biocultural factors in school achievement

D E B O R A H L . C R O O K S / U N I V E R S I T Y OF K E N T U C K Y

Biocultural Factors in SchoolAchievement for Mopan Childrenin Belize

We were once much smarter people.—Maya caretaker at archaeological site, Toledo District, Belize

MOPAN MAYA CHILDREN of San Antonio, Toledo, Be-lize, face a number of challenges to their success inschool. Mayan ethnicity, language, and traditional eco-nomic strategies are rendered disadvantageous forschool achievement by historical, social, cultural, andeconomic forces. Statistics on poverty and child growthindicate that Mayan children are poorer and morepoorly nourished than other Belizean children, and theydo less well on national achievement exams. In addi-tion, they attend school structured on the British modeland taught in English. In the past, biocultural ap-proaches might model a causal pathway from poverty tonutritional status to cognitive development to schoolachievement. But the interrelationships among poverty,nutrition, and school achievement are more compli-cated than can be described by simple formulas or re-ductionist models. To understand these relationships,we need a combination of quantitative and qualitativeinformation.

In research conducted between November 1990and July 1991, I utilized both quantitative and qualita-tive techniques to explore the complex web of relation-ships among nutritional status, household factors, andschool performance for Mopan Maya children. Combin-ing quantitative and qualitative techniques in biocultur-al research, however, requires a rethinking of the re-search process. To incorporate qualitative methodologies

DEBORAH L. CROOKS is an assistant professor in the Department ofAnthropology, University of Kentucky. Lexington, KY 40506.

into traditional statistical models of biological anthro-pological research is difficult because of the distinctlydifferent nature of the two kinds of data produced. Oneapproach to this problem is discussed by WilliamDressier (1995), who advocates using ethnography upfront in the research design to construct culturallymeaningful variables that can then be quantified fortesting. Resulting models are able to account for locallevels of meaning within the framework of "scientific"research.

Another approach is that taken by Thomas Leather-man (in press) and Brooke Thomas (in press). This po-litical-economic approach uses quantitative measuresof biological outcome but contextualizes the numbersby focusing "upstream" on conditions that producestress and either facilitate or constrain coping re-sponses. Thus households within communities under-going rapid social and economic change become a criti-cal focus, as does the agency of individuals andhouseholds in negotiating with local, national, and in-ternational forces to produce well-being.

Another approach is that used by Francis Johnstonand Setha Low (1995) and Katherine Dettwyler (1992),which utilizes social behavior and cultural process asqualitative phenomena to assist the reader in interpret-ing quantitative results. In this approach, ethnographyreturns anthropology to biological anthropological re-search by providing a picture of how people go abouttheir lives and by situating human biology in ecological,social, political, economic, and cultural contexts. Moreimportantly, the approach allows research participantsto explain for themselves what is important in theirlives, how they go about making decisions, what theycan and cannot do, and what the constraints are under

American Anthropologist 99(3):586-601. Copyright © 1997. American Anthropological Association.

S C H O O L A C H I E V E M E N T I N B E L I Z E / D E B O R A H L . C R O O K S 5 8 7

which they live. This approach provides invaluable in-sight into the quantitative data in that it acknowledgesthe lived experiences from which the numbers wereproduced.

The research reported here was a first attempt onthe part of this anthropologist to bring together quanti-tative and qualitative data, incorporating elements of allthree approaches. I recognize that more extensive eth-nography than was allowed in this research, or perhapscollaboration among multiple researchers, would im-prove the ability of the two kinds of data to inform eachother, but that awaits the next field project. Neverthe-less, even the limited ethnographic data collected inthis research assisted in interpreting the biological dataand yielded subtle insights into the cultural and socialrealities that shape the lives of Mopan children and theirfamilies in Belize.

Background Research

Over the years, in an attempt to understand the re-lationship between poverty and learning, much re-search focused on undernutrition and cognitive devel-opment. This research was predicated on animalresearch that identified deficits in brain structure as aconsequence of early undernutrition.1 Research withpeople, however, suggested a more complicated sce-nario, one that involved factors that went beyond sim-ple nutrition and involved the quality of the child's over-all environment (Winick et al. 1975). For example,studies of children in nonpoverty environments indi-cated that cognitive deficits resulting from severe un-dernutrition were often reversible.2 And for undernour-ished children in poverty environments, "stimulating"experiences could increase cognitive test performance(Grantham-McGregor et al. 1983; Sinisterra 1987). Theseand other studies indicated that long-term functionalconsequences of undernutrition vis-a-vis cognitive de-velopment and behavior depended less upon the nutri-tional insult itself and more on the context in which itoccurred.3

Extending this scenario to school achievement, wefind that the situation is equally complex. Schoolachievement is the result of an interplay among a vari-ety of short- and long-term biological and social factors.On an individual level, poverty can lead to poor healthand nutrition, compromising cognitive development,behavior, school attendance, and thus achievement. Atthe household level, limited resources may not allow forthe purchase of school supplies and books, or paymentof various school fees, and where children are economi-cally productive members of households, a family's willor ability to release children from work may be lacking.At the national level, inequitable distribution of re-

sources may lead to inferior school infrastructure, re-duced operating expenses, and lack of quality teachersin schools situated among predominantly poor popula-tions.

Along with these biological and social constraints,cultural factors may act as barriers to school achieve-ment as well. For example, in developing countries,school is often taught in the language of colonizers, alanguage that is foreign to many indigenous childrenand compromises their ability to learn. And school cur-ricula and pedagogy may be structured by cultural val-ues equally as foreign. Where these conditions exist, es-pecially where they exist concomitant with poverty,children may have great difficulty succeeding at school.Where economies, politics, and social structures arechanging, this lack of success at school may translate tolack of economic success as adults, producing intergen-erational poverty.

A limited body of literature deals with the relation-ship between nutrition and school achievement in de-veloping countries, but it indicates a strong link be-tween the two. The recent publication of the follow-upto the Institute of Nutrition of Central America and Pan-ama (INCAP) intervention study of 1969-77 is an exam-ple.4 In this study, Reynaldo Martorell (1995) and Er-nesto Pollitt and his colleagues (1995) provide evidencefor a long-term association between early nutrition andintellectual performance for a generally poor popula-tion, such that nutritional supplementation enhances"the educational returns to schooling" (Martorell 1995:1135S) and acts as a "social equalizer." Other re-searchers report that achievement scores are higher(Clarke et al. 1991; Popkin and Lim-Ybanez 1982), gradeattainment is higher (Jamison 1986; Moock and Leslie1986), school duration is longer (Sigman and Neumann1989), and grade scores are higher (Galler et al. 1984a)for well-nourished children compared to the under-nourished growing up in poor environments. Thesemostly quantitative studies indicate a complex relation-ship that may be indirect—that is, occurring via cogni-tive deficits and/or behavioral deficits—and is oftencomplicated by factors such as parental literacy, gen-der, school absence, relative socioeconomic status, andpoverty.

One reviewer of this article questioned both therelevance of the concept of poverty measured by own-ership of material goods for people in many parts of theworld and the ethnocentricity of characterizing thelives of "peasant farmers" as unsatisfying. But the peo-ple of San Antonio often refer to themselves as "poor,"especially in conversations with outsiders. San Antoniois a landlocked community with a large and growingpopulation Subsistence farming is becoming less of anoption for everyone, and it is not desired by everyone.Parents told me that "farming is not profitable anymore,"

5 8 8 A M E R I C A N A N T H R O P O L O G I S T • V O L . 9 9 , N o . 3 • S E P T E M B E R 1 9 9 7

that it "won't have much for these children." Many par-ents hope that education will enable their children tofind work off the farm, out of the bush. As one father putit, "Now, in these modern days, for every job you needqualification [sic]—even if it's a washerwoman. If notfor my education, I wouldn't be what I am today."

