on ethnic differentials in population behaviour in Southeast Asia has definite implications for national population policies and programmes. These policies and programmes are usually implemented or launched on a national level with very little consideration for the ethnic diversity of the country, and are therefore often perceived, on an ethnic group or community level, as being ethnically based, unacceptable or even biased. Perceptions like these certainly have a direct effect on the manner in which such policies and prograJ:llUleS are implemented and received. With conditions as the foregoing in mind, and in view of the importance of, and the lack of information on, the relationship between dimensions of ethnic identity and population, the Institute in 1975 got together with a group of interested research scholars from Indonesia, Malaysia, the Philippines, Singapore and Thailand and began an investigation of such relationships and their implications for population policies and progress. Entitled "Culture and Fertility in Southeast Asia", this investigation consisted of two separate but linked activities, divided into Phase I and Phase II of the project, with the former focused largely on the analysis of secondary data and the latter on material generated by planned fieldwork and the administration of a questionnaire. The work that follows forms part of the "country monographs" growing out of Phase I of the project. These monographs, like the project itself, have

11

been made possible through the co-operation and support of a number of individuals and organizations, particularly the International Development Research Centre (IDRC), Ottawa, the various country team leaders and their colleagues, and the two co-ordinators of Phase I, Dr. Rodolfo Bulatao and Dr. Ong Jin Hui. To all of them we say, thank you. We are also particularly grateful to Dr. Ong jin Hui who, in addition to general co-ordinative responsibilities, helped to edit the manuscripts on which the country monographs are based. Whilst thanking all contributors to, and participants m, the project, and wishing the monographs all the best, we hope it is clearly understood that the responsibility for facts and opinions expressed in this publication rests exclusively with the authors and their interpretations do not necessarily reflect the views and policies of the Institute or its supporters. 5 September 1980

Kemial S. Sandhu Director Institute of Southeast Asian Studies

PREFACE

This volume is part of a series of monographs on culture and fertility m Southeast Asia. They arose out of a regional research project, "Culture and Fertility in Southeast Asia", initiated by the Institute of Southeast Asian Studies, and involving researchers from Indonesia, Malaysia, the Philippines, Singapore and Thailand. The first phase of the Culture and Fertility project concentrated on demographic variables and issues in relation to ethnic and other cultural variables. Each monograph, focusing on a particular country, is developed essentially on the following lines. First, the historical background and the development of the present social structure are detailed. Following this, the patterns of interaction among the various ethnic groups are considered, with particular attention paid to the identification of variables which may influence fertility patterns of the various groups. Against this backdrop, the population policies and family planning programmes are explained. The demographic structure and its changes in tandem with these policies and programmes are then analysed. In order to determine the impact of key variables on fertility patterns, a secondary analysis of available data was carried out, utilizing multiple classification analysis. There are differences in the substance of the individual country monographs, largely on account of the nature and availability of materials and documentation. All the same, there has been quite an adequate coverage of the areas deemed important in all the volumes. Indeed, the chapters on national population policies and family planning and the ones on demographic structure are almost complete in detail; in addition, they are comparable with one another. On the other hand, the chapters based on secondary analysis have problems of comparability because secondary data were used. Since these data sets were not originally designed to answer to the project's research model, there were difficulties in comparing variable definitions and operationalization. Even more problematic was the fact that some variables were not available in the data sets. Furthermore, the data sets were not comparable in population coverage and time-frame (the target year was 1970). For all these reasons, this chapter in each of the monographs should not be treated as anything more than a preview or pretest of the research model. Seen in this perspective, it not only provides a useful means of identifying relevant explanatory variables but also shows that variations do indeed exist in a number of areas between ethnic groups and between countries.

IV

Variations and shortcomings of the type above notwithstanding, the five monographs on the whole do provide a useful background to the identification of relevant ethnic variables. Moreover, the lack of comparability of data is currently being corrected in the second phase of the project. A standardized core questionnaire with additional peripheral and specific queries has been designed, based on the findings of the country studies of Phase I. Ultimately, it is expected that the analysis of data collected in Phase II will culminate m an in-depth examination of the relationship between ethnicity and fertility.

1 September 1980

Ong Jin Hui Editor, Country Monographs Culture and Fertility in Southeast Asia, Phase I

LIST OF TABLES

Table No. 1

Target Number of Acceptors, 1971-76, by Method

9

2

Average Number of Children Ever Born Per 1,000 Ever Married Women by Religion and Age

16

Average Number of Children Ever Born Per 1,000 Ever Married Women, by Religion and Urban-Rural Status -Unstandardized and Standardized for Age

16

Proportions of Women Ever Married and Marital Fertility Rates of Women at Reproductive Ages in 1960 and 1978

18

5

Analysis of Variance of Children Ever Born

21

6

Multiple Classification Analysis of Children Ever Born

22

7

Analysis of Variance of Desired Family Size

25

8

Multiple Classification Analysis of Desired Family Size

26

9

Analysis of Variance of Wife's Age at First Marriage

28

10

Multiple Classification Analysis of Wife's Age at First Marriage

29

11

Analysis of Variance of Current Use of Contraception

32

12

Multiple Classification Analysis of Current Use of Contraception

33

13

Analysis of Variance of Children Ever Born {Thai)

35

14

Multiple Classification Analysis of Children Ever Born {Thai)

36

15

Analysis of Variance of Children Ever Born {Chinese)

37

3

4

VI

Table No.

16

Page

Multiple Classification Analysis of Children Ever Born {Chinese)

38

17

Analysis of Variance of Children Ever Born (Moslem)

40

18

Multiple Classification Analysis of Children Ever Born (Moslem)

41

19

Analysis of Variance of Desired Family Size (Thai)

43

20

Multiple Classification Analysis of Desired Family Size (Thai)

44

Analysis of Variance of Desired Family Size (Chinese)

45

21

22 Multiple Classification Analysis of Desired Family Size (Chinese)

46

23

Analysis of Variance of Desired Family Size (Moslem)

47

24

Multiple Classification Analysis of Desired Family Size (Moslem)

48

Analysis of Variance of Wife's Age at First Marriage (Thai)

50

Multiple Classification Analysis of Wife's Age at First Marriage (Thai)

51

Analysis of Variance of Wife's Age at First Marriage (Chinese)

52

Multiple Classification Analysis of Wife's Age at First Marriage (Chinese)

53

Analysis of Variance of Wife's Age at First Marriage (Moslem)

54

25

26

27

28

29

V1l

Table No. 30

31

32

33

34

35

36

Multiple Classification Analysis of Wife's Age at First Marriage (Moslem)

55

Analysis of Variance of Current Use of Contraception (Thai)

56

Multiple Classification Analysis of Current Use of Contraception (Thai)

57

Analysis of Variance of Current Use of Contraception (Chinese)

58

Multiple Classification Analysis of Current Use of Contraception (Chinese)

59

Analysis of Variance of Current Use of Contraception (Moslem)

60

Multiple Classification Analysis of Current Use of Contraception (Moslem)

61

CONTENTS

FOREWORD PREFACE LIST OF TABLES

Ill

v

1:

CULTURAL DIVERSITY

1

II:

INTERGROUP BEHAVIOUR

3

III:

POPULATION POLICIES -· PAST AND PRESENT

7

IV:

DEMOGRAPHIC CHARACTERISTICS

11

V:

CULTURAL DETERMINANTS OF FERTILITY

18

BIBLIOGRAPHY

63

The Institute of Southeast Asian Studies

The Institute of Southeast Asian Studies was established as an autonomous organization in May 1968. It is a regional research centre for scholars and other specialists concerned with modern Southeast Asia. The Institute's research interest is focused on the many-faceted problems of development and modernization, and political and social change in Southeast Asia. The Institute is governed by a twenty-four member Board of Trustees on which are represented the National University of Singapore, appointees from the government, as well as representatives from a broad range of professional and civic organizations and groups. A ten-man Executive Committee oversees day-to-day operations; it is chaired by the Director, the Institute's chief academic and administrative officer. The responsibility for facts and opzmons expressed in this publication rests exclusively with the authors and their interp~retations do not necessarily reflect the views or the policy of the Institute or its supporters.

"Copyright subsists in this publication under the United Kingdom Copyright Act, 1911, and the Singapore Copyright Act (Cap. 187). No person shall reproduce a copy of this publication, or extracts therefrom, without the written permission of the Institute of Southeast Asian Studies, Singapore."

CULTURE AND FERTILITY The Case of Thailand

by

Suchart Prasithrathsint Likhit Dhiravegin Chavalit Siripirom

Research Notes and Discussions Paper No. 22 Institute of Southeast Asian Studies

1980

1:

CULTURAL DIVERSITY

The population of Thailand is approximately 43 million. Most of this population comprise the Thai people, but there are, in the main, two distinct minority groups characterized by district: the Thai-Moslems and the Chinese. A number of other minorities ·· the Vietnamese, Kampuchean refugees, hilltribes and such ·· also exist. The latter, however, will be ignored in the present discussion.

The Thai Historical Background There is much ongoing debate as to the ongm of the Thais. One commonly accepted thesis is that the Thais existed in tribes in various parts of China. From about 69 B.C. there was a steady migration of these groups to the Indochinese Peninsula. Consequently, various kingdoms came to be established at Sukhothai, Ayuthaya, Thonburi and Bangkok. 1 Except for a brief spell of Burmese domination in 1767, these kingdoms underwent indigenous changes to develop into present-day Thailand.

Thai Culture The focus of Thai culture can be said to be Theravada Buddhism. This religion of the Thais incorporates two influential notions ·· karma and meritaccumulation for happiness in present and future reincarnations. "Karma" emphasizes individual responsibility for fate and perhaps this explains the Thai's tendency to prize individualism very highly. Consequently, family ties amongst the Thais are not very strong. The matrilocal family system ·· prevalent in rural Thailand ·· has quite easily given way to nuclear family households in urban Thailand. Most of the Thais, however, live in rural areas where they are chiefly occupied with rice-growing. However, there are also Thais who hold important positions in government and other institutions. 1 Rang_ Syamauda, A History of Thailand (Bangkok: pp. 6-8.

Thai Watana Panich Co. Ltd., 1973),

2

The Chinese Historical Background Chinese immigration to Thailand can be traced back to as early as the fifteenth century. 2 The largest influx occurred around the nineteenth and early twentieth centuries, mainly as a result of the famine and civil disorders which characterized these periods. This flow of migrants was mainly from Guangdong province and it has given Thailand four of its main Chinese dialect groups, viz., the Teochius, the Hakka, the Cantonese and the Hainanese. The other distinct Chinese dialect group found here is that of the Hokkiens from Fujian province. Of these five groups, the Teochius, comprising 56% of the total Chinese population in Thailand, dominate numerically and they are followed by the Hakkas and Hainanese with 16% and 12% of the total Chinese population respectively. The Hokkiens and Cantonese make up only 7% of the total.

Cbin.t-&t Cuttutt Dialect differences notwithstanding, the culture of the Chinese can be said to be based on Confucianism. This is easily observable in the emphasis given to ritualistic ancestor-worship and the highly stressed "code of ethics" of five relationships ·· ruler and ruled; parents and children; older brother and younger brother; husband and wife; and friends. It will be noted that of these five, three are pertinent to familial relationships. Taking all this into consideration, it may be seen that family is an important fact of Chinese culture. The Chinese have a patrilocal family system based on a system of seniority and gerontocracy. Whilst individuals may seek self-progress, it is family fortunes that are given real concern. Ancestor-worship· further underlines a preference for male children; in fact, according to Confucian tradition, the more sons one begets, the greater one is blessed. However, though these elements persist strongly, they have simultaneously allowed Chinese adoption of Mahayana Buddhism as a major religion. In addition, many second generation Chinese have quite easily assimilated significant aspects of Thai culture.

2 See, for instance, William G. Skinner, Chinese Society in Thailand: (Ithaca: Cornell University Press, 1957).

An Analytical History

3

Occupationally, however, the Chinese remain a distinct group. The majority are engaged in wholesale and retail businesses -- ranging from street stalls and small retail shops to large industries.

The Thai-Moslem Historical Background The Thai-Moslem;. -- as their title denotes -- are Thai, but they should be distinguished from the majority by their adherence to Islam. Their adoption of this religion has given rise to a "new" Thai minority because of Islam's very distinct characteristics.

Thai-Moslem Culture Thai-Moslem culture is significantly dominated by the Islamic faith. The strength of this domination is best seen by the existence of separate schools and laws for Thai Moslems. Unlike the Chinese case, assimilation into Thai culture, not surprisingly, is norexistent. Most of the Thai Moslems do not even speak Thai. The patrilocal family structure allows a man to take four wives, provided he can support all of them. Family ties are sufficiently strong, but the stress is on Islamic brotherhood that goes beyond mere narrow limits. Thai-Moslems are concentrated in four border provinces m South Thailand, where they are engaged on rubber plantations, rice fields and in orchards.

II:

INTERGROUP BEHAVIOUR

Sino-Thai Relations Till recently, Sino-Thai relations have been smooth and conflict-free.

As

4

mentioned, the Chinese began migrating to Thailand from about the fifteenth century. During this early period of migration, the Chinese dealt mainly with Thai Kings and were mostly involved with them over trading. Chinese migrants were mainly males. They tended, therefore, to settle down with Thai women. The children -- known as "lukjir" -- were quickly and effortlessly assimilated into Thai society. This easy assimilation was allowed by the fact that Chinese culture -- dominated as it was by Confucianism -- had ample room for the adoption of Thai culture. Social mingling was also without barriers. Most important, perhaps, was an absence of conflict of economic interests. Trading was not highly valued in Thai culture. Additionally, the corv~e system, patronage bond and slavery engaged the majority of Thai males -- leaving a social vacuum which the Chinese easily filled. Then again the abolishment of the corvee system and store emancipation in the reign of Rama V greatly increased the demand for Chinese labour. This happily coincided with the period of massive migration of the Chinese to Thailand. The nse of Chinese nationalism at the end of the nineteenth century interrupted this conflict-free history. Hostility began to develop and precautionary measures against the threat of Chinese influence were taken. The communist threat led to the adoption of containment policies. However, despite this setback, the lack of physical, cultural and religious differences among the Thais and Chinese prevailed and continued to boost Sino-Thai co-operation and assimilation. An interesting feature of this is that despite some important prejudice about "the uncouth Chinaman", the Thais are never inclined to reject anyone of Chinese ancestry who speaks and behaves like a Thai. 3 The Chinese, for their part -even whilst having stereotyped notions of the "husu" (barbarian), lazy, loose and not very intelligent Thai -- continue to adopt Thai patterns of culture. For instance, of the approximately 10,000 cases of registration for changes of names annually, 90% are Chinese wanting to change their last names to Thai names. 4

Thai and Thai-Moslem Intergroup Relations Geographic location (that is, conc~!ntration of Thai-Moslem> in southern border provinces), language (viz., Malay) and Islam as a chosen religion are S William G. Skinner, Chinese Society in Thailand: University Preas, 1957), p. 381.

