infant development study guide
yht9318
kelleretal.20091.pdf
International Journal of Behavioral Development 2009, 33 (5), 412–420
http://www.sagepublications.com
© 2009 The International Society for the Study of Behavioural Development
DOI: 10.1177/0165025409338441
Distal and proximal parenting
Parenting is a cultural practice that enables children’s develop- ment and the acquisition of competence in a particular socio- cultural environment from birth onwards (Greenfield, Keller, Fuligni, & Maynard, 2003; Keller, 2003; Keller, Yovsi et al., 2004; LeVine, 1970; Liu et al., 2005). Two different parenting styles during the first months of life have been described in the literature (Keller, 2007): the proximal and the distal style. The proximal parenting style is characterized by bodily proximity and body stimulation. The distal parenting style is character- ized by face-to-face contact and object stimulation, i.e., communication through the distant senses. Distal and proximal styles can be regarded as two alternative parenting strategies which are related to the sociodemographic profile of particular contexts.
The proximal parenting style is predominant in traditional subsistence societies (Keller, Lohaus et al., 2004), in which socialization goals that embody relatedness, obedience, and hierarchy are preferred (Kağıtc‚ıbas‚ı, 1996; Keller, Kärtner, Borke, Yovsi, & Kleis, 2005; Keller, Lamm et al., 2006). It could be shown that this style reinforces closeness and warmth (Bandura, 1989; Hetherington & Frankie, 1967). Further- more, longitudinal studies showed that the early experience of proximal parenting is related to an early development of com - pliance (Keller, Kärtner et al., 2005; Keller, Yovsi et al., 2004).
The distal parenting style emphasizes autonomy and sepa- rateness which have been demonstrated as precursors of independent agency. It is valued in Western industrial and postindustrial middle-class families, in which competition, individual achievements, self-enhancement, and equality constitute the preferred socialization goals (Keller, 2007; Keller, Borke, Yovsi, Lohaus, & Jensen, 2005; Keller, Hentschel et al., 2004; Keller, Kärtner et al., 2005; Keller, Lamm et al., 2006; Keller, Lohaus et al., 2004). In longitudinal studies we have demonstrated that the early experience of distal parent- ing leads to an early development of self-recognition (Keller, Kärtner et al., 2005; Keller, Yovsi et al., 2004).
The assumption of distal and proximal parenting as alterna- tive strategies is based on findings demonstrating that high amounts of body contact are consistently associated with low amounts of mutual visual engagement and vice versa across cultures (Keller, Lohaus et al., 2004; Keller, Yovsi et al., 2004; LeVine, 2004). Further confirmation of this argument is offered by studies on other primates. Bard and colleagues (2005) demonstrated that in chimpanzees mutual gaze was inversely related to maternal cradling. When mother and infant are in constant physical contact, there is little mutual gaze. These researchers further argue that mutual engagement can be primarily tactile, which is evolutionarily the most basic pattern, found in most non-human primates and humans in rural eco-cultural environments (Bard, 2002; Stack, 2001). When physical contact is reduced, mutual engagement shifts to the visual (Lavelli & Fogel, 2002).
In this article, we want to systematically demonstrate that proximal and distal parenting styles represent different parent- ing strategies related to different sociodemographic profiles
Distal and proximal parenting as alternative parenting strategies during infants’ early months of life: A cross-cultural study
Heidi Kellera, Joern Borkea, Thomas Staufenbiela, Relindis D. Yovsia, Monika Abelsa, Zaira Papaligourab, Henning Jensenc, Arnold Lohausd, Nandita Chaudharye, Wingshan Lof and Yanjie Sug
Cultures differ with respect to parenting strategies already during infancy. Distal parenting, i.e., face- to-face context and object stimulation, is prevalent in urban educated middle-class families of Western cultures; proximal parenting, i.e., body contact and body stimulation, is prevalent in rural, low-educated farmer families. Parents from urban educated families in cultures with a more inter - dependent history use both strategies. Besides these cultural preferences, little is known about the relations between these styles as well as the behavioural systems constituting them. In this study there- fore, the relations between the styles and the constituting behaviours were analysed in samples that differ with respect to their preferences of distal and proximal parenting. The hypothesized differences between the samples and the negative relationship between distal and proximal parenting, as well as between the respective behavioural systems can clearly be demonstrated. Furthermore, the impact of the sociodemographic variables with respect to the parenting strategies can be shown. Results were discussed as supporting two alternative parenting strategies that serve different socialization goals.
Keywords: culture; infancy; mother–child interaction; parenting strategies; socialization goals
Correspondence should be sent to Heidi Keller, University of Osnabrück, Culture & Development, Artilleriestr. 34, Osnabrueck 49076, Germany; e-mail: [email protected]
The first author was granted funds from the German Research Council for this study.
aUniversity of Osnabrück, Germany. bAristotle University of Thessaloniki, Greece. cUniversity of Costa Rica, Costa Rica. dUniversity of Bielefeld, Germany. eLady Irvin College, India. fAsian Pacific American Legal Center, USA. gUniver- sity of Peking, China.
412-420 JBD338441 Keller:210 x 280mm 30/07/2009 10:58 Page 412
that can be found systematically in very different cultural surroundings. We hypothesize that distal parenting is shown by highly educated urban middle-class mothers. These mothers have their first baby in their late twenties and early thirties and usually have few children, as higher levels of formal education correlate with lower numbers of children (Alvarez, Brenes, & Cabezas, 1990). Formal education emphasizes cognitive func- tioning as segregated from conative and emotional aspects of thought (Serpell & Hatano, 1996) and thus analytical inde- pendence (Greenfield, 1994). Formal education also affects mother–infant interactions, increasing face-to-face communi- cation (Richman, Miller, & LeVine, 1992; Tapia Uribe, Levine, & Levine, 1994). Distal parenting can therefore be regarded as connected with formal education in urban middle-class families.
Proximal parenting, however, is assumedly regarded as appropriate by rural mothers who have their first baby in their late teens or early twenties. Their living environment is char- acterized by little access to formal education. The family is the educational unit where learning is mainly observational (Greenfield, 2004), and can be characterized as “legitimate peripheral participation” (Lave & Wenger, 1991; Rogoff, 2003). This type of education centres on social intelligence (Mundy-Castle, 1974) based on a hierarchical apprenticeship model (Keller, 2003).