Therefore, I argue that research of this type is par-ticularly important in countries like Belize where ac-cess to resources is limited for many and where it maybe ethnically and/or geographically influenced.5 AsaadShoman (1994), a Belizean, points to a strong relation-ship between poverty and education in Belize. SinceMayans are the poorest of the poor there, "the most dis-advantaged and exploited peoples in the land" (Shoman1994:297), the value of education may have particularsalience for this population of Belizeans.

The Research Setting: Belize Today andYesterday

Belizean radio and newspapers often describe Be-lize as "the Caribbean beat in the heart of Central Amer-ica" (SPEAR 1990). Situated south of Mexico and east ofGuatemala on the Caribbean coast of Central America,Belize has a unique history. Unlike the rest of CentralAmerica, which was colonized by the Spanish, Belizewas colonized by the British. Its location and pattern ofcolonization, based on timber extraction rather thanplantation agriculture, created a unique dialectic amongcolonizers, indigenous people, imported populations ofslaves and indentured servants, and immigrants fromthe north (Bolland 1988). These interactions ultimatelyled to patterns of stratified social relations which stillcharacterize Belizean society.

Today, the dominant culture in Belize is Creole, thefundamental elements of which are European and Afri-can. For the Maya, as well as other ethnic groups (suchas Garifuna, Mestizo, East Indian, and Chinese), this ne-cessitates taking part in a two-tiered system, the suc-cessful negotiation of which may be predicated on com-petence in both their own culture and the dominantCreole culture. One way to gain competence in the nationalculture may be through the educational system, but thisis made problematic by classism, racism, economic bar-riers (all embedded in a complex history), and present-day engagement with the international community.

Belize's position in the world economy is tenuousat best. It is a developing country, the economy of whichrests on a narrow agricultural base of sugar, citrus, andbananas. As such, it relies heavily on a volatile worldmarket. The 1980s recession hit the Belizean economyparticularly hard, following which agreements with theInternational Monetary Fund and United States Agencyfor International Development led to structural adjust-

ment. This adjustment required that Belize privatizemany of its assets and institute development policiesthat emphasized low-wage labor and export-oriented in-dustries, particularly agriculture and tourism (Shoman1994). As a result, disparities in wealth increased whileexpenditures for social programs decreased, leading todeterioration of the health and education systems, espe-cially in the rural areas where most Belizean Mayanslive.

Maya are the original inhabitants of Belize, and Ma-yan ruins there date from 1200 B.C., the pre-Classic pe-riod.6 Scholars suggest that by the late Classic period,around A.D. 800, anywhere from 400,000 (Bolland 1986)to 1 million Mayans (Shoman 1994) were living in Be-lize, supported by complex agriculture that included ir-rigation and ridged-field systems. This agricultural basealso supported great ceremonial centers and a complexculture that included mathematics, astronomy, andwriting, as well as the arts and institutionalized religion.

At the time of Western contact in the early 16th cen-tury, Classic Maya society was no longer in evidence inBelize, but Mayan communities continued to thrive.Throughout the 16th century and into the 17th, first theSpanish and then the British sought to exert controlover Maya in Belize, encountering fierce resistance.Historians have suggested that Belize was uninhabitedat the time of British colonization, indigenous Mayanshaving abandoned the area in response to earlier Span-ish encroachment. This, however, is a colonizer's fic-tion. Mayans were present and continued to resist.8

Clashes were constant as British loggers encroached onMayan land (Bolland 1986). When the logging industrydeclined late in the 19th century and the British turnedto farming, they recognized the potential of Mayans asagricultural laborers. Changing their policy towardMayans, the British incorporated them into the colonialprocess but denied them access to permanent lands, en-suring their dependency and contribution to the econ-omy as wage laborers.

Thus the story of Mayans in Belize is a long one, go-ing back further than our knowledge. By the time ofWestern contact, Mayan society was already millenniaold, and today Mayans continue to migrate to Belize(Bolland 1986; Thompson 1988). One group, MopanMayans from San Luis in the Peten, came to Belize inprotest against the oppressive policies of Guatemalanlandowners in the 1880s (Gregory 1984; Thompson1930), finally settling at the site of San Antonio, Toledo,where this research took place.9

San Antonio, Toledo District, Belize

It's 4:00 in the morning and the roosters are crow-ing. I hear our neighbor chopping the morning firewood:


thunk. .. thunk. . . thunk Within an hour babies arecrying and the noises of women and children becomeclear. Everyone is up and busy.

I lie in bed and listen to mothers bark directions inMaya. The little ones, still sleepy, whine, and the olderchildren scurry about their chores. Eusebio goes off tothe farm to do chores for his grandmother.10 When he re-turns, he frantically searches for his schoolbooks, re-membering an assignment that is due today.

I can smell the smoke of the morning fire; the oldestdaughter, Margarita, must be preparing the cookinghearth. She is 14 and no longer attends school; shewould like to continue, but her parents say she is too old(she has reached menarche). I hear water hitting thebottom of the metal teapot and then the sounds of Mar-garita grinding the morning corn. In a little while comesthe pat-pat-slap of tortillas being made.

Next door, the sounds are of children taking care ofchildren. The littlest ones are washed, cleaned, and fedby their older sister, who then heads to the pump to"back" water.

By the time I make it to the family kitchen, the dayis well underway for the families of San Antonio. Themen and some women have gone off to their farms orother jobs, the rest of the women are getting their ownmorning meal, and the children, it seems, have done afull day of work before they leave for school. (See Fig-ure 1.)

* * * * *

The village of San Antonio, Toledo, lies in the mostsouthern and remote district of Belize.11 Often calledthe "Forgotten District" because of its geographical andpolitical distance from the capital, Toledo is one of thepoorest districts in a poor country. It has the highestrate of unemployment in the nation, 46 percent, and thehighest percentage of people in the lowest income cate-gories (Central Statistical Office 1991).

Health statistics in Toledo District reflect povertyas well. Infant mortality is 21.8 per thousand, the high-est in the country. Compared to other developing coun-tries, this is fairly low, but newer aggregate national sta-tistics reported in Belize Today (Centers for DiseaseControl 1993) indicate the rate is probably twice that.This may be especially true in rural areas, where infantmortality often goes unrecorded. My experiences in SanAntonio indicate that this is the case. But even with lowrates, infant mortality does not always provide an accu-rate picture of health. As countries develop and infantmortality goes down, child health does not always fol-low at the same rate. This appears to be the case inToledo District, where malaria is endemic and worm in-festation affects the majority of children. Despite thesehigh rates of illness, health services in Toledo District

Figure 1A family of children in San Antonio. On the day I visited, the twoolder girls spent much of their time attending to the two youngerchildren: they played with them, washed the baby's clothes, fed himhis bottle, and the oldest helped in the kitchen. Photo by DeborahL. Crooks.

are limited (Crooks 1994b) and the lack of potablewater is a significant risk to public health (District Medi-cal Officer, personal communication, May 25,1991).

Indices of educational status in Toledo District arealso low. Although education is universal and compul-sory, it is of uneven quality, the rural areas being at thegreatest disadvantage. Toledo has the lowest percent-age of people reaching the secondary school level (6.2percent) (Central Statistical Office 1991). In addition,children of Toledo exhibit the poorest performance onthe Belize National Selection Examinations (BNSF.s)(Central Statistical Office 1990). This exam is given at theend of the primary school career and is the benchmarkthat determines who is eligible to go on to secondary

5 9 0 A M E R I C A N A N T H R O P O L O G I S T • V O L 9 9 , N o 3 • S E P T E M B E R 1 9 9 7

school. The Ministry of Education states that it assumesequal distribution of "intelligence" among children in allareas of Belize; therefore geographic variation on theBNSEs is likely due to unequal distribution of resources(Ministry of Education n.d.). (See Figure 2.) This is sup-ported by other statistics. For example, Toledo Districthas the highest proportion of "untrained teachers"among all districts (Chief Education Officer Raymond,personal communication, May 14, 1991).12

The economy of Toledo District, especially outsideof the district capital, Punta GordaTown, is farm based.Toledo produces most of the country's corn, rice, andbeans and some pork for domestic consumption, alongwith honey, cacao, and citrus for export (Wilk andChapin 1990). Mopan Maya are traditionally subsis-tence farmers, but now it is common for San Antoniomen and women to combine a number of economicstrategies. Some households cash-crop, raising every-thing from the national staples of corn, rice, and beansto cacao (from which chocolate is made) to peppers,pineapples, and plantains. Other San Antonians work

for wages at the nearby lumber mill or army base, or runsmall shops; some raise animals for sale or producecrafts. Many San Antonian households successfullycombine strategies, becoming relatively wealthy by Ma-yan standards, but production of traditional crops is sel-dom abandoned completely. During my stay in San An-tonio, I knew of only two men who no longer went to theplantation, one because he had been crippled by snake-bite and had to find other employment. When my re-search assistants were ranking families by socioeco-nomic status, one female-headed household was loweredfrom rank 2 to rank 1 (the lowest) when my assistant re-membered there was no one to farm for the household.On the other hand, the children of many of these fami-lies are now working and living out of town. One was ahigh school principal, another was a policeman; therewere a number of teachers and members of the BelizeDefense Force, and one woman was a successful mar-keter of indigenous crafts in Belize City.