An Analytical History (Ithaca:

Cornell

4 Boorootpat, Kh~atpai, Panha Chonltloomnoi Nni Pratesthai (Minority problems in Thailand]. (Banakok: Prae Pittaya, 1972), p. 97.

5

major barriers to the interactiqn and communication between the Thai majority and the Thai-Moslem minority. These factors also create crucial barriers to assimilation. History has also left its negative mark; those border provinces that the Thai-Moslems have traditionally concentrated in were once tributary states of Siam. Conflict as a predominant facet of Thai and Thai-Moslem relations throughout history has been well summed up by the fact that "the heroes for the Thai majority are invaders for the Muslim minority, and heroes of the Moslem minority are rebels of the Thai majority." The hostile feelings that exist till today probably form the chief reasons behind the contemporary existence of various Muslim movements in Southern Thailand -- for example, the Pattani United Liberation Organization (PULO), the Posu group, Wamaeduesa, Poyae group, Nai Seng group, and so on. 5 To the Thai Government and people, the Thai-Moslems appear to lack a Thai national identity. Hence, their loyalty to Thailand is suspect; it is believed that Thai-Moslems consider themselves Malays for they prefer to speak in Malay, study the Malay language and preserve a culture that is similar to the Malays. The average Malay is well aware that Thai-Moslellli refuse to further their studies in Thai schools beyond that which is compulsory. The Thai-Moslems, on the other hand, perceive the majority, especially the government, as being hypocritical and contemptuous of them. To make matters worse, they believe that the Thai Government is trying to destroy their religion and identity -- a serious contention, for religion is a highly emotional matter. Aware of the latter's suspicions, Thai authorities have taken a guarded, pluralistic approach towards the Thai-Moslems. For instance, the government has legalized polygamy among the Thai-Moslems a."'ld granted the right of leave of absence for religious activities in Mecca. 6 Also, in October 1963, a programme of Malay language training was launched for officials assigned to posts in the southern provinces.

5 Arong Sutthaaart, The Problem of Conflicts in the Four Southern Provinces (Bangkok: Phitak Pracha Publishing Co., 1976), pp. 129-130. 6 "Decree on leave with pay of the civil and judicial officials," Government Gazette, Book 82, Part 6 (August 1965 ), Article II, Subsection.

6

Simultaneously, attempts to integrate Thai-Moslems and create national consciousness amongst them have also been made. One such example is the introduction of admission quotas for Thai-Moslem students to universities in Bangkok without their having to take entrance examinations. Even so, the most salient feature of Thai and Thai-Moslem relationship is the constant political unrest in Southern Thailand. This includes many separatist movements since 1932. Significantly, at least one of these movements -- the National Liberation Army of Pattani, founded in 1960/61 -- defines itself as having a religious mission.

Intergroup Perceptions and Fertility There has not been any single study on intergroup perceptions among the Thai Chinese and Thai-Moslems in relation to one another's fertility-patterns. However, impressionistic evidence and personal experiences allow for the following hypothesis. First, the Thais feel that the Chinese Confucian beliefs explain their desire for large families and many sons and their unwillingness to practise family planning. Whilst this belief is not groundless, it is not entirely true either. Secondly, the Thais deem themselves to be sexually superior to the Thai-Moslems for it is believed that they have comparatively superior reproductive organs. These attitudes towards the Chinese and Thai-Moslems, however, have no influential bearing on Thai fertility. Family planning is adopted on an individual basis and it has never become conscious behaviour on an ethnic level. The fear of becoming outnumbered is quite nonexistent except among some members of the ruling ""elite. The latter's fears relate to Chinese fertility among the Chinese in Thailand as well as in the People's Republic of China (PRC). Some consider a large population desirable on a military power rationale. 7 However, such arguments are comparatively rare. On the whole, there is consensus that this fear is baseless whilst family planning is necessary.

7 See Tawee Rangkum, "Pratesthai Tongkarn Nayobai Prachakom Nai Puchubun Rue Mai Yangrai" [Does Thailand need a population policy at present?], Paper presented at the First National Seminar on Population, 1963.

7

III:

POPULATION POLICIES -- PAST AND PRESENTS

Population Policy Prior to 1971 During the first half of the present century, Thailand's official stance on population was predominantly pronatalist. Over sixty years ago, when there were only about eight million people in Thailand, the Minister of Interior said that we needed five to six times more people. Health services first established in the early part of the twentieth century had lower mortality and higher fertility as one of their significant goals. Early marriages were then encouraged to "make the nation powerful". As late as 1956, bonuses were offered for large families. It remained for a World Bank economic mission m 1958-59 to recommend that the government seriously consider the adverse effects of the population growth rate on economic development. Consequently, the Office of the Prime Minister established a series of committees to study the problem and to make recommendations to the Cabinet. In addition, between 1963 and 1968, three National Population Seminars were held.

Throughout this period, with the exception of one committee, the reports of the various committees and the three seminars all warned of the dangers of a too rapid population growth. However, it was only after the third population seminar in 1968 that the Cabinet asked the National Economic Development Board (NEDB) to prepare a set of final recommendations. During the period prior to an actual population policy declaration, the Cabinet did agree that the voluntary practice of family planning was permissible and, in 1967, the Prime Minister affixed his signature to the United Nations World Leaders' Statement on Population. Beginning in 1968, the Cabinet endorsed the development of family planning services by the Minister of Public Health on a research basis. In late 1969 and early 1970, the NEDB, together with the Ministry of Public Health and the Institute of Population Studies, in a comprehensive report to the Cabinet, strongly recommended the adoption of a population control 8 For a detailed account of the historical development of population policy in Thailand, see Suchart Prasithrathsint, "Historical Development of the Studies of Population Problems and Population Policy in Thailand," in his "Economic and Fertility Behaviour of the Rural People in Thailand" (Ph.D. Thesis, Brown University, 1971), pp. 48-85.

8

policy. In March 1970, the Cabinet accepted the re)liJrt and declared that "the Thai Government (supports) voluntary family planning in order to resolve various problems concerned with the very high rate of population growth, which constitutes an important obstacle to the economic and social development of the nation."

Population Policy 1972-76 With the Thai Government's announcement, the Ministry of Public Health established the National Family Planning Project (NFPP). A five-year proposal had by then been drawn up for inclusion in the Five-Year Social and Economic Plan (1972-76) of the NEDB. The specific objectives of the NFPP are as follows: (1)

to reduce the population growth rate from over 3% to about 2.5% by the end of 1976,

(2)

to inform and motivate eligible women, particularly those living m rural and remote areas, and to make services readily available throughout the country, and

(3)

to integrate family planning activities with overall maternal and child health services and thus to mutually strengthen the activities in these closely related fields.

The five-year plan for family planning related primarily to activities directly under the jurisdiction of the Ministry of Public Health, including general services, training, public information and programme research and evaluation. New acceptor targets have been set for the period 19 71-7 6 and are shown in Table 1. Based on performances to date, these targets are realistic, being based on calculations of woman years of protection and number of births prevented by this programme; thus the objective of lowering the growth rate from over 3.0% to 2.5% by 1976 IS thought to be within reach. There is no official policy regarding migration and no effort has been made to restrict internal migration. However, it is recognized that the very rapid growth of Bangkok has created severe economic strains. Thus, the current policy aims at creating alternative urban "growth poles", particularly in the Northeast Region, from which migration has been heavy due to inadequate

9

Table 1:

Target Number of Acceptors, 1971-76, by Method

Year

Oral Pill

IUD

Sterilization

1971

200,000

80,000

20,000

300,000

1972

235,000

90,000

25,000

350,000

1973

280,000

90,000

30,000

400,000

1974

280,000

90,000

35,000

405,000

1975

280,000

90,000

40,000

410,000

1976

280,000

90,000

40,000

410,000

1,555,000

530,000

190,000

2,275,000

AD six years

All Three Methods

nonagricultural employment opportunities. This is to be done by deliberate building-up of the infrastructure (electricity, water supply, etc.) and by encouraging location of new industries there. Regional development study groups are also developing similar master plans for other regions of the country. It should also be noted that the family planning programme itself is viewed as one measure to deal with growth in the urban areas. The Greater Bangkok Master Plan prepared by the Ministry of Interior called attention explicitly to the growth caused by internal increase (as well as migration) and proposed family planning services as a remedial programme.

Population Policy 1977-81 I The Fourth National Economic and Social Development Plan of Thailand (1977-81) plans to decrease the population growth rate from 2.5% to 2.1%; the birth rate from 34.5/1,000 to 29.0/1,000; the mortality rate from 9.0/1,000 to 8.0/1,000. If this is achieved, the population will increase by 2.2% per annum throughout this period.

The target was determined by examining existing capital and manpower for family planning services also. Emphasis will be placed on people in remote

10

areas, slums, certain minorities in the South, and potential family planning acceptors. In order to achieve the target, the two important strategies, viz., family planning with a target of 3.03 million acceptors by the end of 1981, and population education, are to be implemented. In addition, measures to encourage migration to less populated areas are to continue. Refugees from Indochina, too, will have to be guarded against. As a corollary to all these, the plans include a heavy emphasis on reducing problems and diseases related to malnutrition, especially in infants, young children and pregnant women; decreasing unequal distribution of public health services among the population; improving the efficiency of the existing public health centres; and increasing the number of doctors, nurses and other para-medical personnel.

Organizational Structure The Cabinet, at the time of the declaration of the population policy, established a National Population Policy Committee which is chaired by the Minister of Health, with the Secretary General of the NEDB and the Deputy Minister of Health as the Deputy Chairman. Members include the undersecretaries of all ministries and the directors of all pertinent specialized agencies, such as the National Research Council, the National Statistical Office, and the Institute of Population Studies. The primary responsibility for the implementation of a national family planning programme was assigned to the Ministry of Public Health, which serves, together with the NEDB, as co-ordinator of all population activities. The Ministry of Public Health provides curative and preventive health services throughout Thailand, particularly in the rural areas. It is interesting to note that health facilities in urban areas, particularly in Bangkok, are operated by the individual municipalities which are responsible to the Ministry of Interior. Other agencies, such as the Ministry of Defence (responsible for military hospitals), also provide health care. There are three departments in the Ministry of Public Health: Medical Services, Health and Medical Sciences, each headed by a Director General. The Government of Thailand is a centralized government, so there is direct control by each central ministry over units in the provinces. Provincial governors,

11

appointed by the Ministry of Interior, also have some authority over governmental personnel assigned to their provinces. Thus senior personnel in each province have, in effect, two bosses: their own Director General and the Governor (see Figure 1).

IV:

DEMOGRAPHIC CHARACTERISTICS

Five years after the World Bank Expert Group had pointed out the significance of the demographic problems, the first social science survey research project to study fertility and factors affecting it was conducted in 1964. Since then, numerous scientific studies of fertility and family planning have been nndertaken. A few attempts using the census data and other sources of information to study fertility differentials have also been made. 9 Such studies reveal that there are urban-rural fertility differentials. One study shows that, for the nation as a whole, the average number of live births per 1,000 evermarried women was 4,261 in 1960. But the fertility level varied from a high 4,461 for the rural agricultural population to a low 3,3 75 for those living in Bangkok. Bangkok's level, about 25% below that of the most rural part of Thailand, confirms the significant impact of the urban metropolis on fertility levels in the country. However, the predominantly rural character of Thailand accounts for the high fertility level characterizing the country as a whole. 10 It is clear that, in both Bangkok and provincial urban centres, fertility is still well above replacement levels. Thus, rapid population growth in Thai cities in the foreseeable future is certain, even if rural migration is somehow slowed down or eliminated.

9 Thip Chalothom, "Fertility Levels and Differentials in Thailand.'' International Union for the Scientific Study of Population, Sydney Conference (Canberra: Australian National University, 1967), p. 355; A. Das Qlpta, S. Chotechanapibal, Thip Chalothom, and W. Siripak, "Population Perspective of Thailand," Sankhya: The Indian Journal of Statistics, Series B, Vol. XXVII (1965), Parts 1-2; United Nations, Population Bulletin, No. 7, With Special Reference to Conditions and Trends of Fertility in the World (New York: United Nations, 1965); Thip Chalothom, "Differential Fertility of Thai Women" (Paper presented at the Seminar in Demographic Research Population Studies Centre, University of Pennsylvania, 1966), pp. 15-29. 10 S. Goldstein, "Religious Fertility Differentials in Thailand," Population Studies, Vol. XXIV, No. 3 (Novaalaer 1.970), p. JJa.