Cultural prototypes
The two parenting strategies are assumed to be closely related to the dimensions of interpersonal distance and the dimension of agency in Kağıtc‚ıbas‚ı’s model (1996). Interpersonal distance has the end points of separateness and relatedness and agency has the end points of heteronomy and autonomy. Although several combinations are possible Kağıtc‚ıbas‚ı (1996) describes three prototypical models: independence (separate- ness and autonomy), interdependence (relatedness and heteronomy) and psychological interdependence (relatedness and autonomy). She relates these models to different contex- tual patterns: the model of independence characterizes Western urban middle-class families, the model of interdependence characterizes non-Western rural families and the autonomous- related (psychological interdependence) model characterizes urban educated middle-class families from non-Western societies. For this study data were collected in communities which can be considered as representing these three cultural models.
The first group of samples, representing the cultural model of independence, therefore consists of urban, educated middle- class families from Western societies. We collected samples of Euro-American families from Los Angeles, German families from Marburg, and Greek families from Athens. Euro- American, German, and Greek middle-class families have been described as endorsing a distal parenting strategy with a high amount of eye contact and object play (Keller, Demuth, & Yovsi, 2008; Keller, Hentschel et al., 2004; Keller, Kärtner et al., 2005; Keller & Lamm, 2005; Keller, Papaligoura et al., 2003; Keller, Yovsi et al., 2004; LeVine, 2004).
The second group of samples, representing the cultural model of interdependence, consists of rural farming families with little formal education. We recruited samples from the ethnic tribe of Cameroonian Nso and Indian Gujarati villagers. Nso and Gujarati villagers have been described as practising proximal parenting strategies characterized by high amounts of body contact and body stimulation (Abels et al., 2005; Keller,
2003; Keller, Abels et al., 2005; Keller, Kärtner et al., 2005; Keller, Lohaus, Völker, Elben, & Ball, 2003; Keller, Voelker, & Yovsi, 2005; Keller, Yovsi et al., 2004; Keller, Yovsi, & Voelker, 2002; Nsamenang, 1992; Nsamenang & Lamb, 1994).
For our third group of samples, representing the cultural model of autonomous-relatedness, we collected samples from San José (Costa Rica), Beijing and Taiyuan (China), Delhi (India), and urban educated Nso (Cameroon). These families are expected to value autonomy as well as relatedness due to their urban educated lifestyle in a society that still holds beliefs especially for family life that is traditionally oriented towards relatedness (Kağıtc‚ıbas‚ı, 1996). They should therefore practise distal as well as proximal parenting styles. Theoretically, they could be expected to express the same amount of autonomous related behaviours as the parents with an independent cultural model and the same amount of relatedness oriented behaviours as the parents with an interdependent cultural model. However, previous studies revealed an intermediate position (Keller, 2007; Keller, Yovsi et al., 2004).
We tested the following hypotheses:
1 We expected the four parenting systems (body contact, body stimulation, face-to-face context, object stimulation) as well as the two parenting styles (proximal and distal) to express systematic variations across the three cultural models. We expected the contexts representing the independent cultural model to score highest on face-to-face context and object stimulation, expressing the distal style of parenting and the samples representing the interdependent cultural model to score highest on body contact and body stimulation, repre- senting the proximal style of parenting. With respect to samples representing the model of autonomous relatedness, we expect them to score middle on all systems, expressing the two styles on a medium level.
2 We assumed that the sociodemographic profile substantially defines the parenting style. However, culture has a surplus meaning that accounts for additional variance in the prediction of parenting styles.
3 Since we assumed that the two styles as well as the consti- tutive parenting systems are part of alternative parenting strategies, we expected distal and proximal parenting as well as their components to be negatively related.
Method
Participants
Participants in the present study were 214 families from nine different cultural communities; 39 of the families lived in rural areas and 175 in urban areas; 130 were primiparae, 79 multi- parae, while for 5 families this information was missing. There were no significant differences concerning the gender of the infants (χ2(8) = 5.63, ns). The assessment took place when the youngest child in the family was about three months old (plus/minus one week). All children were physically healthy at the time of assessment (Table 1).
The Western urban middle-class samples
The Euro-American, German, and Greek samples consisted of urban middle-class families with mostly first-born infants. The mean mothers’ age was 31.49 (SD = 3.80) and the mean age of the fathers 34.98 (SD = 4.86) years. Mothers and fathers
INTERNATIONAL JOURNAL OF BEHAVIORAL DEVELOPMENT, 2009, 33 (5), 412–420 413
412-420 JBD338441 Keller:210 x 280mm 30/07/2009 10:58 Page 413
414 KELLER ET AL. / DISTAL AND PROXIMAL PARENTING AS ALTERNATIVE STRATEGIES
T a
b le
1 D
es cr
ip ti
on o
f th
e sa
m pl
es
A ge
Y ea
rs o
f fo
rm al
s ch
oo lin
g
G en
de r
B ir
th o
rd er
M ot
he rs
F at
he rs
M ot
he rs
F at
he rs
C on
te xt
C u lt u ra
l sa
m pl
es N
M al
e F
em al
e F
ir st
L at
er M
S D
M S
D M
S D
M S
D R
u ra
l U
rb an
E u
ro -A
m er
ic an
s 2 0
7 a
1 3
a 1 4
a 6
a 3 5 .0
0 3 .6
1 3 6 .7
5 4 .8
4 1 6 .8
9 1 .5
2 1 6 .7
5 2 .1
7 0
2 0
G er
m an
s 3 1
1 5
a 1 6
a 3 1
a 0
a 2 9 .5
8 4 .2
6 3 3 .6
8 5 .2
1 1 4 .1
3 3 .0
2 1 3 .7
7 3 .0
5 0
3 1
G re
ek s
2 9
1 5
a 1 4
a 1 9
a 1 0
a 2 9 .9
0 3 .5
2 3 4 .5
2 4 .5
3 1 3 .3
4 3 .1
0 1 2 .4
5 3 .1
3 0
2 9
C o st
a R
ic an
s 1 9
1 0
a 7
a 9
a 8
a 2 7 .0
6 4 .8
5 3 1 .4
0 6 .6
4 8 .6
5 2 .5
7 —
— 0
1 9
C h
in es
e 2 0
7 a
1 3
a 2 0
a 0
a 2 7 .5
5 2 .6
7 3 1 .3
0 3 .0
2 1 4 .1
0 1 .8
9 1 4 .8
0 2 .7
1 0
2 0
U rb
an N
so 2 1
7 a
1 2
a 7
a 1 2
a 2 9 .9
5 6 .4
3 —
— 1 1 .5
0 .9
7 —
— 0
2 1
U rb
an I
n d
ia n
s 3 5
1 7
a 1 6
a 1 4
b 2 0
b 2 9 .0
3 3 .4
2 3 2 .3
6 4 .0
0 1 5 .5
3 1 .3
1 1 6 .1
8 1 .0
0 0
3 5
R u
ra l
N so
2 3
8 a
1 5
a 9
a 1 4
a 2 6 .7
4 5 .5
5 3 5 .4
7 6 .0
1 7 .0
4 .9
8 6 .5
2 1 .5
6 2 3
0 R
u ra
l In
d ia
n s
1 6
7 a
9 a
7 a
9 a
2 2 .6
9 4 .7
5 2 6 .8
1 5 .9
8 2 .7
5 3 .4
0 8 .4
7 4 .0
0 1 6
0 T
o ta
l 2 1 4
9 3
c 1 1 5
c 1 3 0
d 7 9
d 2 8 .8
5 5 .1
1 3 2 .9
7 5 .4
9 1 2 .1
8 4 .6
6 1 3 .0
3 4 .2
6 3 9
1 7 5
N ot
es .