Because of their close connection with the marketeconomy, the people of San Antonio travel to Punta

Figure 2Typical classroom at San Luis Rey School in San Antonio. There are few resources available to Mopan children compared to those availableto children in other towns and districts. Photo by Deborah L. Crooks.


Gorda quite often, a trip of one and a half hours. In1990-91, the bus ran four times per week, transportingvillagers for a variety of reasons, including buying orselling at the market or shops, doctor appointments, of-ficial business with the offices of the Agricultural Minis-try or Social Security, and visits with friends or relativesin town. The bus also transports the occasional touristto San Antonio, where travelers from all over the worldcome to stay at the sole guesthouse in the interior ofToledo. Thus the all-weather road, which came throughto San Antonio around 1940, has been a prime instru-ment of change for this community for some time (Gre-gory 1984, 1985).

* * * * *

I remember the first time I came to San Antonio. Webounced along a dirt road through the oppressive after-noon heat for what seemed like forever. The old schoolbus was crammed with Garifuna and Mayan women andchildren; a few men were at the back sitting among largeparcels of seed, other agricultural supplies, cases of popand beer, and cartons of goods with which to stock thevillage shops. I expected to find a small village of look-alike thatch-roofed huts, with peasant farmers tendingtheir nearby fields. But this was not the San Antonio Ifound and have come to know. Yesterday during myhousehold visits, Maria and I passed a number of newhouses being constructed out of concrete block. "Whyso many concrete houses these days?" I asked her."There's much money in town now," she told me; thenshe lowered her voice, "ganga [marijuana]," she whis-pered. But today on the bus, as we pass thatch-roofedstructures with dirt floors, Caribbean-looking raised-wooden structures, and the less common concrete-block homes with zinc roofs, some with two stories, Iknow that it is more than the marijuana trade. The eco-nomic success of many San Antonio families, in addi-tion to choice of lifestyle from more traditional to less,creates the panoply of house styles. On the outskirts ofthe village, my friend Marcelo smiles and waves fromhis shop. We stop, unload supplies for his store and hischicken-raising enterprise, and say good-bye to hiswife, who has been selling tamales in town. We continuethrough the village; more stops, more people get offwith their envelopes, packages, and boxes, and finally Iam home. The men unload supplies for another store,along with huge bags of pig feed for two farmers whoraise hogs some distance back in the bush. I deliver thegoods I purchased for my family: plastic bowls, cilantro,and bread. Miss Regina is tending the family shop whileher husband is off delivering bags of cacao for sale toHershey. She wants an accurate accounting of the costof the goods I bought. "Miss Debi, you paid too much forthe cilantro." I usually do.

The ResearchI utilized both quantitative and qualitative methods

as well as population and household perspectives.Quantitative methodologies are described elsewhere(Crooks 1994a, 1994b) and include collection of data onchild growth, health, and household characteristics suchas demography, inventory, and economic activities. Inaddition, I interviewed parents in their homes concern-ing their attitudes toward education and their hopes andaspirations for their children. The interviews were con-ducted in English when parents were comfortable usingthis language, or in Mopan Maya via an interpreter.These interviews were structured but open-ended, al-lowing parents an opportunity to provide further infor-mation beyond that addressed in the interview proto-col. Many people took advantage of this opportunity.

My research provides four sets of analyses. Thefirst documents the growth status of school-aged chil-dren, indicating a growth pattern that is typical of LatinAmerican populations experiencing chronic undernu-trition: a high percentage of children are stunted, butfew are wasted by World Health Organization (WHO)criteria (Crooks 1994a). In addition, most children haveadequate body composition, as measured by arm cir-cumference, arm fat, and arm muscle, although there isgreat variation (Figures 3 and 4). WHO criteria call forcomparing child growth to National Center for HealthStatistics (NCHS) references via z-scores, which mea-sure the distance of the child's individual measurefrom the mean of the reference group. Where samplechildren exhibit heights below -2 standard deviations(SDs) of NCHS references, they are said to be "stunted,"representing chronic undernutrition. Where weightsare below -2 SDs, they are said to be "wasted," repre-senting acute undernutrition. In addition, weight-for-height is often used as a measure of current nutritionalstatus, but since there is controversy over this measure,

0

-0.5

• 1

z-score -1.5

-2

•2 5

3

1 1 + • • -

-

9 10

— - i V 1

11 12

Age

- - % ^

o

13

- i

-c

•

14

— heightI -o- weight

- • - arm circum

-"• arm muscle• arm fat

Figure 3Nutritional status of San Antonio boys, as indicated by height,weight, arm circumference, arm muscle, and arm fat z-scores.Z-scores for height and weight calculated via NCHS data; z-scoresfor arm circumference, muscle, and fat calculated via NHANES data


Figure 4

Nutritional status of San Antonio girls, as indicated by height,

weight, arm circumference, arm muscle, and arm fat z-scores.

Z-scores for height and weight calculated via NCHS data; z-scores

for arm circumference, muscle, and fat calculated via NHANES data.

arm circumference and the derived measures of arm fatand arm muscle are included here. Since NCHS data donot include these measures, z-scores were calculatedvia National Health Examination Survey (NHANES)reference data (Frisancho 1990). The pattern of under-nutrition evident in these figures usually results from asynergistic action between inadequate nutrition interms of quantity and/or quality of diet and high rates ofillness.

The second analysis includes a number of environ-mental factors that were hypothesized to influence chil-dren's growth and nutritional status. These factors werehousehold-level measures but were considered in con-

text of both local and national-level policies and statis-tics (Crooks 1994b). Regression analysis indicated thatchild growth in San Antonio is most influenced by localsocioeconomic status, birth order, home environment,and gender (Table 1). This occurs in a part of the coun-try where unemployment and health risks are high andwhere access to health resources is limited.

The third analysis, the relationship between nutri-tional status and school achievement, and the fourth,the qualitative analysis of interviews with parents, arereported here. Given the findings of the first two analy-ses and given the relationship between malnutrition,cognitive/behavioral functioning, and school achieve-ment reported in the literature, I hypothesized a posi-tive relationship between nutritional status and schoolachievement. But given that this relationship is not asimple one and is based on the literature, I used dis-criminant analysis to explore the relationship betweenschool achievement and a number of independent vari-ables, including growth and nutritional status, house-hold demographic characteristics, and aspects of the"stimulative" quality of the home.