Figure I:

Population/Family Planning Organization Chart

I

I ,,

I' '

I PROVINCIAL HEALTH OFFICES

~

NATIONAL ECONOMIC DEVELOPMENT BOARD

---

~~~

''

'

NATIONAL FAMILY PLANNING PROGRAMME

ADMINISTRATION

I

__...-,_

.................

I

RURAL HEALTH CENTRES

I

MINISTRY OF PUBUC HEALTH

I

C'l'

I

NATIONAL POPULATION POLICY COMMITTEE

OTHER MINISTRIES AND SPECIALIZED AGENCIES

DEPARTMENT OF HEALTH

CABINET

I

I RESEARCH AND EVALUATION

~-----

I TRAINING AND SUPERVISION

-~

--

I

DEPARTMENT OF MEDICAL SERVICES

-~

I

PUBLIC INFORMATION

I 84 PROVINCIAL HOSPITALS

I

13

It was found that the average number of children ever hom to the rural ever married women was 1 7% higher than that of children hom to women residing in Bangkok. It was speculated that, in a population where deliberate family limitation was not extensive, age at entry into marriage would be a very important determinant of a woman's cumulative fertility. Such a speculation was confirmed by a strong inverse association between age at first marriage and cumulative fertility of both rural and urban women. 11

Another study found that women who expect large families are those who start their marriage life relatively early in the reproductive cycle (20-22 years of age), live with a perceived inadequate income for normal household expenses, have conservative attitudes towards legalization of abortion, and have husbands who are less exposed to mass media and entertainment (as measured by infrequent movie attendance). Women who prefer large families are those who live with a moderate monthly income (Baht 1,001-1,500),1 2 and are primarily traditional as measured by definite sex preference for children, infrequent movie attendance, segregated leisure activities, and no postmarriage work experience. 13 With regard to socioeconomic status and fertility, it was found that in the rural areas, those who were most favourable towards family planning were the wealthiest (as measured by the scores of material possession) whilst those who were the poorest were less in favour of family planning.l 4 Actual fertility among rural couples was also found to be in excess of ideal fertility, except for the youngest groups. This was interpreted as being due to the couples' lack of ability to control fertility effectively. One explanation offered was that the rural people resorted to birth control rather late, that is, they probably had not practised it either sufficiently early in the reproductive cycle or with adequate regularity to have a significant impact on the number of children ever hom. On the average, most couples first used birth control only after the fifth pregnancy. 15

11

Suchart Prasithrathsint, "Economic and Fertility Behaviour of the Rural People in Thailand" (P.D. Thesis, Brown University, 1971), pp. 167-171.

12

US$1 is roughly equivalent to Baht 25.

13

Chavalit Siripirom, "Fertility Differentials in Chiangmai: Socio-Economic Status, LifeStyle, and KAP' Characteristics" (Dissertation, Syracuse University, 1976).

14

Suchart Prasithrathsint, Some Factors Affecting Fertility and Knowledge, Attitude and Practice of Family Planning Among Rural Thai Women, Institute of Population Studies, Working Paper No. 2 (Bangkok: Chulalongkorn University, 1973).

15

Ibid.

14

Besides place of residence, perhaps no other social variable has been associated with fertility differentials as frequently as education. It is generally expected that fertility and educational levels are inversely related. Much of the evidence so far accumulated support this inverse relationship. 16 There is some question, however, about the extent to which these relationships hold either prior to or in the very early stages of the transition from high to low fertility, as well as during the post-transition period. 1 7 Nevertheless, data from the 1960 census indicates that, in both rural and urban Thailand, illiterate women have higher fertility rates than literate women, although the difference was larger in urban areas. In addition, a general inverse relationship between fertility and years of schooling was established for both rural and urban women.18

Within the rural sample, because the number of women who received more than a minimum elementary education is very small, it is difficult to examine differentials beyond the base range of 0-4 years of schooling for most age groups. Within this range in all age groups, however, women with no schooling had the greatest number of children ever born and, among the youngest women, those who had completed the fourth grade had the lowest number. The differences, however, are not large. For the urban sample, it is possible to differentiate within a much wider range of educational levels since a fair proportion of women would have attended schools beyond the fourth grade. For all age groups, there is a general inverse relationship between children ever born and years of schooling: women with either no schooling or with only 1-3 years had the most children.19

16

David M. Heer, "Educational Advance and Fertility Change," International Population Conference (London, 1969), Vol. Ill (Liege: International Union for the Scientific Study of Population, 1971), pp. 1903-1914.

17

Clyde V. Kiser, "Educational Differentials in Fertility in Relation to the Demographic Transition," Intematior:.al Population Conference (London, 1969), Vol. Ill (Liege: International Union for the Scientific Study of Population, 1971), pp. 1926-1935.

18

S. Goldstein, A. Goldstein and P. Tirasawat, The Influence of Labour Force Participation and Education on Fertility in Thailand, Institute of Population Studies, Research Report No. 2 (Bangkok: Chulalongkorn University, 1972).

19

Suchart Prasithrathsint, Some Factors Affecting Fertility and Knowledge, Attitude and Practice of Family Planning Among Rural Thai Women, Institute of Population Studies, Working Paper No. 2 (Bangkok: Chulalongkorn University, 1973); S. Goldstein, "lnB.uence of Labour Force Participation and Education on Fertility in Thailand," Population Studies, Vol. XXVI, No. 3 (November 1972).

15

As to the relationship between woman's labour force participation and fertility in Thailand, available data shows that a large majority of Thai women are members of the labour force and most of them are engaged in farming. The analysis points to a differential relation between labour force participation and fertility in rural, agricultural places and in the urban centre of Bangkok. In the latter, fertility of women in the labour force is lower than that of housewives. In rural, agricultural places, the fertility of employed women is higher, although minimally so. This pattern suggests that the greater separation of work and family roles among employed women in the urban centres lowers fertility of urban working women, whereas the general absence of such conflict m rural society results in minimal effect of labour force participation on fertility. 20 Differentials also exist on the basis of occupation. Women engaged in farming have by far the highest fertility -- 4,503 per 1,000 ever married women, standardized for age. Women engaged in sales work have the next highest level of fertility -- 3,887 per 1,000. They are in turn followed by women engaged m crafts, women in professional and administrative work, and finally, Women in service and transportation. For the last three groups, fertility levels varied within the relatively narrow range of 3,500 to 3,600, levels approximately 1,000 below those of women working on farms. 2 1 With regard to ethnic fertility differentials, almost no studies have been done. This is partly due to the lack of a conceptual framework for ethnic identification. However, a study of religious fertility differentials has been found to be relevant. According to this analysis, based on the 1960 census data, 93.6% of the population are Buddhist, 3.9% are Moslem, and 1. 7% are Confucian. Together, the remaining minority groups of Christians, Hindus and Others constitute less than 1% of the Thai population. It was found that religious fertility differentials do exist. (See Tables 2 and 3.) When both age and urban-rural status are controlled, Confucian women have slightly more children than Buddhists, while both these groups average considerably more children than do Moslem women. 22 20

Suchart Prasithrathsint, "Female Labour Force Participation and Fertility Behaviour in Rural Thailand," Journlll of Social Sciences, Vol. 8, No. 4 (October 1971), pp. 34-48; S. Goldstein, "Influence of Labour Force Participation and Education on Fertility in Thailand," Population Studies, Vol. XXVI, No. 3 (November 1972).

21

S. Goldstein, "Influence of Labour Force Participation and Education on Fertility in Thailand,' Population Studies, Vol. XXVI, No. 3 (November 1972).

22

S. Goldstein, "Religious Fertility Differentials in Thailand," Population Studies, Vol. XXIV, No. 3 (November 1970), p. 324.

16

Table 2:

Average Number of Children Ever Born Per 1,000 Ever Married Women by Religion and Age

Average Number of Children Ever Born

Age

Total·

Buddhist

Moslem

13-24

1,230

1,038

1,918

1,222

25-34

3,341

3,029

3,545

3,329

35-49

5,784

4,231

5,008

5,703

50 and over

5,782

4,221

3,930

5,675

Total

4,320

2,988

3,998

4,261

Total Standardized for Age

4,367

3,391

3,842

4,261

Confucian

Source: S. Goldstein, "Religious Fertility Differentials in Thailand," Population Studies, Vol. XXIV, No. 3 (November 1970), p. 324.

Table 3:

Average Number of Children Ever Born Per 1,000 Ever Married Women, by Religion and Urban-Rural Status -- Unstandardized and Standardized for Age

Urban-Rural Status

Buddhist

Moslem

Confucian

Total

Index (Rural, agricultural= 100)

Unstandarcfized Bangkok Other urban, nonagricultural Urban, agricultural Rural, nonagricultural Rural, agricultural

3,350 3,834 4,246 3,904 4,474

3,714 2,675 3,250 2,990 3,214

3,603 4,123 4,829 4,750 4,657

3,278 3,801 4,222 3,866 4,410

Total

4,320

2,988

3,998

4,261

Bangkok Other urban, nonagricultural Urban, agricultural Rural, nonagricultural Rural, agricultural

3,319 3,812 4,01!1 4,019 4,517

Standardized for Age 3,902 3,520 3,802 3,875 3,613 4,045 3,261 4,270 3,411 4,708

3,375 3,778 3,997 3,982 4,461

Total

4,367

3,391

3,342

4,261

Total Standardized for Urban-Rural 4,357

3,389

3,550

4,261

Source:

74 86 96 88 100

76 85 90 89 100

S. Goldstein, "Religious Fertility Differentials in Thailand," Population Studies, Vol. XXIV, No. 3 (November 1970), p. 330.

17

Community surveys in Yala, Potharam and Bangkhen indicate that more frequent divorces and remarriages may have reduced Moslem birth rates. 23 However, it is unlikely this adequately explains the lower birth rates amongst the youngest age group of Moslems f4:>r Moslems tend to marry earlier. It is more probable that in some ways Moslem women in Thailand either voluntarily or involuntarily limit the number of children they bear to a greater extent than do either Buddhist or Confucian women. Reliance on older traditional methods of birth control or higher rates of sterility due to poor health may be contributory factors. With respect to fertility trends in Thailand, a decline in fertility is evident m most fertility measures. 24 For instance, a comparison of marital fertility rates m 1960 and 1970 reveals a definite decrease in all age groups. The proportion of t:Yer married women at reproductive ages also shows a declining trend in all but the first two youngest age groups (see Table 4). In fact, fertility rates in many parts of the country have declined rapidly during the past decade because of increasing availability of acceptable birth control methods in the rural areas and family planning campaigns carried out by both public and private sectors. Rapid urbanization and increasing economic pressure on parents to have small families have also contributed to the fertility decline. It is expected that existing socioeconomic fertility differentials will become much less in the near future.

23

See Soontaree Suvipakit, A Pilot Study of Family Health in Thai Muslim Communities in South Thailand, Yala Project, Research Report No. 1 (Bangkok: National Research Council, 1969); Institute of Population Studies, The Potharom Study, Research Report No.4 (Bangkok: Chulalongkorn University, 1971); Amos H. Hawley and Visid Prachuabmoh, "Family Growth and Family Planning: Responses to a Fal!l!ly: Jllanning Action Program_ iii a Rurar·Districfof Thilliiild;'T Demography, Vol. TII (1966), pp. 319-331; Visid Prachuabmoh and James T. Fawcett, "Fertility Control in Rural Thailand: Some Results of a Demonstration Project in Potharam District," International Population Conference (Sydney, 1967) (Liege: International Union for the Scientific Study of Population), pp. 492-500; Donald 0. Cowgill et al., Family Planning in Bangkhen, Thailand (Bangkok: Mahidol University, 1969}.

24

Suchart Prasithrathsint, "Trends and Differentials in Fertility," in United Nations, Economic Commission for Asia and the Pacific (ESCAP), Country Monograph Series No. 3: Population of Thailand {Bangkok: United Nations, 1976), Table 67, p. 66 and Table 71, p. 68.

18

Table 4:

Proportions of Women Ever Married and Marital Fertility Rates of Women at Reproductive Ages in 1960 and 1978

Proportioosof Women Ever Married Age Groups

Marital Fertility Rates

1947

1960

1970

1960

1970

15-19

19.4

13.8

19.0

422.4

353.1

20-24

70.0

61.2

62.0

464.1

466.8

25-29

89.1

85.7

84.3

397.4

362.8

30-34

94.3

93.2

91.9

323.2

283.9

35-39

96.0

95.7

94.7

269.3

228.3

40-44

96.8

96.8

96.1

139.9

119.9

45-49

97.1

97.3

96.9

36.5

32.7

Source:

Suchart Prasithrathsint, "Trends and Diffetentials in Fertility," in United Nations, Economic Commission for Asia and the Pacific (ESCAP), Country Monograph Series No. 3: Population of Thailand (Bangkok: United Nations, 1976), Table 67, p. 66 and Table 71, p. 68.

V:

CULTURAL DETERMINANTS OF FERTILITY

In this chapter, an attempt is made to discover the extent and pattern of the relationship between fertility and fertility-related measures on the one hand and that between a set of independent variables and covariates on the other hand. The major purpose of the following analysis is to find out the significance of ethnicity relative to other independent variables. Fertility and fertility-related measures used as dependent variables are the number of children ever born, woman's desired family size, woman's age at first marriage and birth control practice. The set of independent variables used in the following analysis are ethnicity, place of residence, woman's education, husband's education and the level of household income. The covariates are woman's work status, woman's current age and length of marriage duration.

19

Data Source As far as data is concerned, vanous existing sources of data have been