a F
ro m
t w
o f
am il ie
s th
is i
n fo
rm at
io n
w as
m is
si n
g. b
F ro
m o
n e
fa m
il y
th is
i n
fo rm
at io
n w
as m
is si
n g.
c F
ro m
s ix
f am
il ie
s th
is i
n fo
rm at
io n
w as
m is
si n
g. d
F ro
m fi
ve f
am il ie
s th
is i
n fo
rm at
io n
w as
m is
si n
g.
412-420 JBD338441 Keller:210 x 280mm 30/07/2009 10:58 Page 414
hold high levels of formal education (14.55 (2.66) years of formal schooling). They lived as nuclear families in apartments and houses. Usually, infants had rooms of their own already. During the early months of infants’ lives, mothers were the exclusive caretakers who spend most of the time in “mother–infant isolation”. The Euro-American sample was recruited in 2002, the German and the Greek samples in 1999.
The rural village samples
The Cameroonian Nso families lived in the North-Western grassfields of Cameroon, the Indian Gujarati families live in the Nandesari area. The samples consisted of rural farming families who lived patrilocally mainly in extended households. The mean age of the mothers was 24.71 (5.15) and the mean age of the fathers 31.14 (5.99) years. The level of formal education was low to minimal (6.19 (2.48) years of formal schooling). Mothers were infants’ primary caretakers during the early months of life, however, the mother–infant dyad was embedded in a close knit network of kin and neighbours. During the night, infants co-slept with their mothers. The samples were assessed in the years 2000, 2001, and 2003.
The non-Western urban middle-class samples
The urban Costa Rican, Chinese, Indian, and Cameroonian samples were characterized by relatively high educational levels (13.46 (1.74) years of formal schooling; lowest among the Costa Rican mothers). The mean age of mothers was 28.40 (4.34) and the mean age of the fathers 31.69 (4.55) years. Families were three generational or nuclear with kin living close by, so that the mother–infant dyad lived in constant interactional networks.
The families in the different samples followed different lifestyles as related to religion with the Costa Rican and the Nso families being Catholic. The Nso families all belonged to the ethnic group of the Nso with a traditional sociocultural system they also follow. The Indian families followed Hinduism and the Chinese families Confucianism. Assessments were made in 2001 in Costa Rica, in 2002 in China and in 2003 in Cameroon and India.
Procedure
The data collection was done by local researchers and research assistants who were trained in interview techniques, video - taping, and demographic assessment by supervisors from the German coordination centre of the study. In all cases, the field researchers were members of the cultural community under study.
Information about the character of the study was provided to all the participants at the first contact. The instruction informed the family that we were interested in child develop- ment across cultures. Because of our interest concerning inter- action behaviour in everyday life and in the families’ natural surroundings, the families were visited at home. The visits started with a warm-up and familiarization phase which consisted of an informal conversation with the families, deter- mined to some extent by the cultural community. This was followed by the interview assessing demographic information. Thereafter, a non-standardized free-play situation with only the mother and the baby was videotaped. The mothers were therefore asked to play in the way they usually do without any
further instruction. All communication was done in the mother’s native language.
Because our intention was to examine a setting in every culture that allowed the assessment of similarities as well as differences in parenting, we decided to focus on free-play situations between mother and infant. Although the studied cultural communities differ substantially in their definitions of the adequate care of small babies, free-play situations can be found in all cultural environments, although to different degrees depending on the workloads of the mothers. Individ- ual bouts of play have a duration of 5 to 10 minutes length, defined by infants’ attention spans during that developmental phase. Free-play situations required the infant to be awake and fed but there were no further specifications with respect to content or duration (Keller, Lohaus et al., 2004; Keller, Voelker, & Yovsi, 2005; Keller, Yovsi, & Voelker, 2002).
In order to familiarize the families with the videotaping procedure, we recorded care and other routine situations, prior to the actual recording of the free-play mother–infant inter - action. These practice situations were not included in the video analysis. The length of the free-play episodes was about 10 minutes (M = 9.02, SD = 2.68). The videotapes in the Greek (M = 7.16, SD = 2.45) and in the rural Cameroonian (M = 7.22, SD = 3.00) sample were significantly shorter than the videotapes in the Costa Rican sample (M = 10.53, SD = 4.53). These discrepancies emerged from different amounts of drop-out crying and/or sleepy/sleeping behaviour (299 sleepy and 98 crying intervals in the Greek sample, 59 sleepy and 163 crying intervals in the Cameroonian sample, 101 sleepy and 111 crying intervals in the Costa Rican sample) and because of differences in the total length of the tapes. If the babies fell asleep, the videotaping was stopped and not continued. Because of the ratio scores described below, these differences were controlled for in the statistical analysis. All participants received a small gift as an acknowledgement of their participation.