Discriminant analysis is an analytical techniquethat tests the relationship between a categorical de-pendent variable and multiple metric and dummy inde-pendent variables. It involves deriving the linear combi-nation of multiple independent variables that bestdifferentiates between two (or more) groups of the de-pendent variable (Hair et al. 1987; Norusis 1988). It isparticularly useful in complex anthropological research

Table 1Regression analysis for growth z-scores and household variables

Outcome variable Significant predictor variable B SE B Beta Mult R Variation explained

Stature

Weight

Arm circumference

Arm muscle

Arm fat

HOME scoregender (M/F)

HOME scorebirth-order rank

SESbirth-order rank

SESHOME scoregender (M/F)

ageSESbirth-order rankgender (M/F)

.24-.38

.16

.68

1.25.56

1.10.13.29

.07

.61

.36-.15

.07

.17

.06

.28

.30

.21

.39

.06

.13

.03

.19

.14

.07

.39-.25

.34

.29

.45

.29

.32

.27

.25

.29

.35

.28-.23

.47

.47

.55

.54

.61

21.8%

22.5%

30.0%

29.4%

37.6%

Note: Final regression models for each anthropometric outcome variables tested are given, along with the regression coefficients(B) for individual variables in the final model and the standard error of the coefficient (SE B). But since variables are measured indifferent units, the magnitude of individual regression coefficients may be easily misisnterpreted. Therefore, units of measure arestandardized and coefficients are provided as beta coefficients (Beta) for easier comparison. Finally, the Multiple R (Mult R), whichrepresents the correlation coefficient between the observed value of the dependent variable and the predicted value based on the finalmodel, is also provided; squaring Mult R provides the percentage of variation in the outcomes variable explained by the model.


because it accounts for relationships among inde-pendent variables as well as their individual relation-ships with the dependent variable. In this case, the de-pendent variable is school achievement as measured byyear-end grades. "Achievers" are those with a grade-point average of 70 or better, and "nonachievers" arethose with a grade-point average below 70. This is thecutoff for passing, set by the school principal. Using dis-criminant analysis, we can identify how school achiev-ers' lives may differ from those of nonachievers.

The sample population included 63 children ran-domly selected from all children in standards 3-6(grades 5-8) at San Luis Rey School. Standards 3-6 werechosen to control for possible bias resulting from lan-guage ability. Most children in San Antonio begin schoolspeaking Mopan Maya; they learn English, the languagein which school is taught, at varying rates. The teachersand school principal felt that, by standard 3, fluency inEnglish was no longer a problem for most children.

Based on the hypothesized model, the followingvariables were entered into the discriminant analysis:

(1) Growth. Two growth variables were entered.Height-for-age z-score (HAZ) represented long-termchronic undernutrition, while upper-arm fat z-score(UFAZ) represented current nutritional status.

(2) Early malnutrition. A considerable body of lit-erature indicates the most damaging influence of under-nutrition occurs prenatally and in the first few years ofa child's life (for example, Martorell 1995; Pollitt et al.1995). The variable (MALNUTR) was used to indicatewhether a child was diagnosed by the village healthnurse with undernutrition prior to the age of five. Theinformation was obtained from the health clinic files.

(3) Sociodemographic measures. Household so-cioeconomic status (SES) was assigned based on theaverage ranking of families by three village research as-sistants. This measure correlated well with other often-used measures of SES, including house style, presenceof electricity, and possession of modern appliances,among others (see Crooks 1994b). Dependency ratio(DEPEND) was calculated as the number of unem-ployed offspring in the home divided by the number ofadults in the home. No single age cutoff was used to de-termine this designation since many children leaveschool as early as age 14 to marry or work full-time fortheir families, while others may still be in school as lateas the age of 20. Father's and mother's literacy (FATH-LIT, MOTHLIT) was assessed via parents' self-report.

(4) School. Two school-related variables were util-ized. The first is the number of days absent from school(DAYSGONE), which may be due to illness, work re-sponsibilities, or other reason. The second is the gradelevel of the child (GRADE), that is, standards 3-6.

(5) Home environment. The child development lit-erature indicates that development is, in part, a function

of the stimulation the child receives in the home. Thisstimulation can come from various sources, includingmaterial aspects of the home environment such as vi-sual and audio stimulation (e.g., pictures on the walls,access to music), stimulating experiences outside thehome (e.g., visits with friends), and the kinds and qual-ity of caretaker/child interactions. The variable used inthis research (HOME) is derived from the Growth Fos-tering Materials and Experiences subscale of Caldwelland Bradley's HOME Instrument (Caldwell and Bradley1984), modified for use in San Antonio, and includessuch items as access to radios, books, and pictures inthe home.

(6) Other. Gender was also included in the analysis.

Results of the Discriminant Analysis

Stepwise discriminant analysis was performed us-ing the computer program Statistical Package for theSocial Sciences (SPSS-PC+), version 2.0. Results are pre-sented in Table 2, including goodness-of-fit statistics.Standardized discriminant function coefficients, whichrepresent the relative contribution of the variable to thefunction, are presented; these must be interpreted withcaution, however. As in regression analysis, the value ofthe coefficient for a particular variable depends on theother variables in the function. Therefore, variableswith large coefficients may be powerful discriminants,

Table 2Results of stepwise discriminant analysis of household variables andschool achievement

Variables in the model(in order of significance)

Father's literacyGrade levelDays absentGenderSESArm fat z-scoreHOME score

Variables not in the model:

Dependency ratioHeight z-scoreMalnutritionMother's literacy

Goodness-of-fit statistics:

EigenvalueCanonical correlationVariation explainedWilks's lambdaChi squareSignificanceHit rate

Standardizedcoefficient

.75

.49-.62

.49-.57

.32

.26

————

0.750.65

42.80%0.57

31.590.00

85.71%

Loading

.52

.44-.29

.18

.05

.25

.28

-.23.19

-.13.06


HAZ

UFAZ

Main

SES

depend

fathlit

mothlit

daygone

grade

HOME

gender

HAZ

.38#

-.15

.20

-.20

-.24*

-.04

-.05

-.06

.38#

-.30 +

UFAZ

.38#

-.14

.35 +

-.30 +

.08

-.004

.14

.42#

.31 +

-.25*

Main

-.15

-.14

.05

.08

-.19

.08

-.03

-.02

-.17

.10

SES

.20

.35 +

.05

-.08

.43#

-.03

-.05

.09

.36

.03 +

De-pend

-.20

-.30 +

.08

-.08

-.12

.03

-.05

-.31 +

-.20

-.02

FathLit

.24*

.14

-.19

.43#

-.11

.07

-.004

.19

.24*

.08

MothLit

-.04

.08

.08

-.03

.03

.07

.19

.23*

-.16

.10

DaysGone

-.05

-.004

-.03

-.05

-.05

-.004

.19

.12

-.08

.15

Grade

.06

• 42#

-.02

.09

-.31 +

.19

.23 +

.12

.22*

-.05

HOME

.38#

.31 +

-.17

.36 +

-.20

.24*

-.16

-.08

.22*

-.09

Gen-der

-.30 +

-.25*

.10

.03

-.02

.08

.10

.15

-.05

-.09

Figure 5Chart showing Pearson correlation coefficients among independent variables used in discriminant analysis. *p < .05; +p < .01; #p < .001

but the reverse cannot be assumed for variables withsmall coefficients; these may be partialed out becauseof colinearity with other variables (Hair et al. 1987;Norusis 1988). In this model, a number of variables areweakly but significantly correlated (see Figure 5), butdiscriminant analysis is robust with respect to this situ-ation; therefore, these low-to-moderate correlations donot preclude use of the procedure.

Discriminant loadings, which measure the relation-ship between the variables and the function, may pro-vide a better assessment of individual contributionsthan do coefficients. These are presented for all vari-ables entered into the analysis and indicate that the twobest predictors of achievement in the final model are fa-ther's literacy and grade in school, followed by days ab-sent from school, the home environment, and currentnutritional status (arm fat z-score). Both gender and so-cioeconomic status, originally specified in the model,have loadings that are generally considered insignifi-cant (below .20; Hair et al. 1987). It is likely that thesewere partialed out because of their correlation withother variables in the model.

Other statistics in Table 2 indicate that the functionis a good discriminator of achievement. The eigenvalue,or ratio of the between-groups to within-groups sum ofsquares, is sufficiently large, and the chi-square is sig-nificant. The square of the canonical correlation indi-cates that 42.8 percent of the variance in achievement isaccounted for by the function. The "hit rate," or the per-cent of cases correctly classified by the function, is

much improved over that which would be expected bychance alone, that is, 83.9 percent versus 50.0 percent.