studied to see if any of them is suitable for secondary data analysis. To be suitable, the data source must have the minimum amount of information needed for cross-national comparative study; that is, it must have all the dependent as well as independent variables and the covariates that would be used in all countries involved in this cross-national research project. In other words, it must have data on ethnicity, number of live births, desired family size, woman's age at first marriage, birth control practice, place of residence, woman's current age, marriage duration, couple's education and household income. Several fertility studies and surveys have been conducted since 1958, but the majority of them were confined to the Thais only. Few of them dealt with ethnic fertility differentials: one was concerned with the fertility of the Moslem; another made a comparative study of the Thai and the Moslem women, and a third examined religious fertility differentials, using the 1960 population census data. The national census data was first examined. It was found that several variables were missing. Among these were desired family size, woman's age at first marriage, household income and contraceptive practice. Thus, census data could not meet the objective and the requirements of the research project. An attempt was also made to investigate another possible source of data -- a National Longitudinal Survey of Social, Economic and Demographic Change conducted by the Institute of Population Studies in both rural and urban areas of the country. Even though the Survey had all the variables needed, it was confined to the Thai majority. The numbers of Chinese and Moslems in the survey were very small, insufficient for a comparative analysis. As mentioned, few attempts have been made to study possible relationships between culture and fertility, the focus here being on ethnic differentials. The major available data -- used in the following secondary analysis -- comes from three pilot studies conducted in 1977. These studies, on the differential fertility of the three main ethnic groups, were carried out by graduate students of the Department of Social Sciences and the Faculty of Environment and Resource Studies of the Mahidol University. The surveys, conducted as field practice requirements for their courses on research methodology and population studies, were supervised by the staff of the Department of Social Sciences.

20

The three pilot fertility sample surveys were carried out in the city of Bangkok and in the rural fringe or suburban areas of Bangkok. The Moslems were selected from their major concentrations in Bangkok and its outlying areas. The Chinese and the Thai were selected from various residential and commercial areas in Bangkok and its suburbs. No attempt was made to make the samples representative of their ethnic· groups. Altogether 883 women were interviewed, of which 327 were Thais, 267 Chinese, 274 Moslems, and 15 of unknown ethnic identity. The majority of them, 74 7 cases, lived in the city of Bangkok. Their ages ranged from 16 to 53 years, with a mean of 33.2 years. The length of marriage ranged from 1 to 34 years with a mean of 11 years. Given the nature of the data source, conclusions reached from the analysis of such data must be regarded as tentative.

Analysis of Data Live Births With respect to fertility, the number of children ever born will be used as the first dependent variable. The independent variables include ethnicity, place of residence, couple's education level and household income. The covariates are woman's work status, woman's current age and length of marriage duration. The multiple classification analysis shows that the independent variables and the covariates could explain about 25% of the variance of the dependent variable. After being adjusted, both the independent variables and the covariates could explain about 61% of the variance of the dependent variable. As for the order of the degree of association of the independent variables with the dependent variable, woman's education is found to be the most significant factor, followed by husband's education, ethnicity, household income and place of residence, respectively. However, when adjusted for the effects of other independent variables, the order of significance of these variables changes. While woman's education still remains the most significant factor, ethnicity becomes the second most important, followed by husband's education, household income and current residence. When controlling for both the independent variables and the covariates, the explanatory power of these variables increases and ethnicity becomes the most significant factor followed by woman's education, husband's education, type of residence and level of household income (see Table 5).

21

Table 5;

Analysis of Variance of Children Ever Born

Source of Variation

Significance ofF

Sum of Squares

DF

Mean Square

F

Main Effects Ethnicity Place of Residence Woman's Education Husband's Education Household Income

1,267.496 86.837 3.464 226.718 68.240 22.726

11 2 1 3 3 2'

115.227 43.419 3.464 75.573 22.747 11.363

41.758 15.735 1.255 27.387 8.243 ,.4.118

0.000 0.000 0.263 0.000 0.000 0.0 11 'Z

Covariates

1,843.979 4.324 1.955 479.089

3 1 1 1

614.659 4.324 1.955 479.089

222.751 1.567 0.709 173.621'

0.000 0.211 0.400 0.000

Explained

3,111.475

14

222.248

80.542

Residual

1,959.174

710

2.759

Total

5,070.648

724

7.004

Woman's Work Status Woman's Current Age Marriage Duration

0.0

Covariate Raw Coefficient Woman's Work Status = -0.224 Woman's Current Age = -0.015 Marriage Duration = 0.229 Number of· Cases

868

The results could be interpreted to mean that, other things being equal, ethnicity is the most significant factor affecting fertility. The level of education that women had attained is also another factor related to fertility. The extent and the direction of all the independent variables are shown in Table 6.

Ethnicity The data also shows thatMoslems had the largest number of children ever born with the Chinese at the other end and the Thais in the middle. However,

22

Table 6:

Multiple Classification Analysis of Children Ever Born

Variable and Category

N

Unac:ljuated Dcv'n

Ethnicity Thai Chinese Moslem

223 236 266

-0.57 -0.78 1.18

Place of Residence Rural Fringe 117 Urban 608

-0.04 0.01

Woman's Education 0-3 years 66 4 years 347 130 5-12 years 13 years+ 182

1.67 0.88 -0.82 -1.70

Husband's Education 0-3 years 56 4 years 290 5-12 years 152 13 years+ 227

1.55 1.06 -0.55 -1.3 7

Household Income 1,100 and less 74 1,101-3,800 346 3,801+ 305

Eta

0.34

0.01

0.46

0.43

0.97 0.50 -0.81 0.27

Adjusted for Independents Dcv'n

-0.13 -0.49 0.54

-0.16 0.03

1.44 0.61 -0.52 -1.33

0.83 0.31 -0.44 -0.30

-0.16 -0.21 0.28

Beta

Adjusted for Independents and Covariatcs Dcv'n

-0.12 -0.22 0.30 0.17

0.03

0.35

0.15

0.09

-0.26 0.05

0.23 0.13 -0.25 -0.16

0.00 0.15 -0.21 -0.05

-0.11 0.04 -0.02

Beta

0.09

0.04

0.07

0.05

0.02

Multiple R 2

0.250

0.614

Multiple R

0.500

0.783

23

when controlling for differences in current residence and woman's income, the differences among the three ethnic groups become less. With further control for differences in female labour force participation, marriage duration and woman's current age, the differences become even smaller. This means that the seemingly large differences in fertility among the three ethnic groups are largely due to their social and economic differences.

Place of Residence It is interesting to note that women in the urban areas of Bangkok averaged a slightly larger number of live births than women living on the rural fringe of Bangkok. This difference could be explained by the increasing suburbanization of Bangkok. People who live in the suburban areas are better educated and seek better housing conditions. Therefore, they are likely to aspire for better living and hence control their family size.

Education The data also shows a negative relationship between fertility and the level of education attained by both husband and wife. Those who had fewer years of education had a larger family size than those who had more years of schooling. When controlling for the independent variables and for both the independents and the covariates, the direction of the relationship remains unchanged, even though the differences become smaller.

Household Income Regarding the relationship between the level of household income and the number of children ever born, a negative association between them is found. Those who had lower incomes averaged a larger family size than those who had higher incomes. When controlling for the differences in all other independent variables, a curvilinear relationship is found. Those in the middle income group had the smallest family size while those in the highest income group had the largest. However, when controlling for both the independents and the covariatcs, those in the middle income group had the largest family size and those in the highest income group averaged a slightly larger family size than those in the lowest

24

mcome group. It should be noted that when controlling for the differences in the independent variables and for both the independents and the covariates, the income fertility differentials become less and the direction of the relationship also changes in the manner explained above. In brief, the fmdings show that ethnicity is the most significant factor related to fertility as measured by the number of live births. It is followed in order of significance by woman's education, husband's education, place of residence and household income. Most of the directions of the relationship between the dependent variables and the independent variables are readily explicable in the Thai social and economic context.

Desired Family Size

In addition to the actual family size, an effort was made to study the differences in family size as desired by people in different social, economic and ethnic groups. It was found that woman's education is the variable that is most highly associated with desired family size. Following this in order of the degree of ~sociation are husband's education, ethnicity, household income and place of residence. The order of degree of association does not change when controlling for the independent variables, but each independent variable is less associated with the dependent without the control. However, the .order of the degree of association changes slightly when controlling for both the independents and the covariates. While woman's education retains its position, ethnicity and husband's income share the second position, and the level of household income 1s least associated with the dependent variable (see Tables 7 and 8). With respect to ethnic differences in desired family size, the data shows that Moslems desired the largest family size and the Chinese desired the smallest one. The differences between the three ethnic groups become significantly smaller when controlling for the differences in place of residence, woman's education, husband's education and the level of household income. The differences become even less significant when controlling for both the independent variables and the covariates which include female labour force participation, woman's current age and marriage duration. In fact, the differences between the Thai and the Chinese disappeared. However, the Moslems still desired a larger family size than the other two ethnic groups.

25

Table 7:

Analysis of Variance of Desired Family Size

Source of Variation

Sum of Squares

Main Effects Ethnicity Place of Residence Woman's Education Husband's Education Household Income

745.059 34.724 6.035 136.665 38.993 10.267

11 2

Covariates Woman's Work Status Woman's Current Age Marriage Duration

637.565 16.161 12.848 220.748

3 1 1

Explained

Mean Square

F

67.733 17.362 6.035 45.555 12.998 5.134

25.301 6.485 2.254 17.016 4.855 1.918

0.000 0.002 0.134 0.000 0.002 0.148

1

212.522 16.161 12.848 220.748

79.385 6.037 4.799 82.458

0.000 0.014 0.029 0.000

1,382.624

14

98.759

36.890

0.000

Residual

1,841.852

688

2.677

Total

3,224.475

702

4.593

DF

3 3 2

Significance of F

Covariate Raw Coefficient Woman's Work Status = -0.442 Woman's Current Age= -0.038 Marriage Duration 0.159 Number of Cases

868

Place of Residence and Desired Family Size Insofar as the place of residence is concemed, it was found that those in the suburban areas of Bangkok desired a smaller .family size than those in the urban centre. The difference becomes greater when controlling for the independent variables. It becomes even greater when controlling for both the independents and the covariates. This means that there are other factors related to the characteristics of people living in the two areas of residence and these characteristics are more strongly associated with desired family size when the independent variables and the covariates are controlled. Research is needed to understand better the relationship between place of residence and desired family size.

26

Table 8:

MultiEle Classification Analysis of Desired Family Size

Variable and Category

N

Unadjusted Dev'n

Eta

Adjusted for Independents Dev'n

Beta

Adjusted for Independents and Covariates Dev'n

Beta

Ethnicity Thai

220

-0.54

-0.14

-0.10

Chinese

221

-0.48

-0.29

-0.10

Moslem

262

0.86

Rural Fringe

113

-0.12

Urban

590

0.02

66

1.37

1.12

0.40

4 years

343

0.65

0.46

0.12

5-12 years

127

-0.60

-0.39

-0.22

13 years+

167

-1.41

0.31

0.36

0.17 0.13

0.06

Place of Residence -0.26

-0.22 0.02

0.04

0.05 0.05

0.04

Woman's Education 0-3 years

0.45

-1.09

0.34

-0.24

0.10

Husband's Education 0-3 years

56

1.34

0.77

0.29

4 years

288

0.74

0.19

0.07

5-12 years

145

-0.29

-0.23

-0.12

13 years+

214

-1.16

0.42

-0.30

-0.10 0.15

0.06

Household Income 1,100 and less

74

0.76

-0.06

-0.03

1,101-3,800

339

0.40

-0.15

-0.02

3,801+

290

-0.66

0.26

0.19

0.03 0.07

0.01

Multiple R2

0.231

0.429

Multiple R

0.481

0.655

27

Education and Desired Family Size, Desired family size is negatively associated with the level of education attained by the couples. Insofar as woman's education is concerned, women who had less than four years of education desired 4.69 children while those who had more than twelve years of schooling desired only 2.16 children, a difference of 2.53 children. When controlling for differences in ethnicity, place of residence, husband's education and the level of household income, the differences in the average desired family size among the women in various educational levels become less. When additional control is made on marriage duration, woman's current age and female labour force status, the differences become even less. Those who had the least education differed from those with the highest education by only .64 child. Husband's education is also negatively associated with desired family size. When controlling for the independent variables, the differences in the average family size desired by men with different levels of education are much less than in the case of woman's education. The difference between the least educated husband and the most educated husband in their desired family size is only .39 child. None the less, the direction of the relationship between the two variables remains the same. Thus, it could be concluded that the level of education attained by women and their spouses is a significant factor related to the desire for a given family size.