Measures
The coding of parenting systems
Parenting strategies were assessed with the behavioural systems conceptualized in the component model of parenting (CMP, Keller, 2002; Keller, Lohaus et al., 2004), which allowed a culture-sensitive description through the flexible combination of the different systems; body contact, body stimulation, face- to-face context, and object stimulation. These systems reflect universal parenting systems which occur in any cultural context. However, the amount of the individual systems as well as their combinations vary across cultures (Keller, 2007). Following the description above and based on previous analysis (Keller, Kuensemueller et al., 2005), these four parenting systems can be compiled into a distal (face-to-face context and object stimulation) and a proximal (body contact and body stimulation) parenting style.
The videotaped free-play interactions were analysed by members of the German coordination centre of this study through a computer-based video analysis system (Voelker et al., 1999). First, the whole registered time of free-play inter - action was divided into 10-second intervals. In a second stage, intervals in which the child was not fussy/crying or sleepy/asleep were identified and each of these intervals
INTERNATIONAL JOURNAL OF BEHAVIORAL DEVELOPMENT, 2009, 33 (5), 412–420 415
412-420 JBD338441 Keller:210 x 280mm 30/07/2009 10:58 Page 415
received a separate code for the four parenting systems: body contact system, face-to-face context, object stimulation, and body stimulation.
If, during the interactional episodes, the mother or child could not be clearly seen on the video, these events were coded as not visible (due to extensive training of the researchers hardly any non-visible situations occurred, less than .04% of the episodes). Therefore, measurements, as well as reliabilities, were based on episodes during which the relevant behaviour was not obscured in any way.
In the following, the exact coding of the four parenting systems is presented.
Body contact system. Body contact was coded each time when one or more of the following body contact positions occurred for at least 5 seconds of one 10-second interval: both legs of the child are in contact with the mother, both legs and parts of the torso of the child are in contact with the mother, the whole or almost the whole body of the child is in contact with the mother. Then the percentage of “body contact” intervals per dyad was calculated.
Body stimulation system. Body stimulation was coded each time when one or more of the following stimulation behaviours occurred within one 10-second interval: the whole body of the child is moved, body parts of the child are moved, the child is touched repeatedly and all situations during which the infants body is stimulated with the mother’s face; e.g., kissing. Then the percentage of body stimulation intervals per dyad was calculated.
Face-to-face context. The face-to-face system was defined as effort of the mother to position her body and head to her infant in a way that allowed face-to-face exchange. Face-to-face positions were coded with a time sampling method when the mother created a situation for at least 5 seconds of the 10- second interval that allowed face-to-face exchange. In any other case “no face-to-face positions” was coded. In a next step the percentage of face-to-face intervals per dyad was calculated.
Object stimulation system. The object stimulation system was defined as the effort of the mother to attract the attention of the infant to an object. Some mothers used toys such as puppets or rings and other mothers used household objects or things from nature around them such as spoons or wooden sticks. The kind of object depended on the respective cultural context and the preferences of the mother. Object stimulation was coded when an object was included in the interaction within a 10-second interval. In the next stage the percentage of object stimulation intervals per dyad was calculated.
Proximal parenting style. For the proximal parenting style, the average of the categories body contact system and body stimulation system was calculated.
Distal parenting style. For the distal parenting style, the average of the categories face-to-face context and the object stimulation system was calculated.
Inter-rater reliability. The reliabilities were calculated on the basis of a sub-sample of 10 video sequences that were randomly collected from the complete set of samples (4.67%). Cohen’s Kappa was .86 for the body contact system, .90 for the body stimulation, .85 for the face-to-face context, and .99 for the object stimulation.
Results
The mean percentages of the occurrence of the four parenting systems – body contact, body stimulation, face-to-face context, and object stimulation – are shown in Table 2 separately for the nine cultural samples. Averaging over the parenting systems we included proximal parenting (body contact and body stimula- tion) and distal parenting (face-to-face context and object stimulation) as additional variables.
416 KELLER ET AL. / DISTAL AND PROXIMAL PARENTING AS ALTERNATIVE STRATEGIES
Table 2 Means and standard deviations (in parentheses) for parenting systems by samples
Body Body Face-to-face Object Proximal Distal contact stimulation context stimulation parenting style parenting style
Sample N M (SD) M (SD) M (SD) M (SD) M (SD) M (SD)
Euro-American sample 20 .43a (.31) .55aa (.22) .57a (.22) .45a (.35) .49a (.21) .51a (.23) German sample 31 .26a (.29) .45aa (.22) .84a (.18) .41a (.29) .35a (.20) .62a (.15) Greek sample 29 .20a (.28) .56aa (.24) .77a (.18) .39a (.35) .38a (.22) .58a (.20) Costa Rican sample 19 .57a (.30) .63aa (.19) .59a (.29) .08a (.13) .60a (.18) .33a (.17) Chinese sample 20 .32a (.33) .59aa (.17) .60a (.26) .44a (.29) .46a (.19) .52a (.15) Urban Nso sample 21 .52a (.36) .59aa (.28) .56a (.31) .24a (.37) .56a (.29) .40a (.17) Urban Indian sample 35 .38a (.33) .60aa (.23) .54a (.24) .29a (.31) .49a (.22) .42a (.19) Rural Nso sample 23 .83a (.22) .66aa (.22) .63a (.27) .01a (.02) .74a (.12) .32a (.14) Rural Indian sample 16 .57a (.32) .48aa (.18) .40a (.22) .02a (.06) .52a (.16) .21a (.10) Independent 80 .28a (.30) .51aa (.23) .75a (.22) .41a (.32) .40a (.22) .58a (.19) Autonomous-related 95 .44b (.34) .60ba (.22) .57b (.27) .27b (.31) .52b (.22) .42b (.18) Interdependent 39 .72c (.29) .59ab (.23) .53b (.27) .01c (.04) .65c (.20) .27c (.13)
Note. abcMeans in the same column that do not share superscripts differ at p < .05.