Qualitative Analysis

"Dioos." I walk very slowly through the front yard. Iam cautiously watching for dogs, but I also do not wantto appear too suddenly in the open doorway of this fam-ily's home. "Dioos," calls my research assistant andgood friend, Regina. She is just ahead of me, alwaysmindful of the dogs, always ready to protect me from asudden bite. "Mind the dogs, Miss Debi." As they snarl atus, she barks back at them—something in Mopan—andthey retreat. It has been a long walk in the humid heat tothe Chos' house; they live outside of the village. Just be-fore we reach our destination, we smell smoke and I canfeel the heat of the fields being burned prior to planting.As we turn a corner in the road, I see the fire. While I getmy camera out of my pack, Regina explains the ritualthat will ensure successful planting,

You put the corn seed in a bag and put it up on a high place.Then the man burns copal all around it. This is done atnight. He prays in Maya, sometimes in English, to the Soulof the Hills and Animals—and to God—and asks that theanimals don't get into the corn. After, he lights a candle andit burns through the night. Early in the morning, he goes tothe plantation and buries copal in the center of the field.Then he waits for his helpers.

"Dioos," comes an answer from the doorway. Fran-cisca invites us in. She hurriedly tidies the floor andplaces small carved wooden stools for us to sit on. The


house is a large traditional home; it is light and airy, un-usual for San Antonio houses, which tend to be dark in-side. While we wait for her husband to come, I ask Fran-cisca if she will speak with me in English or if she wouldprefer Mayan. She prefers Mayan, so Regina translatesour small talk. I ask after her family, her children. She ismore comfortable with me than most women I inter-view, because I have been here before. I know this fam-ily and this farm outside of town in the midst of the rainforest. "Welcome, Miss Debi." Soon Lucio, a successfulfarming entrepreneur, comes in with a smile and a hand-shake. He has been out burning the fields. We begin totalk in English. He tells me that he is a self-taught man;when he was a boy, he "fooled around" in school andlater realized the importance of education. He gotbooks, he read, and he "traveled to other towns to learnabout the world." When I ask about the importance ofeducation, he tells me education is very important. "Idon't want my children to be like me," he says. "Makethey find it easier."

The Chos, like many people in San Antonio, tell methey have lived a hard life. While they value fanning andtheir own traditions and customs, they have experi-enced increasing demands for cash with little conse-quent means to produce it. Parents voice the opinionthat education will benefit their children by providingthe means to future advancement in a variety of ways(Table 3). One father says,

School is important: to study, to see what is good, it's betterthan to work on a farm. It is important for my children tobe better when they get grown, not like I am. My parentsdidn't send me to school. I want my children to be better.

A mother told me in reference to her daughter,

I am sending her to school to read and try to explain herself,for her future. I like to educate my children. I grew up inthe bush and had no education; if my mother had sent meto school, I wouldn't be sitting here. [She says this with hereighth child at her breast]

But while parents want their children to have betterlives, the routes to that betterment may vary. While ana-lyzing the qualitative data, I noticed that parents ofschool achievers seemed to respond somewhat differ-ently than parents of nonachievers to the question "Whyis school important?" (Table 3). Parents of schoolachievers most often mentioned school's value for get-ting a good job, defined by them as something otherthan subsistence farming for boys and marrying earlyfor girls. In addition, the next most often mentioned ex-planations were the ability to deal with outsiders andthe chance for a better future. These parents wantedtheir children to be better able to interact with non-

Table 3Most often mentioned responses to the question "Why is schoolimportant?" for parents of achieving and nonachieving students. (Note:Most parents gave more than one response.)

Response

To get better jobTo read and writeFor their future (life)To talk to outsidersTo be better/not like meFarming not profitableTo grow/sell betterFor a better/good lifeTo get wide knowledge/higher

educationFor their living

Achievers

37232626239933

3

Nonachievers

32363214181111114

0

Mayans, to be better equipped than they had been to ne-gotiate a changing world. An achieving student's fathertold me,

Education is important. Many people are unable to answerquestions, as we are talking here today. In the future,farming won't have much for these children. With educa-tion, they could find a job. It is sad for those who are noteducated.

An achieving student's mother says,

Yes, school is important. Because when you can read andwrite, you can move from one place to others; you can talkwith people and understand those around you. Not like me.

Parents of nonachieving students also expressedhope that their children would be well-equipped fortheir future lives and that they would be able to get"good" jobs. But they were less likely to express a con-cern over the ability to deal with outsiders and morelikely to express the more limited goal of being able toread and write. One mother says, "School is important:to read and write and to do math. We are poor; with edu-cation, we could find work."

Finally, in order to evaluate whether parents' hopesand aspirations for their children differed betweenschool achievers and nonachievers, I asked parentswhat their child would be when she or he grows up. Ta-ble 4 indicates the most often mentioned answers. Par-ents of school achievers were more likely to express as-pirations toward a "professional" job (such as teacheror medical professional) than were parents of non-achieving students. Parents of nonachieving studentswere more likely to mention the traditional occupationsof farmer or housewife; or they could not say what theirchild would be. But whether parents' aspirations areleading to differential achievement or whether parents


Table 4Most often mentioned responses to the question "What will your childbe in the future?" for parents of achieving and nonachieving students

Response

TeacherFarmer/house/stay homeI don't knowDoctor/nurse/dentistGo onto higher studiesWhat they want to beAgricultural extension agent/

agricultural projectShop clerkA sister (a nun)

Achievers

31141414996

33

Nonachievers

73632007

11

40

have different aspirations because of differentialachievement is not known.

DiscussionThe original hypothesized association between

growth in stature and school achievement was not sup-ported by the data from this population of poor Mayanchildren. Why this is so is unclear, since a number ofother studies do report such an association.13 But re-search in Guatemala suggests that in some cases, socialand cultural factors may override nutritional ones as im-portant predictors of school achievement. Workingwith children of various socioeconomic and grade lev-els in Guatemala City, Barry Bogin and Robert MacVean(1983) reported no association between growth andachievement scores, except for low correlations amongcognitive tests and height and weight for boys in grade 3and height for boys in grade 5. On the other hand, theydid find consistent associations between socioeco-nomic status and growth and socioeconomic status andachievement. They concluded that socioeconomic fac-tors were at least as important as nutritional factors inschool achievement, if not more so, especially whereundernutrition is moderate rather than severe, as it is inSan Antonio. In a later study among Mayan children,Bogin and MacVean (1987) again reported a lack of as-sociation between nutritional status and school out-come, in terms of school continuation. They suggestedthat cultural ideals concerning the purpose of schooland the economic value of children are more importantthan health and nutrition in determining school continu-ation. A similar factor may be operating in San Antonio,or it may be—as reported by Johnston and Low (1995),who also found no direct effect of stature on schoolachievement—that the relationship is indirect, medi-ated through cognitive development, a variable that Idid not measure.

On the other hand, the lack of association betweenstature and school achievement for San Antonio chil-dren may be due to the pervasive influence of stuntingwithin this population. Data reported elsewhere(Crooks 1994a) indicated that 66 percent of the samplechildren were stunted; another 28 percent were below-1 SD. Therefore, there may not be enough "chronicallywell-nourished" children in this small sample to indi-cate statistical significance with achievement.

Current nutritional status, measured here by esti-mated arm fat, does predict school achievement. Cur-rent nutritional status likewise contributed to schoolachievement for Nepalese children (Moock and Leslie1986), Chinese children (Jamison 1986), and Philadel-phia children (Karp et al. 1992). For Jamaican children(Clarke et al. 1991), weight-for-height predicted schoolachievement, but the association disappeared once so-cioeconomic variables were controlled. This again indi-cates the effect may be indirect, and Naomi Clarke andher colleagues (1991) argued for an indirect relation-ship between current nutritional status and schoolachievement via attendance, a relationship that doesnot appear to hold in San Antonio (that is, there is nocorrelation between the two; see Figure 5). Barry Pop-kin and Marisol Lim-Ybanez (1982) also argued for an in-direct relationship via the effects of hunger on the abil-ity to concentrate for poor Filipino school children, afinding not uncommon in the United States but one thatwas not tested in San Antonio.