Desired Family Size and Household Income With respect to the association of the level of household income with desired family size, the data shows that, on the average, those with a lower household income tend to have a larger desired family size than those with a higher household income. However, when controlling for the differences in the ethnic composition, place of residence and couple's education, a positive association between the two variables is found. Further control for the differences in female labour force participation, woman's current age and marriage duration shows that the positive association between the two variables remains but the differences m the desired number of children among the three income groups become less. In fact, among the relationships between the dependent variable and the independent variables, when analysis is controlled for all independent variables

28

and the covariates, the level of household income has the least association with the dependent variable. In other words, the level of household income is the least important variable in explaining desired family size among the chosen independent variables.

Woman's Age at First Marriage Another factor that is usually considered to be a significant factor affecting fertility is woman's age at first marriage. Again, among all the independent variables, woman's level of education has the most bearing on the age at first marriage. Even when controlling for all other independent variables and the covariates, female education still ranks the same. Next in order of the degree of association are husband's education, household income and ethnicity. The hierarchical order of these independent variables does not change when controlling for all other independent variables. However, when controlling for differences in female labour force participation, woman's current age and duration of marriage, ethnicity becomes the second most important factor after woman's education and place of residence becomes the variable that is least associated with woman's age at first marriage (see Tables 9 and 10).

Table 9:

Analysis of Variance of Wife's Age at First Marriage

Source of Variation

F

Significance ofF

435.543 96.825 24.182 190.353 43.248 37.639

34.545 7.680 1.918 15.098 3.430 2.985

0.000 0.001 0.167 0.000 0.017 0.051

1 1

549.020 549.019

43.546 43.546

0.000 0.000

5,339.996

12

445.000

35.295

0.000

9,027.230

716

12.608

14,367.227

728

19.735

Sum of Squares

DF

4,790.977 193.651 24.182 571.058 129.744 75.278

11 2 1 3 3 2

549.020 549.019

Explained Residual

Main effects Ethnicity Place of Residence Woman's Education Husband's Education Household Income Co variates Woman's Current Age

Total

Covariate Raw Coefficient Woman's Current Age -0.118 Number of Cases 868

Mean Square

29

Table 10:

MultiEle Classification Analysis of Wife's Age at First Marriage

Variable and Category

N

Unadjusted

Adjusted for Independents

Dev'n

Dev'n

Eta

Beta

Adjusted for Independents and Covariates Dev'n

Beta

Ethnicity Thai Chinese Moslem

226 237 266

0.99 1.72 -2.37

117 ·612

0.10 -0.02

0.41

-0.08 0.81 -0.66

0.14

-0.07 0.85 -0.70

0.15

Place of Residence Rural Fringe Urban

0.01

0.43 -0.08

0.32 -0.06 0.04

0.03

Woman's Education 0-3 years 4 years 5-12 years 13 years+

67 347 131 184

-0.99 -2.05 0.65 3.76

57 290 154 228

-1.15 -2.27 0.36 2.93

74 348 307

-1.89 -1.56 2.22

0.54

-0.57 -1.23 0.07 2.47

0.34

-1.15 -1.33 0.24 2.75

0.38

Husband's Education 0-3 years 4 years 5-12 years 13 years+

0.50

-0.28 -0.68 0.59 0.53

0.13

-0.68 -0.67 0.65 0.58

0.14

Household Income 1,100 and less 1,101-3,800 3,801+

0.43

-0.18 -0.40 0.50

0.10

-0.13 -0.21 0.27

0.05

Multiple R 2

0.333

0.372

Multiple R

0.577

0.610

30

Ethnicity and Age at First Marriage On the average, Thai women get married at about 22.3 years of age. Among the three ethnic groups, Moslem women married the earliest and Chinese women the latest. The ages at which the Moslem and Chinese get married are 19.9 and 23.26 years respectively. The results are consistent with the number of live births. These findings confirm the general hypothesis that the earlier the women get married, the greater the number of live births. When controlling for all other independent variables and the covariates, the differences in age at first marriage among the three ethnic groups are much reduced but the general pattern still remains. This implies that as women of different ethnic groups become similar in social and economic aspects of life, they tend to be similar in other aspects of life, including age at first marriage.

Place of Residence and Age at First Marriage It is interesting to note that those who lived m the suburban areas or on the rural fringe of Bangkok married later than those in central Bangkok. The finding is again consistent with the differences in children ever born and desired family size. However, the differences between them are too small to be significant.

Education and Age at First Marriage Insofar as education is concerned, it is generally expected that those who are better educated tend to marry later than those who are less educated. The data generally confirms this, but the relationship is found to be somewhat curvilinear. The relationship remains the same when controlling for the differences in other independent variables and the covariates. The average age at first marriage of those who had 4 years of education or less is about 21 years. Those who had 5 to 12 years of schooling tend to get married at 22.5 years of age and those who had more than 12 years of education, on the average, get married at about 25 years of age. Woman's age at first marriage is also associated with husband's education. Women whose husbands are less educated get married earlier than those whose

31

husbands are better educated. This is probably due to the selection of a spouse from a similar educational background. The association between husband's education and woman's age at first marriage becomes less when controlling for the differences in ethnic composition, place of residence and household income. However, the pattern of the differences remains the same.

Household Income and Age at First Marriage The data also shows that women from lower mcome households tend to marry earlier than those from higher income households. Among the three income groups, those with the lowest income differ from those with the highest income by 4.1 children. However, when controlling for the differences in ethnicity, place of residence, couple's education, female labour force participation, duration of marriage and age composition, the differences become very small -less than 1 child. In brief, among all the independent variables, it appears that couple's education is the most significant factor associated with woman's age at first marriage. Ethnicity is also an important factor, with or without control for the differences in other independent variables and the covariates.

Current Contraceptive Use Among all the dependent variables, contraceptive use is the one least associated with the set of independents and covariates, with R2 of .06. Among the independent variables, ethnicity is the variable that is most associated with the dependent variable, followed by place of residence, husband's education, household income and woman's education, respectively. Among the covariates, the number of children ever born explained the variance of the dependent variable more than any other covariate. The findings imply that there exist significant differences in contraceptive use among different ethnic groups, among people who live in different places of residence, and among women with different actual family sizes (see Tables 11 and 12).

32

Table 11:

Analysis of Variance of Current Use of Contraception

Source of Variation

Significance ofF

Sum of Squares

DF

Mean Square

8.196

11

0.745

3.949

0.000

Ethnicity

5.129

2

2.565

13.591

0.000

Place of Residence

2.084

1

2.084

11.044

0.001

Woman's Education

0.226

3

0.075

0.399

0.754

Husband's Education

0.483

3

0.161

0.854

0.465

Household Income

0.437

2

0.219

1.158

0.315

1.031

4

0.258

1.367

0.244

Woman's Work Status

0.108

1

0.108

0.573

0.449

Woman's Current Age

0.023

1

0.023

0.121

0.728

Marriage Duration

0.036

1

0.036

0.193

0.660

Children Ever Born

0.594

1

0.594

3.150

0.076

9.228

15

0.615

3.260

0.000

Residual

132.277

701

0.189

Total

141.504

716

0.198

Main Effects

Co variates

Explained

Covariate Raw Coefficient Woman's Work Status = 0.036 Woman's Current Age = 0.002 Marriage Duration

-0.002

Children Ever Born

0.018

Number of Cases

868

F

33

Table 12:

Multiple Classification Analysis of Current Use of Contraception

Variable and Category

N

Unadjusted Dev'n

Ethnicity Thai Chinese Moslem

221 235 261

0.04 -0.12 0.08

Place of Residence Rural Fringe Urban

114 603

-0.10 0.02

Woman's Education 0-3 years 4 years 5-12 years 13 years+

63 343 130 181

-0.03 0.03 -0.02 -0.02

Husband's Education 0-3 years 4 years 5-12 years 13 years+

54 286 152 225

-0.06 0.04 -0.06 -0.00

Eta

Adjusted for Independents Dev'n

0.19

0.10

0.06

73 341 303

-0.13 0.02

0.03 0.02 -0.00 -0.05

0.12

0.07

0.12 0.02

0.02 0.02 0.00 -0.05

0.12

0.06

-0.07 -0.00 -0.03 0.04 0.08

-0.04 -0.03 0.04 0.02

0.21

0.22

-0.06 -0.00 -0.04 0.04

0.02 -0.01 -0.00

Dev'n Beta

0.04 -0.13 0.08

0.04 -0.13 0.09

0.09 Household Income 1,100 and less 1,101-3,800 3,801+

Beta

Adjusted for Independents and Covariates

0.07

-0.04 -0.02 0.04 0.08

0.07

Multiple R 2

0.058

0.065

Multiple R

0.241

0.255

34

Factors Mfecting the Number of Live Births by Ethnic Groups In order to understand how factors affect fertility differentially in each

specific ethnic group, a separate analysis is made for each of them. Among the factors chosen to be the independent variables are place of residence, woman's education, husband's education, household income and female labour force participation, and the covariates are woman's current age and marriage duration.

The Thais Among Thai women, the set of independent variables ex;plain 27% of the variance of the dependent variable. The data also shows that the level of education attained is the factor most related to fertility. The relationship is negative as ex;pected. Better educated women averaged a smaller family size than less educated women. A difference of 3.4 children is found between the least and the most educated women. The difference becomes slightly ·less when controlling for place of residence, husband's education, household income and female labour force participation. The difference becomes much less when controlling for differences in women's age composition and their marriage duration. In fact it is reduced to only 1.15 children (see Tables 13 and 14). Next in order of degree of association with the dependent variable are husband's education, woman's work status, household income and place of residence. As for husband's education, it was found that women with better educated husbands averaged a smaller family size than those with less educated husbands. The pattern of the relationship remains the same, although the differences become smaller, when controlling for the differences in other independent variables; the differences become even smaller when women's current age and marriage duration are controlled. It is interesting to note that woman's work status initially shared the same rank as husband's education in terms of degree of association with fertility. However, when controlling for the differences in other independent variables and the covariates, woman's work status ranks behind husband's education though it remains statistically significant. It should be noted that the difference between working and nonworking women is less than that between the least and the most educated women.

35

Table 13:

Analysis of Variance of Children Ever Born (Thai)

Source of Variation

Sum of Squares

DF

Mean Square

F

Significance ofF

Main. Effects

245.751

10

24.575

13.569

0.000

Place of Residence

0.102

1

0.102

0.056

0.813

Woman's Education

67.679

3

22.560

12.456

0.000

Husband's Education

17.263

3

5.754

3.177

0.025

2.991

2

1.495

0.826

0.439

10.137

1

10.137

5.597

0.019

271.549

2

135.775

74.965

0.000

2.035

1

2.035

1.123

0.000

95.427

1

95.427

52.688

0.000

Explained

517.300

12

43.108

23.801

0.000

Residual

380.346

210

1.811

Total

897.646

222

4.043

Household Income Woman's Work Status

Covariates Woman's Current Age Marriage Duration

Covariate Raw Coefficient Woman's Current Age Marriage Duration

Number of Cases

327

=

-0.027 0.189

36

Table 14:

Multiple Classification Analysis of Children Ever Born (Thai)

Variable and Category

N

Unadjusted Dev'n

Eta

Adjusted for Independents Dev'n

Beta

Adjusted for Independents and Covariates Dev'n

Beta

Place of Residence 59

0.25

164

-0.09

0-3 years

16

2.46

2.22

0.89

4 years

64

0.80

0.87

0.31

5-12 years

51

-0.15

-0.22

-0.21

13 years+

92

-0.90

0-3 years

14

2.23

0.64

0.75

4 years

49

0.48

-0.53

-0.40

5-12 years

45

0.25

0.05

0.11

13 years+

115

-0.58

1,100 and less

14

0.59

0.25

0.34

1,101-3,800

75

0.40

-0.17

0.02

134

-0.29

Rural Fringe Urban

-0.01

-0.04 0.07

0.01

0.01

0.00

0.00

Woman's Education

0.49

-0.86

0.47

-0.25

0.17

Husband's Education

0.37

0.13

0.15

0.03

0.13

Household Income

3,801+

0.18

0.07

0.06

-0.05

0.05

Woman's Work Status No

96

0.85

Yes

127

-0.65

0.33 0.37

-0.25

0.06 0.14

-0.04

0.02

Multiple R 2

0.274

0.576

Multiple R

0.523

0.759

37

In terms of income fertility differentials, the data shows that women in a higher income group tend to average a smaller family size than women in a lower income group. The pattern of differences between different income groups remains the same but the differences are significantly reduced and are not statistically significant. Although there is a difference in fertility with respect to place of residence, it is too small to be statistically significant.

The Chinese The same set of independent variables have more explanatory power for the Chinese than for the Thai women. They explain about 3 7% of the dependent variable. The order of the rank of the independent variables, before controlling for the independent variables and the covariates, is similar to that of the Thais. However, with control, the rank order changes significantly. Husband's education is the most significant variable followed by woman's education and female labour force participation (see Tables 15 and 16). Table 15:

Analysis of Variance of Children Ever Born (Chinese)

Source of Variation

Sum of Squares

DF

Mean Square

F

Significance of F

Main Effects Place of Residence Woman's Education Husband's Education Household Income Woman's Work Status

310.667 0.008 43.388 44.626 2.952 0.664

10 1 3 3 2 1

31.067 0.008 14.463 14.875 1.476 0.664

18.550 0.005 8.636 8.882 0.881 0.396

0.000 0.945 0.000 0.000 0.416 0.530

Covariates Woman's Current Age Marriage Duration

296.494 0.398 88.640

2 1 1

148.24 7 0.398 88.640

88.521 0.238 52.928

0.000 0.626 0.000

Explained

607.161

12

50.597

30.212

0.000

Residual

373.461

223

1.675

Total

980.622

235

4.173