412-420 JBD338441 Keller:210 x 280mm 30/07/2009 10:58 Page 416
Differences between the cultural orientations (hypothesis 1)
To test hypothesis 1 concerning the mean differences in the four parenting systems between the different cultural models (independence, autonomous relatedness, interdependence), we first calculated a MANOVA with the cultural model as independent variable and the four parenting systems as dependent variables. The results showed that there were signifi- cant differences concerning the occurrence of the parenting systems F(8,416) = 13.44, p < .01 (Wilks λ = .63, partial χ2 = .21). As follow-up analyses four ANOVAs with the cultural models as independent variables and the parenting system as dependent variable were calculated. All parenting systems reached the level of significance (body contact: F(2,211) = 24.94, p < .01, χ2 = .19; body stimulation: F(2,211) = 3.65, p < .05, χ2 = .03; face-to-face context: F(2,211) = 14.17, p < .01, χ2 = .12; object stimulation: F(2,211) = 25.34, p < .01, χ2 = .19).
In a next step, multiple comparisons (Tukey HSD tests) with the cultural models as independent variables and the parent- ing system as dependent variable were calculated as post-hoc tests. As shown in Table 2, all three groups differed significantly in the predicted direction concerning the amount of body contact and object stimulation. The amount of body stimula- tion differed significantly between the independent (less) and the interdependent samples (more) and the face-to-face context differed significantly between the independent (more) and the autonomous-related samples (less).
With parenting styles as the dependent variable, we were also able to find support for hypothesis 1: the factor cultural model produced a significant effect in the MANOVA, F(4,420) = 21.17, p < .01 (Wilks λ = .69, partial χ2 = .17) and both ANOVAs: F(2,211) = 18.99, p < .01 (partial χ2 = .15) for the proximal style and F(2,211) = 41.37, p < .01 (partial χ2 = .28) for the distal style as dependent variable. From Table 2 it can be seen that the three group means again differed in the predicted direction, which was statistically confirmed by multiple comparisons (Tukey HSD tests).
To further explore the relationships between the three cultural models we conducted a discriminant function analysis in which we predicted the cultural model from the amount of proximal and distal parenting style. The first discriminant function allowed for a statistical significant discrimination of the three groups, χ2(4, N = 214) = 77.31, p < .01 (Wilks λ = 0.69). The second function did not make a significant contri- bution and accounted for less than 0.2% of the between- groups variance. 53.3% of the subjects could be correctly classified into their respective groups. Only nine subjects of the independent group were wrongly assigned to the interdepend- ent group and no misclassification resulted in the opposite direction. Figure 1 plots the discriminant scores of the subjects as points in the plane, spanned by the two discriminant func- tions. As can be seen from the three centroids, the first discrim- inant function accounted for differentiation among the three groups. The groups were ordered as predicted with the autonomous-related cultural model being between the inter - dependent and independent one.
The results thus demonstrated that the autonomous-related cultural model occupied a middle position concerning their parenting style.
Influence of the sociodemographic and cultural profile (hypothesis 2)
To assess the influences of the sociodemographic variables and the cultural models on the two parenting styles predicted in hypothesis 2, we performed two separate hierarchical multiple regression analyses. In a first step, the sociodemographic vari- ables age, gender, and birth order of children as well as age and years of formal education of mother were entered, and in the second step, cultural models. (Because the information about formal education of fathers was missing in two samples and its high correlation with the education of mother in the others, r = .78, p < .01, we omitted this variable from the regression analyses.) The categorical cultural models variable was split into two dummy-coded variables. Table 3 shows a similar pattern for both parenting styles. The three cultural models accounted for a statistical significant additional 8% (proximal) and 9% (distal) of the variance in the parenting styles beyond the variance explained by the sociodemographic variables. The sociodemographic variables alone accounted for 17% (proximal) and 32% (distal) of the variance. The variances of the two predictor groups overlapped considerably, but the cultural models contributed specifically to the prediction of the parenting styles and could not be completely reduced to the included sociodemographic variables.
Relations between distal and proximal parenting and their components (hypothesis 3)
To test the general relations between the parenting systems Pearson Product Moment correlations were calculated among the four parenting systems. Table 4 shows the correlations for the total sample. There were negative correlations between body contact and face-to-face exchange as well as object stim- ulation. There were also negative correlations between object stimulation and body stimulation. There was a positive corre- lation between body contact and body stimulation. Finally, also
INTERNATIONAL JOURNAL OF BEHAVIORAL DEVELOPMENT, 2009, 33 (5), 412–420 417
Figure 1. Discriminant function plot with circled centroids of the three sociocultural orientations.
412-420 JBD338441 Keller:210 x 280mm 30/07/2009 10:58 Page 417
in line with hypothesis 3, we found a significant negative corre- lation between the two parenting styles (r = –.47**, p < .01).
Discussion
This study was aimed at analyzing the structure of the relations between and the cross-cultural variation of the parenting systems: body contact, body stimulation, face-to-face context, and object stimulation, as well as the composite scores: body contact and body stimulation, representing a proximal parent- ing style, and face-to-face context and object stimulation, representing a distal parenting style. In order to test this struc- ture we analysed samples with different sociodemographic profiles that are supposed to enact different amounts and different combinations of these parenting systems. We can confirm the a priori classification of our samples with one exception. The Euro-American participants, who were
supposed to prefer the distal style of parenting, demonstrated the autonomous-related style instead, as they show relatively high amounts of body contact and body stimulation and a relatively low amount of face-to-face context. However, this finding can be regarded as a consequence of the emphasis on object stimulation. In many instances, mothers and babies jointly looked at the same object. This pattern of shared atten- tion in interactional situation is described in the literature mainly for children during the second half of the first year. It thus can be concluded that the Euro-American mothers interacted with their 3 months old babies as if they were in a different, later, developmental phase. This joint attention processes also imply that the mothers hold their babies in their laps, which increases the amount of body contact.
