This article was downloaded by: [University of California Davis]On: 21 November 2014, At: 23:50Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Early Child Development and CarePublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/gecd20

Longitudinal Patterns of Parental Support as Predictorsof Children's Competence MotivationHans th. Meij a , J. Marianne Riksen‐Walraven b & Cornelis F.M. Van Lieshout b

a Department of Developmental Psychology , University of Nijmegen (now at the NetherlandsInstitute of Care and Welfare , Utrechtb Department of Developmental Psychology , University of NijmegenPublished online: 07 Jul 2006.

To cite this article: Hans th. Meij , J. Marianne Riksen‐Walraven & Cornelis F.M. Van Lieshout (2000) Longitudinal Patternsof Parental Support as Predictors of Children's Competence Motivation, Early Child Development and Care, 160:1, 1-15, DOI:10.1080/0030443001600101

To link to this article: http://dx.doi.org/10.1080/0030443001600101

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in thepublications on our platform. However, Taylor & Francis, our agents, and our licensors make no representationsor warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Anyopinions and views expressed in this publication are the opinions and views of the authors, and are not theviews of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should beindependently verified with primary sources of information. Taylor and Francis shall not be liable for any losses,actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoevercaused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Early Child Development and Care, 2000, VoL 160, pp. 1-15Reprints available directly from the publisherPhotocopying permitted by license only

© 2000 OFA (Overseas Publishers Association) N.V.Published by license under

the Cordon and Breach Publishers imprintPrinted in Singapore.

Longitudinal Patterns of Parental Supportas Predictors of Children's CompetenceMotivation

HANS TH. MEIJ1, J. MARIANNE RIKSEN-WALRAVEN2,* andCORNELIS F.M. VAN LIESHOUT2

1 Department of Developmental Psychology, University of Nijmegen(now at the Netherlands Institute of Care and Welfare, Utrecht)2Department of Developmental Psychology, University of Nijmegen

(Received 2 December 1999)

The competence motivation of 77 children at 12 and 30 months was examined inrelation to the quality of support they received from their parents at 6, 9, 12, 18, and30 months. First, the quality of parental support at the separate ages was used topredict the children's competence motivation. Although the quality of parental supportwas moderately stable across time, only weak relationships were found between thequality of parental support the children received at earlier ages and their later competencemotivation. Next, the individual patterns of parental support across time were examinedin relation to children's competence motivation. Using cluster analysis, four groups ofparent-child pairs were identified with similar patterns of parental support across time:(1) increasing, (2) decreasing, (3) stable high, and (4) stable low quality of support.The children in these four groups differed significantly with regard to competencemotivation. Remarkably low levels of competence motivation were found at 30 monthsfor children receiving a decreased quality of parental support over time.

Key words: Competence motivation, parental support, longitudinal, parent-infantinteraction

Robert White (1959) described children's competence motivation as an innate anduniversal motive "to interact effectively with the environment" (p.297). AlthoughWhite did not pay much attention to the existence and development of inter-individual differences in the strength of this motive, such individual variation hasbeen the topic of considerable theorizing and research. Harter (1978, 1981)elaborated on White's conceptual framework and was the first to emphasize the

*Correspondence: J. Marianne Riksen-Walraven, Department of Developmental Psychology,University of Nijmegen, PO Box 9104, 6500 HE Nijmegen, The Netherlands. E-mail: riksen@psych.kun.nl

1

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a D

avis

] at

23:

50 2

1 N

ovem

ber

2014

2 H.TH. MEIJ et al.

social basis of competence motivation. Since then, numerous studies have examinedthe effects of children's social experiences and especially their interactions with theprimary caregiver on the development of competence motivation in the first yearsof life.

The effects of children's social experiences on their competence motivation areassumed to be mediated by the development of such personal agency beliefs asperceptions of control, perceptions of competence, and self-efficacy beliefs (Bandura,1977; Ford & Thompson, 1985; Lewis & Goldberg, 1969; Riksen-Walraven, 1978;Skinner, 1986; Watson, 1966). According to this point of view, infants experiencethemselves as competent when their acts elicit a response from the environment Suchexperiences promote the development of a generalized sense of self as an effectiveagent, which then provides the motivational basis for further exploration and masteryof the environment Given that the most prominent act-effect contingencies in earlyinfancy are experienced in interaction with the social environment, the extent towhich caregivers respond to their children's signals and thus provide opportunitiesfor the infants to experience their own effectiveness may be considered a powerfuldeterminant of children's competence motivation. And indeed, numerous empiricalstudies have shown parental responsiveness to foster motivational development ininfancy and beyond (Ramey, Starr, Pallas, Whitten & Reed, 1975; Riksen-Walraven,1978; Riksen-Walraven & Van Aken, 1997; Watson, 1972).