Covariate Raw Coefficient Woman's Current Age = -0.012 Marriage Duration = 0.189 Number of Cases

267

38

Table 16:

Multiple Classification Analysis of Children Ever Born (Chinese)

Variable and Category

N

Unadjusted

Adjusted for Independents

Dev'n

Dev'n

Place of Residence Rural Fringe Urban

18 218

-0.44 0.04

Woman's Education 0-3 years 4 years 5-12 years 13 years+

30 63 54 89

1.76 0.86 -0.05 -1.17

22 44 64 106

2.18 1.12 0.04 -0.94

7 89 140

1.44 0.71 -0.52

120 116

0.75 -0.78

Husband's Education 0-3 years 4 years 5-12 years 13 years+

Household Income 1,100 and less 1,101-3,800 3,801 +

Woman's Work Status No Yes

Eta

0.06

0.52

0.51

0.31

0.38

-0.02 0.00

1.06 0.72 0.01 -0.87

Beta

Adjusted for Independents and Covariates Dev'n

0.15 -0.01 0.00

0.37

1.54 0.42 -0.20 -0.37

0.02

0.17 0.32 -0.09 -0.23

0.07 -0.08

0.11

0.45 0.37 -0.14 -0.16 0.28

0.04 -0.19 0.12

Beta

0.07

0.04

0.12

0.09 -0.06 0.04

0.14 -0.14

0.02

0.07

Multiple R 2

0.317

0.619

Multiple R

0.563

0.787

39

The data shows a significant negative relationship between Chinese fertility and husband's educational level, even with control for the differences in place of residence, woman's education, household income, female labour force participation, woman's current age and woman's duration of marriage. However, when controlling for other independent variables and the covariates, the fertility differentials become much less. A significant negative relationship is also found between the dependent variable and woman's education. The Chinese who live on the rural fringe of Bangkok are also found to have a smaller family size than their counterparts who li-xe in central Bangkok with or without control for other independent variables. \powever, the difference ~,, between the two groups is small. When further control is ,made for the differences in woman's age composition and duration of marriage, the 'direction of association is reversed. No conclusion can be made as the findings are not statistically significant. With respect to woman's work status, it has been found that women who worked outside the home averaged a smaller family size than those who did not, regardless of whether or not adjustments were made for either the independent variables or the covariates or both. However, the difference between them is not statistically significant. The relationship between the level of household income and fertility has also been found to be negative butJ with control for other independent variables or the covariates or both, the relationship becomes curvilinear or a U-shape type. None of the differences is statistically significant.

The Moslem The same set of independent variables least explain the variance of the dependent variable for the Moslems compared to other ethnic groups. The couple's level of education is statistically significant and most related to the dependent variable, with husband's education being more important than woman's education. The relative significance of husband's education over woman's education is similar to that of the Chinese but opposite to that of the Thais. To some extent, it can be said that Thai women have higher status than their Chinese and Moslem counterparts with the same level of education insofar as family size is concerned. Even though a negative relationship between fertility and couple's education is found, the relationship is somewhat curvilinear. The differences in fertility among women with different levels of education become more pronounced

40

when adjustments are made for differences in the independent variables. A similar relationship is found between fertility and husband's level of education (see Tables 17 and 18).

Table 17:

Analysis of Variance of Children Ever Born (Moslem)

Source of Variation

Sum of Squares

DF

Mean Square

F

6.494 0.841. 6.961 5.532 4.832 0.730

Significance ofF

0.000 0.360 0.000 0.001 0.009 0.394

Main Effects Place of Residence Woman's Education Husband's Education Household Income Woman's Work Status

271.461 3.514 87.288 69.375 40.393 3.052

10 1 3 3 2 1

27.146 3.514 29.096 23.125 20.196 3.052

Covariates Woman's Current Age Marriage Duration

1,277.980 0.323 263.175

2 1 1

638.990 0.323 263.175

152.868 0.077 62.961

0.000 0.781 0.000

Explained

1,549.441

12

129.120

30.890

0.000

Residual

1,057.540

253

4.180

Total

2,606.981

265

9.838

Covariate Raw Coefficient Woman's Current Age= -0.010 Marriage Duration 0.273 Number of Cases

274

With regard to the association between fertility and the level of household income, the relationship between the two variables is of J -shape curvilinear and remains unchanged, even when controlled for differences in other independent variables. However, the differences almost disappear when the differences in other independent variables and the covariates are controlled.

41

Table 18:

Multiple Classification Analysis of Children Ever Born (Moslem)

Variable and Category

N

Unadjusted

Adjusted for Independents

Dev'n

Dev'n

Place of Residence Rural Fringe Urban

40 226

-0.25 0.05

Woman's Education 0-3 years 4 years 5-12 years 13 years+

20 220 25 1

1.37 0.12 -2.15 -0.23

Husband's Education 0-3 years 4 years 5-12 years 13 years+

20 197 43 6

0.47 0.33 -1.67 -0.56

Household Income 1,100 and less 1,101-3,800 3,800+

53 182 31

0.09 -0.11 0.48

Woman's Work Status No Yes

237 29

0.11 -0.88

Eta

0.03

0.24

0.24

0.06

0.10

-0.28 0.05

1.32 0.08 -1.75 -0.08

0.02 0.27 -1.23 -0.27

-0.14 -0.15 1.11

0.04 -0.32

Beta

0.04

0.21

0.18

0.13

0.04

Adjusted for Independents and Covariates Dev'n

-0.70 0.12

0.29 -0.00 -0.19 -0.50

-0.79 0.15 -0.46 0.94

-0.23 0.07 -0.00

0.03 -0.21

Beta

0.09

0.03

0.11

0.04

0.02

Multiple R 2

0.104

0.594

Multiple R

0.323

0.771

42

Similar to the Thais and the Chinese, the Moslems who live on the rural fringe of Bangkok average a smaller family size than those who live in central Bangkok, and those who work outside home have a smaller family size. than the nonworking women. However, the differences are small and not statistically significant. The findings seem to imply that for the Moslem there may be other factors or variables associated with fertility than the set of independent variables and covariates. In brief, fertility of the three ethnic groups is most significantly affected by the level of education of the married couples. For the Thai women, woman's education is more important than her spouse's but for the Chinese and the Moslem, the opposite holds; that is, husband's education is more important than woman's education. Perhaps in the Chinese and l\blem groups, men are by custom more domineering than in the Thai group. Apart from that, the relationships between the dependent variables and other independent variables of the three ethnic groups share similar patterns but the relative significance of the independent variables and the covariates varies among them.

Factors Affecting Desired Family Size by Ethnic Groups Closely related to the number of live births is desired family size. It is expected that the same set of independent variables and the covariates will be related to desired family size of each ethnic group in the same manner as they are with the number of children ever born. Among the three ethnic groups, the desired family size of the Chinese 1s most fully explained by the same set of variables and covariates. These variables least explain the Moslem's desired family size. This implies that these variables and covariates are accordingly related to desired family size for each ethnic group.

The Thais For the Thais, woman's education is the most significant factor in explaining the variance of the dependent variable. The level of education 1s

43

negatively related to desired family size as it is to the number of live births. Woman's work status also negatively affects woman's desired family size. Those who work outside home desire a smaller family size than those who do not. Similar to woman's education, husband's education is negatively re~ated to the dependent variable. It is also found that those who live on the rural fringe of Bangkok desire a smaller family size than those who live in central Bangkok but the difference is not statistically significant. Household income is curvilinearly related to desired family size. The middle income group desires the smallest family size. However, the relationship between the two variables is not statistically significant. Among the covariates, marriage duration is more significant than woman's current age in explaining the variation of the Thai woman's desired family size (see Tables 19 and 20).

Table 19:

Analysis of Variance of Desired Family Size (Thai)

Significance of F

Source of Variation

Sum of Squares

DF

Mean Square

F

Main Effects Place of Residence Woman's Education Husband's Education Household Income Woman's Work Status

188.966 2.826 43.681 19.729 1.145 11.814

10 1 3 3 2 1

18.897 2.826 14.560 6.576 0.573 11.814

11.713 1.752 9.025 4.076 0.355 7.323

0.000 0.187 0.000 0.008 0.702 0.007

Covariates Woman's Current Age Marriage Duration

83.846 8.655 49.265

2 1 1

41.923 8.655 49.265

25.985 5.365 30.537

0.000 0.022 0.000

Explained

272.811

12

22.734

14.092

0.000

Residual

333.957

207

1.613

Total

606.768

219

2.771

Covariate Raw Coefficient Woman's Current Age = -0.055 Marriage Duration 0.137 Number of Cases

327

44

Table 20:

Multiple Classification Analysis of Desired Family Size (Thai)

Variable and Category

Adjusted for N

Unadjusted Dev'n

Eta

Independents Dev'n

Beta

Adjusted for Independents and Covariates Dev'n

Beta

Place of Residence Rural Fringe Urban W~man's Education 0-3 years 4 years 5-12 years 13 years+

Husband's Education 0-3 years 4 years 5-12 years 13 years+

Household Income 1,100 and less 1,101-3,800 3,800+

Woman's Work Status No Yes

58 162

16 64 51 89

-0.01 0.01

2.34 0.47 -0.03 -0.74

14 49 45 112

2.18 0.23 0.32 -0.51

14 75 131

0.33 0.34 -0.23

95 125

0.75 -0.57

0.01

0.50

0.41

0.17

0.39

-0.20 0.07

1.95 0.50 -0.18 -0.61

0.80 -0.48 0.18 0.04

0.12 -0.11 0.05

0.36 -0.27

0.07

0.43

0.19

0.05

0.19

-0.16 0.06 0.06

1.26 0.14 -0.18 -0.22

0.86 -0.43 0.25 -0.02

0.19 -0.05 0.01

0.19 -0.15

0.23

0.19

0.03

0.10

Multiple R 2

0.311

0.450

Multiple R

0.558

0.671

45

The Chinese As for the Chinese, it has been found that the levels of education

attained by both husband and wife are the most significant factors in explaining the variance of family size desired by them. The relationship between the two variables is negative. The data also shows that Chinese women who work outside the home desire a smaller family size than those who do not but the difference is not statistically significant. A negative association between desired family size and household income is also found but again it is not statistically significant. As in the Thai case, the marriage duration of Chinese women is more significant than their current age in explaining variance in desired family Size. Although there are differences in desired family size between those living in central Bangkok and those living in suburban areas of Bangkok and between different income classes, the differences are small and not statistically significant (see Tables 21 and 22).

Table 21:

Analysis of Variance of Desired Family Size (Chinese)

Source of Variation

Sum of Squares

DF

Mean Square

F

Significance of F

12.32!1 0.860 4.609 3.376 1.187 2.080

0.000 0.355 0.004 0.019 0.307 0.151

Main Effects Place of Residence Woman's Education Husband's Education Household Income Woman's Work Status

232.966 1.626 26.143 19.146 4.489 3.933

10 3 3 2 1

23.297 1.626 8.714 6.382 2.245 3.933

Co variates Woman's Current Age Marriage Duration

111.159 10.967 63.686

2 1 1

55.579 10.967 63.686

29.399 5.801 33.687

0.000 0.017 0.000

Explained

344.125

12

28.677

15.169

0.000

Residual

393.226

208

1.891

Total

737.350

220

3.352

Covariate Raw Coefficient Woman's Current Age= -0.066 Marriage Duration = 0.163 Number of Cases

267

1

46

Table 22:

Multiple Classification Analysis of Desired Family Size (Chinese)

Variable and Category

N

Unadjusted Dev'n

Place of Residence Rural Fringe Urban

15 206

-0.02 0.00

Woman's Education 0-3 years 4 years 5-12 years 13 years+

30 62 52 77

1.51 0.75 -0.14 -1.10

Husband's Education 0-3 years 4 years 5-12 years 13 years+

22 44 59 96

1.60 1.05 0.12 -0.92

7 86 128

1.77 0.61 -0.51

116 105

0.71 -0.78

Household Income 1,100 and less 1,101-3,800 3,801+

Woman's Work Status No Yes

Eta

0.00

0.52

0.50

0.34

0.41

Adjusted for Independents Dev'n

0.82 -0.02

0.90 0.51 -0.09 -0.70

0.93 0.42 -0.12 -0.33

0.67 -0.13 0.05

0.18 -0.20

Beta

0.05

Adjusted for Independents and Covariates Dev'n

0.41 -0.03

0.44 0.25 -0.13 -0.28 0.32

0.23

0.08

0.10

0.27 0.36 -0.11 -0.16

0.55 -0.15 0.07

0.22 -0.24

Beta

0.06

0.15

0.12

0.08

0.13

Multiple R2

0.316

0.467

Multiple R

0.562

0.683

47

The Moslems With respect to Moslem women, husband's education is the most significant factor related to desired family size. However, the pattern of the relationship is not clear. It is interesting to note that there tends to be a positive relationship between household income and desired family size but the result is not statistically significant. The data also shows that Moslem women who work outside the home desire a smaller family size than those who do not and those who live in the suburban areas also desire a smaller family than those who live in central Bangkok. The patterns are similar to those of the Thai and the Chinese. Again the differences are small and not statistically significant (see Tables 23 and 24 ).

Table 23:

Analysis of Variance of Desired Family Size (Moslem)

Source of Variation

Significance ofF

Sum of Squares

DF

Main Effects Place of Residence Woman's Education Husband's Education Household Income Woman's Work Status

119.970 1.646 22.488 32.049 16.100 5. 788

10 1 3 3 2 1

11.997 1.646 9.829 10.683 8.050 5.788

2.948 0.404 2.415 2.625 1.978 1.422

0.002 0.525 0.067 0.051 0.141 0.234

Covariates Woman's Current Age Marriage Duration

440.552 0.419 95.518

2 1 1

220.276 0.419 92.518

54.122 0.103 22.732

0.000 0.749 0.000

Explained

560.522

12

46.710

11.477

0.000

Residual

1,013.427

249

4.070

Total

1,573.949

261

6.030

Covariate Raw Coefficient Woman's Current Age = -0.011 0.167 Marriage Duration Number of Cases

274

Mean Square

F

48

Table 24:

Multiple Classification Analysis of Desired Family Size (Moslem)

Variable and Category

N

Unadjusted

Adjusted for Independents

Dev'n

Dev'n

Place of Residence Rural Fringe Urban

40 222

-0.20 0.04

Woman's Education 0-3 years 4 years 5-12 years 13 years+

20 217 24 1

0.68 0.11 -1.47 -1.42

Husband's Education 0-3 years 4 years 5-12 years 13 years+

20 195 41 6

0.53 0.19 -1.13 -0.26

Household Income 1,100 and less 1,101-3,800 3,801 +

53 178 31

0.09 -0.08 0.29

Woman's Work Status No Yes

234 28

0.10 -0.82

Eta

0.03

-0.19 0.03

0.52 0.08 -1.13 -1.05 0.20

0.35 0.15 -0.87 0.04 0.20

0.05

0.12

-0.07 -0.10 0.70

0.05 -0.45

Beta

Adjusted for Independents and Covariates Dev'n

-0.41 0.07 0.03

-0.08 0.03 -0.19 -1.23 0.16

0.15

0.10

0.06

-0.11 0.08 -0.43 0.81

-0.10 0.02 0.07

0.05 -0.42

Beta