Basically, we can confirm different cultural preferences for the parenting systems as well as for the combinations into distal and proximal styles. With the discriminant function analysis we can demonstrate that there is a good fit between the theoreti- cally based grouping and the parenting behaviours (Figure 1). But of course not all mothers were classified correctly. This may due to the fact that there is also variance within one cultural context. In particular, independent orientated soci- eties are also characterized by a variety of individual styles and habits. Because of this, it is more likely that some families from independent living in cultural contexts show interdependent behaviour than the other way round. And of course there is also variance within the autonomous-related samples (Keller, 2007). Both pathways (the independent and the interdepend- ent) are present and important in these contexts but it is possible that some families may focus more on the independ- ent and some more on the interdependent pathway.
We can show that the sociodemographic profile substantially defines the parenting styles. Nevertheless, culture has a surplus meaning that accounts for additional significance in the predic- tion of parenting styles. It is presumably that this surplus meaning is due to other culture-defining variables not explic- itly coded in this study. This could be, e.g., religion, which has not yet been systematically studied with respect to early parent- ing from a cross-cultural perspective. Also, e.g., Confucianism is believed to strongly inform parenting goals and practices (e.g., Chao, 1994). No systematic comparison has been made between Chinese Confucian and non-Confucian parents with similar sociodemographic profiles. Future studies need to further disentangle the meaning of culture, especially with respect to the impact of different cultural traditions.
On an individual level, we can confirm our expectation that the parenting styles as well as the composing systems are negatively related to each other. Also here it becomes evident that face-to-face context and body contact are the dominant systems within the two styles. Our study contributes to the evidence that body contact and face-to-face contact are two alternative parenting strategies, as has been suggested by Bard and colleagues (2005). Body contact can be assumed to be an evolutionary old parenting system (Hofer, 1987). Moreover the somesthetic system, comprising kinaesthetic and cutaneous processes, is the earliest sensory system that develops in the human embryo (Montagu, 1986). Body contact and touch convey emotions (Stack, 2001) and thus meanings, especially in terms of love and care, empathy and the feeling of security. Body contact transmits interactional warmth directly and can be related to the development of closeness and acceptance of norms and values (Keller, 2003; Keller, Yovsi et al., 2004; MacDonald, 1992). Accordingly affectionate touching and
418 KELLER ET AL. / DISTAL AND PROXIMAL PARENTING AS ALTERNATIVE STRATEGIES
Table 3 Product moment correlations and results of blockwise multiple regression for dependent variables proximal parenting style (top) and distal parenting style (bottom)
_
β
Step Variable r Step 1 Step 2 R2 ΔR2
1 Age child –.13 –.01 .02 Birthorder child .33** .30** .25* Gender child .02 .03 .00 Age mother –.20** –.18* –.08 Education mother –.31** –.12 .12 .17**
2 Culture 1 –.39** –.24** Culture 2 .39** .25* .25** .08**
β
Step Variable r Step 1 Step 2 R2 ΔR2
1 Age child .14 –.02 –.04 Birth order child –.43** –.31** –.26** Gender child .04 .03 .05 Age mother .23** .11 .00 Education mother .50** .33** .11 .32**
2 Culture 1 .52** .31** Culture 2 –.50** –.19† .42** .09**
Notes. Birth order is coded 1 = first child and 2 = other; Gender is coded 1 = male, 2 = female; Culture 1 is coded 1 = independent and 0 = other; Culture 2 is coded 1 = interdependent and 0 = other.
N = 167; **p < .01; *p < .01; †p < .10.
Table 4 Correlations of parenting systems
Body Object Face-to-face stimulation stimulation context
Body contact .28** –.46** –.22** Body stimulation — –.39** .13** Object stimulation — .02** Face-to-face context —
Note. N = 214; **p < .01.
412-420 JBD338441 Keller:210 x 280mm 30/07/2009 10:58 Page 418
closeness are observed predominantly as parts of parenting strategies that are oriented towards an interdependent cultural model (Franco, Fogel, Messinger, & Frazier, 1996; Keller, Voelker, & Yovsi, 2005; Konner, 1976).
Face-to-face contact, on the other hand, can be regarded as a rather new parenting system (Keller, 2003). Also with respect to ontogenetic development, the visual system is the last to develop in the human embryo (Gottlieb, 1976). The distal contact through face-to-face communication can be regarded as an alternative to parenting mainly through body contact during the early months of life. Accordingly it has been repeat- edly demonstrated that physical proximity is an important factor in influencing the infant’s engagement during face- to-face play with more distant positions favouring more eye contact (e.g., Lavelli & Fogel, 1998; Stack, Arnold, Girouard, & Welbourne, 1999). However, it is physically not impossible to have body contact and a face-to-face context at the same time.
Our results gain strength from the fact that we analysed parenting strategies from participants with very different cultural models that favour different systems to different degrees. Therefore the negative correlation between the systems as well as the composite scores can be interpreted in terms of basic strategies. Our data confirm that the four parent- ing systems meaningfully describe the interactional experi- ences of small babies and express the cultural emphasis of particular combinations or styles. They also offer a fair method of conceptualizing parenting in culture, since the parenting system constitutes universals, that are differently embodied across cultures. It is also evident that no one strategy is better than another.
Our study also has constraints. The analyses are based on free-play interactional situations. Although we have demon- strated in different studies that free-play situations occur across diverse cultural contexts, their prominence, however, also differs (Keller, 2007). Further studies should include the analysis of everyday routine situations as well as care situations in order to gain a fuller understanding of children’s everyday experiences. Also the analysis of infants’ contribution to the interactional flow would be important for understanding the expression of cultural models in everyday interactional situations.
Generally, proximal and distal parenting can be regarded as parts of two alternative modes of self-development (Greenfield et al., 2003; Keller & Greenfield, 2000; Keller, Kärtner et al., 2005; Keller, Yovsi et al., 2004). So far, research on the origin of self-development has mainly been based on the analysis of face-to-face interactions (Kaye & Fogel, 1980; Keller, Yovsi et al., 2004; Rochat, 2001) with the implicit and explicit assump- tion that infants from birth on are perceived as independent agents with mental states, desires, and preferences. The results of our study underscore the importance of considering alter- native modes of self development. We propose a proximal pathway with an understanding of the self as a primarily rela- tional co-agent. Future studies should help to explain different developmental pathways with different developmental goals leading to different conceptions of the self.