Given the variation in research outcome outlinedabove, it can be argued that models that are limited toexploring a direct relationship between growth andachievement tell us little even when a relationship is in-dicated. The environments in which children grow arecomplex, and child growth is merely a marker of theconditions of those environments. If we are to truly un-derstand the association between growth and schoolachievement, we need to focus more on the conditionsof the environment that produce the growth. Ethno-graphic detail provides the means to grasp the elusivenature of this relationship.

Clearly, in San Antonio poverty is a factor in thelives of its children. The national phenomenon of pov-erty constrains Belizeans in their struggle to achieve,but Mayan children are at greatest risk. Focusing up-stream, we find that inequitable distribution of re-sources to Toledo District, where the majority of thecountry's Mayans live, limits the ability of San Antoniofamilies to respond to stressful social and economiccircumstances. Access to land is limited and is control-led by an outdated reservation system that continues todeny Mayans ownership (Wilk and Chapin 1989). Statis-tics presented earlier indicate that off-farm employmentis limited as well and that social services, particularly

SCHOOL ACHIEVEMENT IN BELIZE / DEBORAH L. CROOKS 597

health and education services, are minimal compared tothe rest of the country.

But within these constraints on well-being, familiesrespond differentially, some more successfully thanothers. In San Antonio there is a range of socioeco-nomic variation that may affect or be related to house-hold attitudes and responses to schooling. Father's lit-eracy, home environment, days absent from school, andgrade in school are primary predictors of schoolachievement. The first three may indicate a differencein parenting styles. That is, the homes of literate fathersmay provide more school-supportive kinds of experi-ences, including help with homework, reading materi-als, national radio broadcasts, and an insistence thatschool not be skipped for other activities. This may befacilitated by higher socioeconomic status; the correla-tions provided in Figure 5 indicate weak-to-moderatebut significant associations of father's literacy with so-cioeconomic status and the HOME score. On the otherhand, there may be another factor not accounted for inthis research that precedes both parenting styles andhigher socioeconomic status in certain families. Moreethnography could help sort this out.

The association between grade level, the second-best predictor, and school achievement is also an inter-esting finding. Class size remains fairly stable, with 40to 50 or more students, through standard 4, after whichthere is a precipitous drop to a class size of 24 in stan-dard 5 and 20 in standard 6.u It may be that children whoadvance to standards 5 and 6 are those who have be-come more successful at negotiating the dominant cul-ture than those who do not advance. It would be inter-esting to compare these children to school leavers interms of their parents' ideas about the value of school,but I have no data on school leavers. I do, however, haveinformation that grade level is negatively associatedwith dependency ratio and positively associated withcurrent nutritional status. Thus it is likely that familieswith children in higher grades are those with fewer de-pendents and can thus invest more heavily in them, pro-ducing better-nourished and better-educated childrenwho then advance to higher levels.

Irma Otzoy and Enrique Sam Colop (1988,1990) ar-gue that, for Mayans in Guatemala, education in thedominant language and culture is an important additionto, not replacement of, their own cultural repertoire.Following Richard Wilk (1991), I take the implication tobe that this provides flexibility in response to both localand national circumstances. This greater adaptive rep-ertoire allows for more flexibility as indigenous peoplesactively negotiate their position with respect to otherslocally and nationally, it allows for the maintenance oftradition within the context of change. This may be-come increasingly important as Belizean developmentpolicies move from export agriculture to tourism, par-

ticularly ecotourism. During a brief return visit to SanAntonio in 1996,1 noticed a number of new guest housesalong the main and back roads. Some of these were inoperation; some had already closed. I also heard of pro-motions in which small groups of tourists were beingbrought to Toledo District for short-term stays in Mayanand Garifuna villages. Thus, as Toledo becomes the tar-get of these kinds of development policies, Mayans arelikely to be interacting more with outsiders on a numberof levels.

This is not to suggest that up until now Mopan Ma-yans of San Antonio have lived an isolated existence.They are active local and national marketers of their ag-ricultural and other products; they seek and utilize na-tional social and health services; many work as wage la-borers; others are politically active, having electedrepresentatives to the National Assembly; and many area major force behind the Toledo Maya Cultural Council,a national indigenous rights and cultural preservationgroup. Thus many residents of San Antonio are dealingwith outsiders on a daily basis through work, throughaccessing services, and through politics. And many SanAntonians indicated the importance of being able to in-teract on a par with outsiders. One Mopan man told me,"When you meet strangers, you can't speak without edu-cation. They may want to form some kind of organiza-tion; for that you need education." Another said, "It isimportant to be able to speak with others, to be able tocommunicate. Otherwise, you stay like you're vexedwith each other [that is, you don't talk]." Finally, an-other parent told me, "It is important to learn; becauselike we [meaning Mayans], we don't understand whatpeople say."

Both Bogin and MacVean (1983) and Johnston andLow (1995) refer to respect as an important concept toindigenous Maya, although the interpretation of respectin these two studies differs. Bogin and MacVean (1983)report that Mayan mothers say that school is importantfor learning manners and respect for authority and thatthis is important for getting a job. Johnston and Low(1995) indicate that parents suggest the value of schoollies in attaining respect (respeto) and obtaining a job.Eve Danziger (in press) finds respect (izlk) to be the ba-sis of relationships among Mopan Maya in San Antonio.Traditional, religious respect is not simply an individualattribute, but a shared relational attribute betweenMopan people. Thus, respect arises out of relationshipsand is demonstrated through appropriate behavior. Al-though she cautions (personal communication, June1996) that this is a concept that operates betwcm Ma-yans only, it may be that as San Antonians interactmore and more with the outside world, obtaining re-spect in outsider relationships is becoming more andmore important as well.


In sum, school achievement for Mopan Maya chil-dren in San Antonio, Toledo, Belize, is the result of so-cial, cultural, and biological factors within a complexsocial, cultural, political, and economic environment.Toledo District, in which San Antonio is located, is apoor district, receiving less than its share of limited na-tional resources. Poverty is a real factor in the lives ofSan Antonio families and compromises the health, nu-trition, and education of the children. But within thisgreater context, variation in household environment isassociated with school achievement for children andmay relate to family attributes and attitudes. Still, cau-tion is advised in this assessment in that these relation-ships are complex, indirect, and often of indeterminatedirection. The fact that over half the variation is left un-explained indicates that other factors may be importantor that the variables in this model were insufficientlyoperationalized or measured. Johnston and Low (1995)point out the difficulty in operationalizing complex con-ceptual variables such as socioeconomic status, stimu-lation, and nutritional status. They further argue thatthe ability to show any significance at all is a testamentto the importance of these factors in the lives of chil-dren; likely, the reason is related to both.

* * * * *

It's graduation day. There is much excitement andthe village is buzzing with it. I have been invited to anumber of graduation parties. Thank goodness I pickedup trinkets in Punta Gorda Town last week; they willcome in handy as gifts. I join the throng heading to thechurch for the ceremony.

"Good morning, Miss Debi, are you coming to thegraduation?"

"I wouldn't miss it!"People in their Sunday-best are coming up the dirt

road: proud fathers in their clean, long-sleeved shirts,buttoned up to the throat despite the heat; excitedmothers in their traditional embroidered blouses andlong skirts or newer satin dresses and colorful hairdecorations; and the graduates themselves, girls inwhite dresses, boys in white shirts and ties. Sixteen chil-dren are graduating, not very many.

As we sit expectantly in church, they come sol-emnly down the aisle to receive their diplomas and totake communion with the rest of us. After the cere-mony, the graduates pose for pictures outside. I amasked by many parents to take a picture for them, and Ido so. As I focus on these recent graduates, I cannothelp but wonder what this education will do for them.Will they be better equipped for their lives? Only fourchildren passed the BNSEs with 50 percent or better, soonly four would ordinarily be eligible to attend secon-dary school in Punta Gorda. But the cutoff for Toledo

has been lowered to 30 percent, and an additional fourwill be eligible to attend.