0.07

0.04

0.09

0.02

0.06

Multiple R 2

0.076

0.356

Multiple R

0-276

0.597

49

In brief, it can be said that the same set of independent variables and covariates are related to desired family size in differing degrees for the three ethnic groups. While woman's education is the most important factor associated with desired family size for the Thais and the Chinese, it is husband's education that is most important for the Moslem. The impact of female labour force participation on the dependent variable is the greatest among the Moslems, while it is greater among the Chinese than among the Thais.

Factors Affecting Age at First Marriage by Ethnic Groups A factor that is generally found to be related to the number of live births and desired family size is woman's age at first marriage. The age at which women get married is affected by a host of economic and social factors. Among these factors are place of residence, woman's education, husband's education, household income and woman's current age. Again one would expect that these variables are variably related to the age at first marriage of different ethnic groups.

The Thai The data shows that the set of independent variables and covariates explain about 32% of the variance of the age at first marriage of the Thai women. Among the independent variables, the level of woman's education is one of two factors related to woman's age at first marriage. The other factor is woman's current age. There is a positive relationship between these and woman's age at first marriage, with or without control for other independent variables and the covariates. No pattern of association is found between woman's age at first marriage and husband's education. A U-curve relationship is found between the dependent variable and household income, although women in the highest income group tend to marry later than those in the lowest income group (see Tables 25 and 26).

50

Table 25:

Analysis of Variance of Wife's Age at First Marriage (Thai)

Sum of Squares

Source of Variation

DF

Mean Square

F

Significance ofF

1,066.877

9

118.542

9.193

0.000

Place of Residence

28.081

1

28.081

2.178

0.141

Woman's Education

400.956

3

133.652

10.364

0.000

Husband's Education

60.926

3

20.309

1.575

0.196

Household Income

73.234

2

36.617

2.840

0.061

238.686

1

238.686

18.510

0.000

238.685

1

238.685

18.510

0.000

Explained

1,305.563

10

130.556

10.124

0.000

Residual

2,772.484

215

12.895

Total

4,078.047

225

18.125

Main Effects

Covariates Woman's Current Age

Covariate Raw Coefficient Woman's Current Age = 0.150

Number of Cases

327

The Chinese The same set of independent variables and covariates explain a greater proportion of the variance of the age at first marriage of the Chinese women rather than of the Thai women. Women's education is positively related to the dependent variable. So is their husband's education. It is interesting to note that the level of household income is also positively related to age at first marriage

51

Table 26:

Multiple Classification Analysis of Wife's Age at First Marriage (Thai)

Variable and Category

N

Unadjusted Dev'n

Place of Residence Rural Fringe Urban

Woman's Education 0-3 years 4 years 5-12 years 13 years+

Husband's Education 0-3 years 4 years 5-12 years 13 years+

Household Income 1,100 and less 1,101-3,800 3,801 +

59 167

17 64 52 93

15 49 46 116

14 77 135

-0.47 0.17

-1.79 -2.61 -0.09 2.18

-2.00 -2.35 0.04 1.23

-0.48 -1.86 1.11

Eta

0.07

0.48

0.35

0.33

Adjusted for Independents Dev'n

Beta

0.64 -0.23 0.09

-1.63 -2.70 -0.14 2.24 0.49

0.53 -0.12 1.09 -0.45 0.14

0.49 -0.93 0.48

0.16

Adjusted for Independents and Covariates Dev'n

0.56 -0.20

-2.82 -2.92 -0.07 2.57

0.55 0.06 0.98 -0.49

0.50 -0.62 0.30

Beta

0.08

0.56

0.14

0.11

Multiple R 2

0.262

0.320

Multiple R

0.511

0.566

52

of the Chinese women, with or without control for other independent variables and the covariates. The relationship could be due to the fact that the kind of work the better off Chinese women do is unfavourable to early marriages {see Tables 27 and 28).

Table 27:

Analysis of Variance of Wife's Age at First Marriage (Chinese)

Source of Variation

Sum of Squares

DF

Mean Square

F

Significance of F

Main Effects 1,390.499 Place of Residence 5.978 Woman's Education 203.082 Husband's Education 28.979 113.639 Household Income

9 l 3 3 2

154.500 t\.978 67.694 9.660 56.819

14.160 0.548 6.204 0.885 5.208

0.000 0.460 0.000 0.449 0.006

Co variates Woman's Current Age

250.634 250.634

1 1

250.634 250.634

22.972 22.972

0.000 0.000

Explained

1,641.133

10

164.113

15.042

0.000

Residual

2,465.801

226

10.911

Total

4,106.934

236

17.402

Covariate Raw Coefficient Woman's Current Age "' 0.158 Number of Cases

267

The Moslem The same set of independent variables and covariates explain less than 10% of the variance of the age at first marriage of the Moslem women. Only one factor, husband's education, is found to be related to the dependent variable at the .01 level of significance. The relationship between the two variables is of a J -curve type. The finding implies that among the Thai Moslem,

53

Table 28:

Multiple Classification Analysis of Wife's Age at First Marriage {Chinese)

Variable and Category

Place of Residence Rural Fringe Urban

Woman's Education 0-3 years 4 years 5-12 years 13 years+

Husband's Education 0-3 years 4 years 5-12 years 13 years+

Household Income 1,100 and less 1,101-3,800 3,801 +

N

Unadjusted

Adjusted for Independents

Dev'n

Dev'n

18 219

1.57 -0.13

30 63 54 90

-1.92 -2.56 -0.49 2.72

22 44 65 106

7 89 141

-2.58 -2.42 -0.90 2.09

-3.70 -2.32 1.65

Eta

Beta

0.56 -0.05 0.11

0.54

Dev'n

0.61 -0.05 0.04

-0.47 -1.66 -0.58 1.67

-1.22 -1.74 -0.52 1.94 0.33

-1.13 -0.53 0.01 0.44 0.47

0.48

Adjusted for Independents and Covariates Beta

0.04

0.38

-1.87 -0.48 0.06 0.55 0.17

0.12

-1.81 -0.68 0.52

-2.28 -0.99 0.74 0.22

0.16

Multiple R 2

0.339

0.400

Multiple R

0.582

0.632

54

men's social and economic characteristics are more crucial than women's. This is probably due to their ethnic customs, family organization and kinship system. More research is certainly needed in this area (see Tables 29 and 30).

Table 29:

Analysis of Variance of Wife's Age at First Marriage (Moslem)

Source of Variation

Sum of Squares

Main Effects

DF

Mean Squares

F

Significance ofF

22.4.?,()9 6.181 8.402 167.671 7.685

9 1 3 3 2

24.930 6.181 2.801 55.890 3.843

1.861 0.461 0.209 4.171 0.287

0.058 0.498 0.890 0.007 0.751

Covariates Woman's Current Age

122.974 122.974

1 1

122.974 122.974

9.178 9.178

0.003 0.003

Explained

347.344

10

34.734

2.592

0.005

Residual

3,416.885

255

13.400

Total

3,764.228

265

14.205

Place of Residence Woman's Education Husband's Education Household Income

Covariate Raw Coefficient Woman's Current Age = 0.08j Number of Cases

274

Factors Affecting Contraceptive Use by Ethnic Groups Adoption of birth control by different ethnic groups is likely to be affected by different sets of variables. Certain variables that may have great influence on the birth control adoption of an ethnic group may not have any influence at all on another ethnic group. The same variable may affect differently the birth control of a different ethnic group.

55

Table 30:

Multiple Classification Analysis of Wife's Age at First Marriage (Moslem}

Variable and Category

N

Unadjusted

Adjusted for Independents

Dev'n

Dev'n

Place of Residence Rural Fringe Urban

40 226

0.40 -0.07

Woman's Education 0-3 years 4 years 5-12 years 13 years+

20 220 25 I

0.05 -0.14 1.26 -0.90

Husband's Education 0-3 years 4 years 5-12 years 13 years+

20 197 43 6

0.80 -0.48 1.33 3.43

Household Income 1,100 and less 1,101-3,800 3,801+

53 182 31

-0.14 0.06 -0.09

Eta

0.04

0.11

0.23

0.02

0.37 -0.07

-0.25 -0.04 0.59 -0.45

0.87 -0.45 1.20 3.39

0.04 0.07 -0.48

Beta

0.04

0.05

0.22

0.05

Adjusted for Independents and Covariates Dev'n

0.20 -0.04

-0.56 -0.06 0.03 -0.51

0.54 -0.46 1.35 3.49

0.00 0.13 -0.79

Beta

0.02

0.09

0.23

0.08

Multiple R 2

0.060

0.092

Multiple R

0.244

0.304

56

As it is felt that place of residence, husband's and wife's education, household income and woman's work status are likely to affect woman's. adoption of birth control, these are chosen as the independent variables. However, one must also take into consideration woman's current age, marriage duration and number of children ever born; therefore these variables are treated here as the covariates. The Thai data shows that, among the Thai women, the level of education attained by them is the single factor that is statistically significant in its relation to the adoption of birth control. Women who are less educated tend to adopt birth control in greater proportion than those who are better educated. This is probably due to the fact that economic hardship on the poorer urban people is greater than that on their better off counterparts. It is interesting to note that, among the covariates, the number of children ever born is significantly related to the adoption of birth control. It should also be noted that the set of independent variables and covariates explain about 10% of the variance of the dependent variables and the result is not statistically significant (see Tables 31 to 36).

Table 31:

Analysis of Variance of Current Use of Contraception (Thai)

Source of Variation

Significance of F

Sum of Squares

DF

Main Effects Place of Residence Woman's Education Husband's Education Household Income Woman's Work Status

2.813 1.346 0.675 0.822 0.043 0.025

10 1 3 3 2 1

0.281 1.346 0.225 0.274 0.022 0.025

1.616 7.723 1.292 1.575 0.145 0.145

0.104 0.006 0.278 0.197 0.704 0.704

Covariates Woman's Current Age Marriage Duration Children Ever Born

0.928 0.109 0.008 0.776

3 1 1 1

0.309 0.109 0.008 0.776

1.778 0.626 0.046 4.458

0.152 0.430 0.830 0.036

Explained

3.741

13

0.288

1.654

0.073

Residual

36.023

207

0.174

Total

39.764

220

0.181

Covariate Raw Coefficient Woman's Current Age = -0.006 Children Ever Born -0.046 Number of Cases 327

Mean Square

Marriage Duration

F

..

-0.002

57

Table 32:

Multiple Classification Analysis of Current Use of Contraception {Thai)

Variable and Category

N

Unadjusted

Adjusted for Independents

Dev'n

Dev'n

Place of Residence Rural Fringe Urban

59 162

-0.14 0.05

Woman's Education 0-3 years 4 years 5-12 years 13 years+

14 64 51 92

0.09 -0.05 -0.04 0.04

13 48 45 115

0.08 -0.12 -0.01 0.04

14 74 133

-0.05 -0.05 0.03

94 127

-0.03 0.02

Husband's Education 0-3 years 4 years 5-12 years 13 years+

Household Income 1,100 and less 1,101-3,800 3,801+

Eta

Beta

-0.14 0.05 0.20

0.11

0.16

0.09

0.20

0.20 0.09 -0.06 -0.06

-0.05 -0.15 0.01 0.07

-0.01 -0.02 0.01

0.20

0.20

0.04

Adjusted for Independents and Covariates Dev'n

-0.13 0.05

0.17 0.06 -0.05 -0.04

-0.08 -0.14 0.01 0.06

-0.02 -0.03 0.02

Beta

0.19

0.15

0.19

0.05

Woman's Work Status No

Yes

-0.02 0.01 0.06

0.03

-0.02 0.02

0.05

Multiple R 2

0.071

0.094

Multiple R

0.266

0.307

58

Table 33:

Analysis of Variance of Current Use of Contraception (Chinese)

Significance of F

Sum of Squares

DF

Mean Square

1.235

10

0.124

0.771

0.656

Place of Residence

0.073

1

0.073

0.454

0.501

Woman's Education

0.276

3

0.092

0.575

0.632

Husband's Education

0.612

3

0.204

1.274

0.284

Household Income

0.129

2

0.064

0.403

0.669

Woman's Work Status

0.013

1

0.013

0.078

0.780

0.244

3

0.081

0.508

0.677

1

0.188

1.177

0.279

Source of Variation

Main Effects

Co variates

F

Woman's Current Age

0.188

Marriage Duration

0.046

1

0.046

0.284

0.594

Children Ever Born

0.025

1

0.025

0.156

0.693

1.479

13

0.114

0.711

0.752

Residual

39.554

247

0.160

Total

41.034

260

0.158

Explained

Covariate Raw Coefficient Woman's Current Age = 0.007 Marriage Duration

-0.004

Children Ever Born

-0.005

Number of Cases

274

59

Table 34:

Multiple Classification Analysis of Current Use of Contraception (Chinese)

Variable and Category

N

Unadjusted

Adjusted for Independents

Dev'n

Dev'n

Place of Residence Rural Fringe Urban

37 224

-0.05 0.01

Woman's Education 0-3 years 4 years 5-12 years 13 years+

19 216 25 1

0.09 0.00 -0.08 0.20

19 194 43 5

-0.07 0.03 -0.11 -0.00

Husband's Education 0-3 years 4 years 5-12 years 13 years+

Eta

0.05

0.10

0.14 Household Income 1,100 and less 1,101-3,800 3,801+

52 178 31

0.04 -0.01 0.00 0.05

Woman's Work Status No Yes

232 29

0.01 -0.05

0.04

-0.04 0.01

0.11 -0.01 -0.04 0.20

Beta

Adjusted for Independents and Covariates Dev'n

-0.05 0.01 0.04

Beta

0.05

0.10 -0.01 -0.04 0.19 0.08

0.09

-0.12 0.03 -0.09 -0.00

-0.10 0.03 -0.09 -0.00 0.13

0.14

0.03 -0.01 0.03

0.03 -0.02 0.03 0.06

0.05

0.00 -0.02

0.00 -0.02 0.02

0.02

Multiple R 2

0.030

0.036

Multiple R

0.174

0.190

60

Table 35:

Analysis of Variance of Current Use of Contraception (Moslem)

Source of Variation

Sum of Squares

DF

Mean Square

Main Effects

4.021

10

0.402

1.766

0.068

Place of Residence

1.103

1

1.103

4.845

0.029

Woman's Education

0.708

3

0.236

1.037

0.377

Husband's Education

0.194

3

0.065

0.284

0.837

Household Income

1.730

2

0.865

3.798

0.024

Woman's Work Status

0.396

1

0.396

1.737

0.189

1.414

3

0.471

2.069

0.105

Woman's Current Age

0.010

1

0.010

0.042

0.837

Marriage Duration

0.046

1

0.046

0.201

0.654

Children Ever Born

1.039

1

1.039

4.561

0.034

5.435

13

0.418

1.836

0.039

Residual

50.325

221

0.228

Total

55.760

234

0.238

Co variates

Explained

Covariate Raw Coefficient Woman's Current Age= 0.002 Marriage Duration

-0.005

Children Ever Bom

0.053

Number of Cases

267

F

Significance of F

61

Table 36:

Multiple Classification Analysis of Current Use of Contraception (Moslem)

Variable and Category

N

Unadjusted

Adjusted for Independents

Dev'n

Dev'n

Place of Residence Rural Fringe Urban

18 217

-0.22 0.02

Woman's Education 0-3 years 4 years 5-12 years 13 years+

30 63 54 88

-0.11 0.01 0.07 ..0.01

Husband's Education 0-3 years 4 years 5-12 years 13 years+

22 44 64 105

-0.11 0.02 -0.02 0.03

Household Income 1,100 and less 1,101-3,800 3,801+

7 89 139

-0.47 -0.02 0.03

120 115

-0.03 0.03

Woman's Work Status No Yes

Eta

0.13

0.11

0.08

0.18

0.06

-0.24 0.02

-0.03 0.06 0.08 0.08

..0.01 0.06 -0.03 -0.01

-0.49 -0.04 0.05

-0.06 0.06

Beta

0.14

0.15

0.06

0.20

0.12

Adjusted for Independents and Covariates Dev'n

-0.25 0.02

Beta

0.15

-0.07 0.04 0.08 -0.05 0.12

-0.07 0.04 -0.02 0.01

-0.49 -0.03 0.05

-0.06 0.06

0.06

0.19

0.13

Multiple R2

0.072

0.097

Multiple R

0.269

0.312

62

The Chinese