References
Abels, M., Keller, H., Mohite, P., Mankodi, H., Shastri, J., Bhargava, S., Jasrai, S., & Lakhani, A. (2005). Early socialization contexts and social experiences
of infants in rural and urban Gujarat, India. Journal of Cross-Cultural Psychol- ogy, 36(6), 717–738.
Alvarez, A.T., Brenes, A., & Cabezas, M. (1990). Patrones de crianza en la familia costarricense. Actualidades en psicología. Instituto de Investigaciones psico - lógicas. Universidad de Costa Rica.
Bandura, A. (1989). Human agency in social cognitive theory. American Psychol- ogist, 44, 1175–1184.
Bard, K.A. (2002). Primate parenting. In M. Bornstein (Ed.), Handbook of parenting, vol. 2: Biology and ecology of parenting (2nd ed., pp. 99–140). Mahwah, NJ: Erlbaum.
Bard, K.A., Myowa-Yamakoshi, M., Tomonaga, M., Tanaka, M., Costall, A., & Matsuzawa, T. (2005). Group differences in the mutual gaze of chimpanzees (Pan Troglodytes). Developmental Psychology, 41(4), 616–624.
Chao, R.K. (1994). Beyond parental control and authoritarian parenting style: Understanding Chinese parenting through the cultural notion of training. Child Development, 65, 1111–1119.
Franco, F., Fogel, A., Messinger, D.S., & Frazier, C.A. (1996). Cultural differ- ences in physical contact between Hispanic and Anglo mother–infant dyads living in United States. Early Development and Parenting, 5(3), 119–127.
Gottlieb, G. (Ed.) (1976). Neural and behavioural specificity. New York: Academic Press.
Greenfield, P.M. (1994). Independence and interdependence as developmental scripts: Implications for theory, research, and practice. In P.M. Greenfield & R.R. Cocking (Eds.), Cross-cultural roots of minority child development (pp. 1–40). Hillsdale, NJ: Erlbaum.
Greenfield, P.M. (2004). Weaving generations together: Evolving creativity in the Maya of Chiapas. Santa Fe, NM: Sar Press.
Greenfield, P.M., Keller, H., Fuligni, A., & Maynard, A. (2003). Cultural pathways through universal development. Annual Review of Psychology, 54, 461–490.
Hetherington, M., & Frankie, G. (1967). Effects of parental dominance, warmth, and conflict on imitation in children. Journal of Personality and Social Psychol- ogy, 6(2), 119–125.
Hofer, M.A. (1987). Early social relationships: A psychobiologist’s view. Child Development, 58, 633–647.
Kağıtc‚ıbas‚ı, C. (1996). Family and human development across countries: A view from the other side. Hillsdale, NJ: Erlbaum.
Kaye, K., & Fogel, A. (1980). The temporal structure of face-to-face communi- cation between mothers and infants. Developmental Psychology, 16(5), 454–464.
Keller, H. (2002). Culture and development: Developmental pathways to indi- vidualism and interrelatedness. In W.J. Lonner, D.L. Dinnel, S.A. Hayes, & D.N. Sattler (Eds.), OnLine Readings in Psychology and Culture, Western Washington University, Department of Psychology, Center for Cross-Cultural Research. Available at: http://www.wwu.edu/~culture.
Keller, H. (2003). Socialization for competence: Cultural models of infancy. Human Development, 46(5), 288–311.
Keller, H. (2007). Culture of infancy. Mahwah, NJ: Erlbaum. Keller, H., Abels, M., Lamm, B., Yovsi, R.D., Voelker, S., & Lakhani, A. (2005).
Ecocultural effects on early infant care: A study in Cameroon, India, and Germany. Ethos, 33(4), 512–541.
Keller, H., Borke, J., Yovsi, R., Lohaus, A., & Jensen, H. (2005). Cultural orien- tations and historical changes as predictors of parenting behaviour. Inter - national Journal of Behavioral Development, 29(3), 229–237.
Keller, H., Demuth, C., & Yovsi, R.D. (2008). The multi-voicedness of in - dependence and interdependence: The case of Cameroonian Nso. Culture & Psychology, 14(1), 115–144.
Keller, H., & Greenfield, P.M. (2000). History and future of development in cross-cultural psychology. In C. Kağıtc‚ıbas‚ı & Y.H. Poortinga (Eds.), Millen- nium Special Issue of the Journal of Cross-Cultural Psychology, 31(1), 52–62.
Keller, H., Hentschel, E., Yovsi, R.D., Abels, M., Lamm, B., & Haas, V. (2004). The psycho-linguistic embodiment of parental ethnotheories. A new avenue to understand cultural differences in parenting. Culture & Psychology, 10(3), 293–330.
Keller, H., Kärtner, J., Borke, J., Yovsi, R.D., & Kleis, A. (2005). Parenting styles and the development of the categorial self: A longitudinal study on mirror self recognition in Cameroonian Nso farming and German families. International Journal of Behavioral Development, 29(6), 496–504.
Keller, H., Kuensemueller, P., Abels, M., Voelker, S., Yovsi, R., Jensen, H., Papaligoura, Z., Lohaus, A., Rosabal-Coto, M., Kulks, D., & Mohite, P. (2005). Parenting, culture, and development: A comparative study. San José, CR: Instituto de Investigaciones Psicológicas.
Keller, H., & Lamm, B. (2005). Parenting as the expression of sociohistorical time: The case of German individualism. International Journal of Behavioral Development, 29(3), 238–246.
Keller, H., Lamm, B., Abels, M., Yovsi, R.D., Borke, J., Jensen, H., Papaligoura, Z., Holub, C., Wingshan, L., Tomiyama, A.J., Su, Y., Wang, Y., & Chaudhary,
INTERNATIONAL JOURNAL OF BEHAVIORAL DEVELOPMENT, 2009, 33 (5), 412–420 419
412-420 JBD338441 Keller:210 x 280mm 30/07/2009 10:58 Page 419
420 KELLER ET AL. / DISTAL AND PROXIMAL PARENTING AS ALTERNATIVE STRATEGIES
N. (2006). Cultural models, socialization goals, and parenting ethnotheories: A multi-cultural analysis. Journal of Cross-Cultural Psychology, 37(2), 155–172.