But there are new challenges to school achieve-ment for these eight children, challenges to be over-come before they can attend their first class. They mustfind a way to pay for school; tuition, books, and roomand board can run as high as $1,550 per year: quite achallenge for poor farmers. They must also find accom-modations in town as the bus does not run between SanAntonio and Punta Gorda on a daily basis. But parentsare reluctant to allow this; they fear that their childrenwill "run around" and "get into trouble." They fear theywill return to the village with "bad ways."

For now, everyone is excited and happy. We movefrom house to house, from party to party. We eat tama-les, barbeque, and cake. We drink soda, beer, and rum.We celebrate the children, their families, and their ac-complishment. It is for their future.

NotesAcknowledgments. I wish to thank Mary K. Anglin, Barry

Bogin, Eve Danziger, Setha Low, Barbara Tedlock, Christo-pher Tourney, and two anonymous reviewers for their veryhelpful comments on the manuscript. I also wish to thank thegovernment of Belize, especially the Ministry of Health andthe Ministry of Education, for allowing this research. Mostimportantly, I thank the people of San Antonio who tried theirbest to teach me the Mayan way, botik. This research wassupported by grants from the Research Development Fundand the Center for Cognitive Sciences of the State Universityof New York at Buffalo in its initial stages, and a dissertationimprovement grant from the National Science Foundation(BNS 9016774), a grant from the Diamond Research Fund ofthe Graduate School Association at SUNY-Buffalo, and agrant-in-aid of research from Sigma Xi, the Scientific Re-search Society.

1. Bedi 1987; Lytle et a]. 1984; Shoemaker and Bloom 1977.2. Beardslee et al. 1982; Lloyd-Still et al. 1974; Stein and

Susser 1985.3. For example, Barrett and Radke-Yarrow 1985; Bogin and

MacVean 1983; Galler et al. 1984a; Galler et al. 1984b; Galleret al. 1983a; Galler et al. 1983b; Irwin et al. 1978; Johnston etal. 1987; Richardson et al. 1972.

4. The INCAP longitudinal study took place in Guatemalabetween 1959 and 1977. It tested the hypothesis that earlyundernutrition would lead to deficits in mental developmentfor Mayan children. It involved nutritional supplementationof two types, one of high-protein/high-calorie content and theother of no-protein/lower-calorie content; both also providedvitamins and minerals. The original findings of the study werereported in a variety of publications but have been recentlyrecapped in various articles in a special supplement to theJournal of Nutrition (Martorell and Scrimshaw 1995). Alsoincluded in this supplement are results of recent analysis ofa follow-up study conducted in 1988-89 to determine thelong-term effects of the intervention.

S C H O O L A C H I E V E M E N T I N B E L I Z E / D E B O R A H L , C R O O K S 5 9 9

5. Although 23 percent of all people in Belize live in pov-erty, 41 percent of those in Toledo District do so (reported inShoman 1994); these are mostly rural, mostly Mayan people.

6. Chase 1990; Ford 1990; Hammond 1991.7. Shoman 1994 does a fine job of recounting the history of

Maya-Spanish encounters during this time. I refer you to thistext.

8. Bolland 1986, 1988. See also Shoman 1994 andThompson 1988.

9. This is a very interesting story. Further detail can befound in Thompson 1930.

10. All the names of people mentioned in this article havebeen changed.

11. Not to be confused with another Mayan village, SanAntonio, Cayo District.

12. Teacher training in Belize is a complicated issue.Teachers can be (1) "degreed" (the highest level involving abachelor's degree from an accredited institution); (2)"trained" (i.e., three different levels of training involving ateacher's certificate or diploma); or (3) untrained (i.e., havingreceived a high school diploma [First Class], having com-pleted 3 years of high school [Second Class], or minimally,having passed the BNSE at 50 percent or greater [FirstTeacher]).

13. For example, Clarke et al. 1991 for Jamaica; Jamison1986 for China; Karp et al, 1992 for urban Philadelphia; Moockand Leslie 1986 for Nepal.

14. One reviewer suggested that it may be that smallerclass sizes in the higher grades may indicate more personalattention and therefore better achievement. But althoughclass size does drop, classroom size remains fairly stable.That is, the lower standards have more classrooms per grade.Therefore, there is not much variation in the number ofstudents per teacher.

References Cited

Barrett, David E., and Marian Radke-Yarrow1985 Effects of Nutritional Supplementation on Children's

Responses to Novel, Frustrating, and Competitive Situ-ations. American Journal of Clinical Nutrition42:102-120.

Beardslee, W. R., P. H. Wolff, I. Hurwitz, R. Pankh, and H.Schwachman

1982 The Effects of Infantile Malnutrition on BehavioralDevelopment: A Follow-up Study. American Journal ofClinical Nutrition 35:437-441.

Bedi, K. S.1987 Lasting Neuroanatomical Changes Following Under-

nutrition during Early Life. In Early Nutrition and LaterAchievement. John Dobbing, ed. Pp. 1-49. London: Aca-demic Press.

Bogin, Barry, and Robert B. MacVean1983 The Relationship of Socioeconomic Status and Sex

to Body Size, Skeletal Maturation, and Cognitive Statusof Guatemala City Schoolchildren. Child Development54:115-128.

1987 Growth Status, Age, and Grade as Predictors ofSchool Continuation for Guatemalan Indian Children.American Journal of Physical Anthropology 73:507-513.

Bolland, Nigel1986 Belize, a New Nation in Central America. Boulder,

CO: West view Press.1988 Colonialism and Resistance in Belize. Benque Viejo

del Carmen, Belize: Cubola Productions.Caldwell, Bettye M., and Robert H. Bradley

1984 HOME Observation for Measurement of the Environ-ment. Revised edition. Little Rock: University of Arkan-sas Press.

Centers for Disease Control and Central Statistical Office,Ministry of Finance, Belize

1993 Health in Belize. Belize Today 7:37-41.Central Statistical Office

1990 Abstract of Statistics 1990. Belmopan, Belize: Cen-tral Statistical Office, Ministry of Economic Develop-ment.

1991 Population Census: Major Findings. Belmopan, Be-lize: Government Printery.

Chase, Diane1990 The Invisible Maya: Population History and Archae-

ology at Santa Rita Corozol. In Precolumbian PopulationHistory in the Maya Lowlands. T. Patrick Culbert and DonS. Rice, eds. Pp. 199-214. Albuquerque: University of NewMexico Press.

Clarke, Naomi M. A., Sally M. Grantham-McGregor, andChristine Powell

1991 Nutrition and Health Predictors of School Failure inJamaican Children. Ecology of Food and Nutrition26:47-57.

Crooks, Deborah L.1994a Growth Status of School-Age Mayan Children in

Belize, Central America. American Journal of PhysicalAnthropology 93:217-227.

1994b Relationship between Environment and Growth forMopan Children in Belize. American Journal of HumanBiology 6:572-584.

Danziger, EveIn press Relatively Speaking: Language, Thought and Kin-

ship among the Mopan Maya. Oxford, England: OxfordUniversity Press.

Dettwyler, Katherine A.1992 The Biocultural Approach in Nutritional Anthropol-

ogy: Case Studies of Malnutrition in Mali. Medical An-thropology 15:17-39.

Dressier, William W.1995 Modeling Biocultural Interactions: Examples from

Studies of Stress and Cardiovascular Disease. Yearbookof Physical Anthropology 38:27-56.

Ford, Anabel1990 Maya Settlement in the Belize River Area: Variations

in Residence Patterns of the Central Maya Lowlands. InPrecolumbian Population History in the Maya Lowlands.T. Patrick Culbert and Don S. Rice, eds. Pp. 167-182.Albuquerque: University of New Mexico Press.


Frisancho, A. Roberto1990 Anthropometric Standards for the Assessment of

Growth and Nutritional Status. Ann Arbor: University ofMichigan Press.