As for the Chinese, the level of education attained by both husbands and Wives is not significantly related to the dependent variable. Place of residence and household income are significantly related to the dependent variable. Chinese women who live in the urban centre tend to adopt birth control in a greater proportion than their suburban counterparts. This is probably due to the former's engagement in business and commercial activities that require active participation which is incompatible with childbearing. The data also shows that Chinese women from a higher income household tend to adopt birth control in a greater proportion than these from a poorer income household, with or without control for other independent variables and the covariates. Among the covariates, the number of children ever born is the most significant factor related to the adoption of birth control. Thus it could be said that among the Thais and the Chinese, decision on the adoption of birth control depends on the number of children that had been born. It should be noted that for the Moslem the same set of independent variables and covariates are not statistically significant in their relationship with birth control adoption. They explain about 10% of the variance of the dependent variable. The hypothesis that the adoption of birth control by different ethnic groups is likely to be affected by different sets of variables and the belief that the same variable or set of variables may affect differently birth control adoption are confirmed by the findings that none of the independent variables and covariates is significantly related to the dependent variable. The set of independent variables and covariates explain less than 5% of the variance of the dependent variable. In brief, empirical evidence has shown that each ethnic group's birth control practice is affected differently by different kinds of variables. For the Thais, birth control practice is related to woman's education and the number of live births. For the Chinese, place of residence, the level of household income and the number of children ever born are significantly related to the dependent variable. For the Moslems, none of these variables nor any of the other independent variables and covariates is significantly related to the practice of birth control. More research is needed in this area to find out what factors are most related to the adoption of birth control by the Moslems.

63

BIBLIOGRAPHY

Amado.

"The Chinese m Thailand." (1964): 25-32.

Bulletin De Le Province De Siam

Anderson, John. English Intercourse with Siam in the Seventeenth Century. London: Kegal Paul, Trench, Trubuer and Co., 1890. Boorootpat, Khajatpai. Panha Chonkloomnoi Nai Pratesthai [Minority problems in Thailand). Bangkok: Prae Pittaya, 1972. Bourgois-Pichat, Jean. "An Attempt to Appraise the Accuracy of Demographic Statistics for an Undeveloped Country: Thailand," in Suchart Prasithrathsint, "Economics and Fertility Behaviour of the Rural People in Thailand." Ph.D. Thesis, Brown University, June 1971.

Business Review Bangkok.

April 1977, pp. 200-201.

Chalothom, Thip. "Fertility Levels and Differentials in Thailand." In International Union for the Scient,fic Study of Population, Sydney Conference. Canberra: Australian National University, 196 7. "Differential Fertility of Thai Women." Paper presented at the Seminar on Demographic Research, Population Studies Center, University of Pennsylvania, 1966. Chammanondha, Suvimon. "The Limitation of the Overseas Chinese's Influence During the Reign of King Rama IV." In Thammasat University Journal, Vol. IV, No. 1 Uune-September 1974), pp. 70-86. Chulalongkom University, Faculty of Arts. Report of Malay Language Training Group 1-4, 1964-65. Bangkok: Curusapha Printing Office, 1965.

64

Coughlin, Richard J. Double Identity: The Chinese in Modern Thailand. Kong: University of Hong Kong Press, 1960. Cowgill, Donald 0., et al. Family Planning in Bangkhen, Thailand. Mahidol University, 1969.

Hong

Bangkok:

Decree on Leave with Pay of the Civil and Judicial Officials. Government Gazette. Book 82, Part 6. August 1965, Article 11, Subsection. Dhiravegin, Likhit, ed. Chonkloomnoi Nai Pratesthai [Minorities in Thailand]. Bangkok: Prae Pittaya, forthcoming. Division of Civil Coordination, Department of Provincial Administration. Issues Concerning Provinces in Southern Border. Bangkok: Local Press of the Department of Provincial Administration, 196 7. Fairbank, John K., et al. East Asia: Houghton Mifflin Co., 1965.

The Modern Transformation.

Boston:

Fisher, Charles A. South-East Asia: A Social Economic and Political Geography. New York: E.P. Entlon and Co. Inc., 1964. Fitzgerald, C.P. 1965.

The Third China.

Vancouver:

University of British Columbia,

Gnokin, Alan Edward. Changing Identity: The Assimilation of Chinese in Thailand. Ann Arbor, Mich.: University Microfilms, 1968. Goldstein, Sidney. "Influence of Labor Force Participation and Education on Fertility in Thailand." Population Studies, Vol. XXVI, No. 3 (November 1972). "Religious Fertility Differentials in Thailand." Population Studies, Vol. XXIV, No. 3 (November 1970 ). Goldstein, Sidney, Alice Goldstein and Penporn Tirasawat. The Influence of Labour Force Participation and Education on Fertility in Thailand. Institute of Population Studies. Research Report No. 2. Bangkok: Chulalongkorn University, 1972.

65

Gowing, Peter G. In Mora and Kaeg: The Status of Muslims in the Philippines and in Thailand. Translated by Banthon Ondam. Bangkok: The Komol Keemthong Foundation Press, 1975.

Sankhya: The Indian Gupta, A. Das. "Population Perspective of Thailand." journal of Statistics, Series B, Vol. XXVII, Parts 1-2 ( 1965). Hawley, Amos H. and Visid Prachuabmoh. "Family Growth and Family Planning: Responses to a Family Planning Action Program in a Rural District of Thailand." Demography, Vol. III (1966): 319-331. Heer, David M. "Educational Advance and Fertility Change." Internatt'onal Population Conference {London, 1969). Vol. III. Liege: International Union for the Scientific Study of Population, 1971. Institute of Population Studies. The Potharam Study. Bangkok: Chulalongkorn University, 1971.

Research Report No. 4.

International Bank for Reconstruction and Development. A Publ£c Development Program for Thailand. Baltimore: Johns Hopkins Press, 1959. Kiser, Clyde V. "Educational Differentials in Fertility in Relation to the Demographic Transition." International Population Conference (London, 1969). Vol. III. Liege: International Union for the Scientific Study of Population, 19 71. Knodel, John and Pichit Pitaktepsombati. "Thailand: Fertility and Family Planning." Studies in Family Planning. Vol. IV, No. 9 (September 1973). Kosinanont, Tuanchai. "Uttra Kumlung Lae Kwamtongkarn Kumlungkhon Nai Karntaharn" [Force and manpower demand in military operation]. The First National Seminar on Population, B.E. 2506 (1963), organized by the National Statistical Bureau of Thailand. Landon, K. Perry. Russell, 1951.

The Chinese in Thailand.

New York:

Russell and

Malakul, Pin. The Regional Development of Education [Karn Pattana Kam Suksa Nai Suan Phumi Pkak] . Bangkok: Kurusapha Printing Co., 1963.

66

Malloch, D.E. Siam: Some General Remarks on Its Productions. Baptist Mission Press, 1852.

Calcutta:

McDonald, Noah A. "The Chinese in Siam." In Siam and Laos as Seen by Our American Missionaries. Philadelphia: Presbyterian Board of Publication, 1884. Mitchison, Louis.

The Overseas Chinese.

London:

Bodley Head Ltd., 1961.

National Family Planning Project, Ministry of Public Health; Population Sector, National Economic Development Board; Institute of Population Studies, Chulalongkom University. Thailand: Population Family Planning. Prepared for the Country Profile Series of Reports published by the Population Council, August 1971.

The Population of Thailand.

Bangkok, 1974.

National Statistical Office. The Survey of Population Change, 1964-1965. Bangkok: The Social Science Association of Thailand Press, 1969. Police Department, Ministry of Interior. Annual Report of B.E. 2516 of the Alien Registration and Revenue Division. Bangkok. Population Sector, National Economic Development Board. The Fourth National Economic and Social Development Plan of Thailand (1977-1981). August 1976. Prachuabrmh, V.and James T. Fawcett. "Fertility Control in Rural Thailand: Some Results of a Demonstration Project in Potharam District. International Population Conference (Sydney, 1967). Liege: International Union for the Scientific Study of Population. Prasithrathsint, Suchart. "Family and Household Structure in Thailand." Paper Presented at the Meeting of the Organization of Demographic Associate, December 1972. "Economic and Fertility Behaviour of the Rural People in Thailand." Ph.D. Thesis, Brown University, 1971.

67

Prasithrathsint, Suchart. "A Review of Fertility Studies in Thailand from 1970 to the Present." Journal of Social Sciences Review (Academic Issue), Vol. 14, Nos. 1-2 (January-March 1976 ).

Some Factors Affecting Fertil£ty and Knowledge, Attitude and Practice of Family Planning Among Rural Thai Women. Institute of Population Studies, Working Paper No. 2. Bangkok: Chulalongkom University, 1973. "Female Labour Force Participation and Fertility Behaviour m Rural Thailand." Journal of Social Sciences. Vol. 8, No. 4 (October 1971 ). "Trends and Differentials in Fertility." In United Nations, Economic and Social Commission for Asia and the Pacific (ESCAP}, Country Monograph Series No. 3: Population of Thailand. Bangkok: United Nations, 1976. Punyodyana, Boonsanong. Chinese-Thai Dzfferential Assimilation in Bangkok: An Exploratory Study. Southeast Asia Program, Department of Asia Studies, Data Paper No. 76. Ithaca, N.Y.: Cornell University, 1971. "The Changing Status and Future Role of the Chinese in Thailand." Trends in Thailand I, ed. M. Rajaretnam and Lim So Jean. Singapore: Singapore University Press, 1973. Purcell, Victor. The Chinese in Southeast Asia. Part III. University Press, 1966.

London:

Oxford

Rangkum, Tawec. "Pratesthai Tongkarn Nayobai Prachakom Nai Puchubun Rue Nai Yangrai" [Does Thailand need a JPpulation policy at the present]. The First National Seminar on Population, B .E. 2506 (1963), organized by the National Statistical Bureau of Thailand. Rangsiyokrit, Laman. "The Way to Approach People in the Southern Border." Thesis Presented to the Faculty of Public Administration, National Institute of Development and Administration, 196 7. Sarasas, Phra.

My Country Thailand.

Tokyo:

Maruzen, 1942.

68

Siripirom, Chavalit. "Fertility Differentials in Chiangmai: Socio-Economic Status, Life-Style, and KAP Characteristics." Dissertation Presented at Syracuse University, 1976. Skinner, G. William. Ithaca, N.Y.:

Chinese Society in Thailand: An Analytical History. Cornell University Press, 1957.

Chinese Society in Thailand.

Ithaca, N.Y.:

Cornell University

Press, 1962.

Leadership and Power in the Chinese Community of Thailand. Ithaca, N.Y.: Cornell University Press, 1958. Smith, Harvey, et al. Area Handbook for Tha£/and. Washington, D.C.: Area Studies, The American University, 1968.

Foreign

Suhrke, Astri. "The Thai Muslims: Some Aspects of Minority Integration." Paczfic Affairs, Vol. XIII; No. 4 {Winter 1970-71). Sutthasart, Arong. The Problem of Conflicts in the Four Southern Provinces. Bangkok: Phitak Pracha Publishing Co., 1976. Suvipakit, Soontaree. A Pilot Study of Fam£/y Health m Thai Muslim Communities in South Thailand. Yala Project, Research Report No. 1. Bangkok: National Research Council, 1969. Syamananda, Rong. Ltd., 1973.

A History of Thailand.

Bangkok:

Thai Watana Panich Co.

Thomlinson, Ralph. Thailand's Population Facts, Trends, Problems and Policies. Bangkok: Thai Watana Panich Co. Ltd., 1971. United Nations. Population Bulletin, No. 7, With Special Reference to Conditions and Trends of Fertility in the World. New York: United Nations. Economic and Social Commission for Asia and the Pacific. Country Monograph Series No. 3: Population of Thailand. Bangkok: United Nations, 1976. Wales, H.G.O. Ancient Siamese Government and Administration. Paragon Book Reprint Corp., 1965.

New York:

INSTITUTE OF SOUTHEAST ASIAN STUDIES LIST OF PUBLICATIONS IN THE RESEARCH NOTES AND DISCUSSIONS PAPERS SERIES 1

M. Mainguy, Economic Problems Related to Oil and Gas Exploration, 1976. 39pp. S$6.00 (Out of print)

2

R. William Liddle, Cultural and Class Politics zn New Order Indonesia, 1977. 21pp. S$4.00 (Out of print)

3

Raja Segaran Arumugam, State and Oil in Burma, 1977. (Out of print)

4

Hilman Adil, Australia's Pob"cy Towards Indonesia During Confrontation, 1962-66, 1977. 90pp. S$8.00 (Out of print)

5

Albert D. Moscotti, Burma's Constitution and Elections of 1974: Book, 1977. 184pp. S$13.00 (Out of print)

6

Thamsook Numnonda, Thailand and the japanese Presence, 194145, 1977. 142pp. S$13.00 (Out of print)

7

Nguyen The Anh, The Withering Days of the Nguyen Dynasty, 1978. 33pp. S$4.00

8

M. Rajaretnam, Thailand's Kra Canal: (Out of print)

9

R.O. Whyte and Pauline Whyte, Rural Asian Women: Environment, 1978. 34pp. S$4.00

10

Ismail Kassim, The Politics of Accommodation: Malaysian General Election, 1978. llOpp.

11

Leo Suryadinata, The "Overseas Chinese" in Southeast Asia and China's Foreign Policy: An Interpretative Essay, 1978. 45pp. S$5.00 (Out of print)

12

Y. Mansoor Marican, Public Personnel Administration in Malaysia, 1979. 21pp. S$4.00

13

Norbert Hofmann, A Survey of Tourism in West Malavsia and Some SocioEconomic Implications, 1979. 48pp. S$5.00

Some Issues, 1978.

36pp.

82pp.

S$5.00

A Source

S$8.00

Status and

An Analysis of the '1978 S$11.00 {Out of print)

14

B.A. Hamzah, Oil and Economic Issues in Brunei, 1980.

15

Lee Yong Leng, The Razor's Edge: Boundaries and Boundary Disputes in Southeast Asia, 1980. 29pp. S$7.50

16

Anton van Naerssen, Location Factors and Linkages at the Industrial Estates of Malacca Town: Implications for a Regional Development Policy in Peninsular Malaysia, 1980. 31pp. S$7.50

17

Pradumna B. Rana, Exchange Rate Risk Under Generalized Floating: Asian Countries, 1980. 20pp. S$5.00

18

Parsudi Suparlan and Hananto Sigit, Culture and Fertility: Indonesia, 1980. 41pp. S$7.50

19

Nor Laily Aziz et al., Culture and Fertility: 92pp. S$12.50

20

Amelia B. Alfonso, Leda L. Layo and Rodolfo A. Bulatao, Culture and Fertility: The Case of the Philippines, 1980. 67pp. S$9.50

21

Chang Chen-Tung, Ong Jin Hui and Peter S.J. Chen, Culture and FertiHty: The Case of Singapore, 1980. 95pp. S$12.50

22

Suchart Prasithrathsint, Likhit Dhiravegin and Chavalit Siripirom, Culture and Fertility: The Case of Thailand, 1980. 68pp. S$9.50

34pp.

S$7.50

Eight

The Case of

The Case of Malaysia, 1980.

THE AUTHORS Suchart Prasithrathsint, Ph.D., is currently Associate Professor, School of Applied Statistics, National Institute of Development Administration, Bangkok. He has conducted extensive re.s.~arch on demography and Thai fertility behaviour and his publications include Fertility Threshold Values of Income and Education in Thailand (Bangkok: National Institute of Development Administration, 1979) and A Micro-Socio-Economic Analysis of Thai Fertility Behaviour: A R'uml Family Perspective (Bangkok: Mahidol University, 1978). Likhit Dhiravegin, Ph.D., is currently Lecturer at the Faculty of Political Science, Thammasat University, Bangkok. His research interests are focused on Thai political attitudes and development, and he has contributed numerous articles to journals and other publications, in Thai as well as in English. Chavalit Siripirom, Ph.D., is currently attached to the Department of Social Sciences, Faculty of Social Science and Humanities, Mahidol University, Bangkok. He obtained his Ph.D. from Syracuse University, New York.