Keller, H., Lohaus, A., Kuensemueller, P., Abels, M., Yovsi, R.D., Voelker, S., Jensen, H., Papaligoura, Z., Rosabal-Coto, M., Kulks, D., & Mohite, P. (2004). The bio-culture of parenting: Evidence from five cultural communi- ties. Parenting: Science and Practice, 4(1), 25–50.
Keller, H., Lohaus, A., Völker, S., Elben, C., & Ball, J. (2003). Warmth and contingency and their relationship to maternal attitudes toward parenting. The Journal of Genetic Psychology, 164(3), 275–292.
Keller, H., Papaligoura, Z., Kuensemueller, P., Völker, S., Papaeliou, C., Lohaus, A., Lamm, B., Kokkinaki, T., Chrysikou, E.G., & Mousouli, V. (2003). Concepts of mother–infant interactions in Greece and Germany. Journal of Cross-Cultural Psychology, 34(6), 677–689.
Keller, H., Voelker, S., & Yovsi, R.D. (2005). Conceptions of parenting in different cultural communities: The case of West African Nso and Northern German women. Social Development, 14(1), 158–180.
Keller, H., Yovsi, R.D., Borke, J., Kärtner, J., Jensen, H., & Papaligoura, Z. (2004). Developmental consequences of early parenting experiences: Self regulation and self recognition in three cultural communities. Child Develop- ment, 75(6), 1745–1760.
Keller, H., Yovsi, R.D., & Voelker, S. (2002). The role of motor stimulation in parental ethnotheories: The case of Cameroonian Nso and German women. Journal of Cross-Cultural Psychology, 33(4), 398–414.
Konner, M.J. (1976). Maternal care, and development among the !Kung. In R. Lee, & I. DeVore (Eds.), Kalahari hunters-gatherers: Studies of the !Kung San and their neighbors (pp. 218–245). Cambridge, MA: Harvard University Press.
Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge: Cambridge University Press.
Lavelli, M., & Fogel, A. (1998). Developmental changes in early mother–infant communication. Infant Behavior and Development, 21, Special ICIS Issue, 522.
Lavelli, M., & Fogel, A. (2002). Developmental changes in mother–infant face- to-face communication: Birth to 3 months. Developmental Psychology, 38(2), 288–305.
LeVine, R.A. (1970). Cross-cultural study in child psychology. In P.H. Mussen (Ed.), Carmichael’s manual of child psychology (pp. 559–612). London: Wiley.
LeVine, R.A. (2004). Challenging expert knowledge: Findings from an African study of infant care and development. In U.P. Gielen & J.L. Roopnarine (Eds.), Childhood and adolescence in cross-cultural perspective (pp. 149–165). Westport, CT: Greenwood Publishing.
Liu, M., Chen, X., Rubin, H.H., Zheng, S., Cui, L., Li, D., Chen, H., & Wang, L. (2005). Autonomy- vs. connectedness-oriented parenting behaviours in
Chinese and Canadian mothers. International Journal of Behavioral Develop- ment, 29(6), 489–495.
MacDonald, K.B. (1992). Warmth as a developmental construct: An evolution- ary analysis. Child Development, 63, 753–773.
Montagu, A. (1986). Touching: The significance of the skin (3rd ed.). New York: Harper and Row.
Mundy-Castle, A.C. (1974). Social and technological intelligence in western and non-western cultures. Universita, 4, 46–52.
Nsamenang, A.B. (1992). Human development in cultural context: A Third World perspective. Newbury Park, CA: SAGE.
Nsamenang, A.B., & Lamb, M.E. (1994). Socialization of Nso children in the Bamenda grassfields of Northwest Cameroon. In P.M. Greenfield & R.R. Cocking (Eds.), Cross-cultural roots of minority child development (pp. 133–146). Hillsdale, NJ: Erlbaum.
Richman, A.L., Miller, P.M., & LeVine, R.A. (1992). Cultural and educational variations in maternal responsiveness. Developmental Psychology, 28(4), 614–621.
Rochat, P. (2001). Origins of self-concept. In G. Bremner & A. Fogel (Eds.), Blackwell handbook of infant development (pp. 191–212). Oxford: Blackwell.
Rogoff, B. (2003). The cultural nature of human development. New York: Oxford University Press.
Serpell, R., & Hatano, G. (1996). Education, schooling, and literacy. In J.W. Berry, P.R. Dasen, & T.S. Saraswathi (Eds.), Handbook of cross-cultural psychology, Vol. 2: Basic processes and human development (2nd ed., pp. 339–376). Boston: Allyn & Bacon.
Stack, D.M. (2001). The salience of touch and physical contact during infancy: Unraveling some of the mysteries of the somesthetic sense. In G. Bremner & A. Fogel (Eds.), Blackwell handbook of infant development (pp. 351–378). Oxford: Blackwell.
Stack, D.M., Arnold, S.L., Girouard, N., & Welbourne, B. (1999). Infants’ re - actions to maternal unavailability in very low birth weight preterm and full- term infants. Paper presented at the Biennial Meeting of the Society for Research in Child Development, Albuquerque, NM, USA, April.
Tapia Uribe, F.M., LeVine, R.A., & LeVine, S.E. (1994). Maternal behavior in a Mexican community: The changing environments of children. In P.M. Greenfield & R.R.T. Cocking (Eds.), Cross-cultural roots of minority child development (pp. 41–54). Hillsdale, NJ: Erlbaum.
Voelker, S. Keller, H., Abels, M., Ahringhoff, B., Borke, J., Irmler, U., Kulks, D., Wilken, M., & Yovsi, R. (1999). Manual to the assessment of early context, and interactional experiences in diverse cultural settings. Department of Development and Culture, Faculty of Human Sciences, University of Osnabrück, Germany.
412-420 JBD338441 Keller:210 x 280mm 30/07/2009 10:58 Page 420