Galler, Janina R., Frank Ramsey, and Giorgio Solimano1984a The Influence of Early Malnutrition on Subsequent

Behavioral Development. III. Learning Disabilities as aSequel to Malnutrition. Pediatric Research 18:309-313.

Galler, Janina R., Frank Ramsey, Giorgio Solimano, L.Thomas Kucharski, and Robert Harrison

1984b The Influence of Early Malnutrition on SubsequentBehavioral Development. IV. Soft Neurologic Signs. Pe-diatric Research 18:826-832.

Galler, Janina R., Frank Ramsey, Giorgio Solimano, andWalter E. Lowell

1983a The Influence of Early Malnutrition on SubsequentBehavioral Development. II. Classroom Behavior. Jour-nal of the American Academy of Child Psychiatry22:16-22.

Galler, Janina R., Frank Ramsey, Giorgio Solimano, WalterE. Lowell, and Elaine Mason

1983b The Influence of Early Malnutrition on SubsequentBehavioral Development. I. Degree of Impairment in In-tellectual Performance. Journal of the American Acad-emy of Child Psychiatry 22:8-15.

Grantham-McGregor, Sally, William Schofield, and LindaHarris

1983 Effect of Psychosocial Stimulation on Mental Devel-opment of Severely Malnourished Children: An InterimReport. Pediatrics 732:239-243.

Gregory, James R.1984 The Mopan: Culture and Ethnicity in a Changing

Belizean Community. Museum of Anthropology Mono-graphs in Anthropology, 7. Columbia: University of Mis-souri-Columbia.

1985 Educational Modernization in Southern Belize. Be-lizean Studies 13:17-43.

Hair, Joseph F., Jr., Rolph E. Anderson, and Ronald L.Tatham

1987 Multivariate Data Analysis, Second Education. NewYork: Macmillan.

Hammond, Norman1991 The Maya and their Civilization. In Cuello, an Early

Maya Community in Belize. Norman Hammond, ed. Pp.1-7. Cambridge: Cambridge University Press.

Irwin, Marc, Patricia L. Engle, Charles Yarbrough, RobertE. Klein, and John Townsend

1978 The Relationship of Prior Ability and Family Charac-teristics to School Attendance and School Achievementin Rural Guatemala. Child Development 49:415-427.

Jamison, Dean T.1986 Child Malnutrition and School Performance in

China. Journal of Development Economics 20:299-309.Johnston, Francis E., and Setha M. Low

1995 Children of the Urban Poor: The Sociocultural Envi-ronment of Growth, Development, and Malnutrition inGuatemala City. Boulder, CO: Westview Press.

Johnston, Francis E., Setha M. Low, Yetilu de Baessa, andRobert B. MacVean

1987 Interaction of Nutritional and Socioeconomic Statusas Determinants of Cognitive Development in Disadvan-taged Urban Guatemalan Children. American Journal ofPhysical Anthropology 73:501-506.

Karp, Robert, Roy Martin, Trevor Sewell, John Manni, andArthur Heller

1992 Growth and Academic Achievement in Inner-CityKindergarten Children. Clinical Pediatrics 31:336-340.

Leatherman, ThomasIn press Changing Biocultural Perspectives on Health in

the Andes. Social and Science and Medicine.Lloyd-Still, J. D., I. Hurwitz, P. Wolff, and M. Schwachman

1974 Intellectual Development after Severe Malnutritionin Infancy. Pediatrics 54:306-311.

Lytle, Loy D., Carla S. Whitacre, and Mark F. Nelson1984 Mechanisms of Nutrient Action on Brain Function.

In Nutrition and Behavior: A Comprehensive Treatise,vol. 5. Janina R. Galler, ed. Pp. 223-265. New York: Ple-num Press.

Martorell, Reynaldo1995 Results and Implications of the INCAP Follow-up

Study. Journal of Nutrition 125:1127S-1138S.Martorell, Reynaldo, and Nevin S. Scrimshaw, eds.

1995 The Effects of Improved Nutrition in Early Child-hood: The Institute of Nutrition of Central America andPanama (INCAP) Follow-up Study. Theme issue. Journalof Nutrition Supplement 125(45).

Ministry of Education, Belizen.d. Examinations. BNSE, CXC, Teachers. [Belmopan, Be-

lize]: Ministry of Education.Moock, Peter R., and Joanne Leslie

1986 Childhood Malnutrition and Schooling in the TeraiRegion of Nepal. Journal of Development Economics20:33-52.

Norusis, Marija J.1988 SPSS PC+ Advanced Statistics V2.0. Chicago: SPSS.

Otzoy, Irma, and Enrique Sam Colop1988 Mayan Ethnicity and Modernization. Paper pre-

sented at the 87th annual meeting of the American An-thropological Association. Phoenix, AZ.

1990 Ethnicity and Modernization among the Maya ofGuatemala. Mesoamerica 11:97-100.

Pollitt, Ernesto, Kathleen S. Gorman, Patrice L. Engle, JuanA. Rivera, and Reynaldo Martorell

1995 Nutrition in Early Life and the Fulfillment of Intellec-tual Potential. Journal of Nutrition 125:1111S-1118S.

Popkin, Barry M., and Marisol Lim-Ybanez1982 Nutrition and School Achievement. Social Science

and Medicine 16:53-61.Richardson, S. A., H. G. Birch, E. Grabie, and K. Yoder

1972 The Behavior of Children in School Who Were Se-verely Malnourished in the First Two Years of Life. Jour-nal of Health and School Behavior 13:276-284.

Shoemaker, William J., and Floyd E. Bloom1977 Effect of Undernutrition on Brain Morphology. In

Nutrition and the Brain, vol. 2. R. J. Wurtman and J. J.Wurtman, eds. Pp. 148-192. New York: Raven Press.

SCHOOL ACHIEVEMENT IN BELIZE / DEBORAH L. CROOKS 601

Shoman, Asaad1994 Thirteen Chapters of a History of Belize. Belize City,

Belize: Angelus Press Limited.Sigman, Marian, and Charlotte Neumann

1989 Cognitive Abilities of Kenyan Children in Relation toNutrition, Family Characteristics, and Education. ChildDevelopment 60:1463-1474.

Sinisterra, Leonardo1987 Studies on Poverty, Human Growth and Develop-

ment: The Cali Experience. In Early Nutrition and LaterAchievement. John Dobbing, ed. Pp. 208-244. London:Academic Press.

SPEAR (Society for the Promotion of Education andResearch)

1990 SPEAR Reports 3: Profile of Belize 1989. N.p.: CubolaProductions.

Stein, Zena, and Mervyn Susser1985 Effects of Early Nutrition on Neurological and Men-

tal Competence in Human Beings. Psychological Medi-cine 15:717-726.

Thomas, R. BrookeIn press Wandering toward the Edge of Adaptability: Ad-

justments of Andean People to Change. In Human Adapt-

ability. S. J. Ulijaszek and R. Huss-Ashmore, eds. Oxford,England: Oxford University Press.

Thompson, J. Eric S.1930 Ethnology of the Mayas of Southern and Central

British Honduras. Field Museum of Natural History Pub-lications, 274. Anthropological Series, vol. 17, no. 2. Chi-cago: Field Museum of Natural History.

1988 The Maya of Belize: Historical Chapters since Colum-bus. Benque Viejo del Carmen, Belize: Cubola Produc-tions.

Wilk, Richard R.1991 Household Ecology: Economic Change and Domes-

tic Life among the Kekchi Maya in Belize. Tucson: Uni-versity of Arizona Press.

Wilk, Richard, and Mac Chapin1989 Belize: Land Tenure and Ethnicity. Cultural Survival

Quarterly 13:41-45,1990 Ethnic Minorities in Belize: Mopan, Kekchi, and

Garifuna. SPEAR Reports 1. Belize: Cubola Productions.Winick, Myron, Knarig Katchadurian Meyer, and Ruth C.Harris

1975 Malnutrition and Environmental Enrichment byEarly Adoption. Science 190:1173-1175.

biocultural factors in school achievement

Documents