After the first year of life, other aspects of parental behavior gain importance aspartial determinants of children's competence motivation. The manner in whichparents support children during problem solving or the performance of a given task,for example, largely determines the extent to which children experience themselvesas competent Parental encouragement and emotional support in the face offrustration typically stimulate children to persist on a task. Careful structuring ofthe task and the provision of clear instructions attuned to the child's cognitive levelmay also increase children's chances of success. By not interfering with theperformance and allowing children to act on their own, moreover, the children'sattribution of success to their own competence is also promoted. Empirical studieswith toddlers and older preschoolers have clearly shown these aspects of parentalsupport to indeed be related to children's task motivation (Busch-Rossnagel, Knauf-Jensen & DesRosiers, 1995; Frodi, Bridges & Grolnick, 1985; Heckhausen, 1993;Lutkenhaus, 1984; Maslin-Cole, Bretherton & Morgan, 1993; Skinner, 1986; Wachs,1987).

In sum, there is ample evidence that the quality of parental support affectschildren's competence motivation in the first years of life. The majority of thestudies in this area have measured quality of parental support at only a single pointin time. That is, the quality of parental support is implicitly assumed to be relativelystable over time, which implies that assessment on one or two occasions is sufficientto predict children's competence motivation. The question, however, is whetherpossible changes in parental support over time should not be considered in attemptsto explain the level and development of children's competence motivation. Thatis, intra-individual variation in the quality of parental support may occur and alsobe related to changes in children's competence motivation.

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a D

avis

] at

23:

50 2

1 N

ovem

ber

2014

LONGITUDINAL PATTERNS 3

One reason to expect stability in the quality of support provided by parents issimply that the quality of parenting is largely determined by relatively enduringparental characteristics that are, in part, the product of the parents' developmentalhistories (Belsky, 1984; Belsky, Crnic & Woodworth, 1995; Kochanska, Clark &Goldman, 1997) and their genetic endowment (Perusse, 1994). Parental behavioris also obviously influenced by much more variable factors, such as the level of lifestress and social support experienced by the parents (Belsky, 1984; Teti & Gelfand,1992; Simons, Lorenz, Wu & Conger, 1993; Vaughn, Egeland, Waters & Sroufe,1979). For this reason, we can expect at least some change in the quality ofparenting over time. The moderate, stability that has been found in the quality ofparental support across the first few years of life (cf . Pianta, Sroufe & Egeland, 1989)leaves room for the existence of different patterns of parental support across time.

Although the importance of studying patterns of development at the individuallevel has been recognized in developmental psychology (Burchinal & Appelbaum,1991), longitudinal research has not addressed individual patterns of parentalsupport over time and how such patterns may relate to children's development Inthe present study, we therefore examined individual patterns of parental supportacross a two-year period and related these observed patterns of support to thedevelopment of children's competence motivation.

METHOD

Participants

The present report is based upon a sample of 77 Caucasian first-born children (41boys, 36 girls) and their primary caregivers (76 mothers, 1 father). The parents andchildren were seen when the children were 6, 9, 12, 18, and 30 months old.Originally, 78 families participated, but one family emigrated when the child was12 months old. The sample comprised no premature infants or infants with abnormalmedical histories. All of the children lived in two-parent, working-class families(semi-skilled and unskilled workers). The recruitment of the families was based onthe municipal records from the city of Nijmegen.

The study was originally designed to be an intervention study. Between the agesof 6 and 12 months, two different intervention programs were carried out. Bothprograms were aimed at improving the quality of support parents provide for theirinfants (i.e. their sensitive responsiveness) and thereby the infants' competencemotivation. Each program was carried out in one randomly selected third of thetotal sample; the remaining third of the sample served as the control group.However, the interventions did not produce the expected differences in quality ofparental support or children's competence motivation (Meij, 1992). In addition,the stability of parental support and the children's competence motivation over timedid not differ for the intervention versus control groups. In light of the above, itwas decided that the combined sample could be considered as homogeneous inthese respects.

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a D

avis

] at

23:

50 2

1 N

ovem

ber

2014

4 H.TH. MEIJ et al.

Procedure and Measures

Quality of Parental Support to the Child

At all five ages, scores for the quality of parental support were derived fromvideotaped parent-child interaction episodes. At 6, 9, and 12 months, we rated theparents' behavior towards the child using the Ainsworth Sensitivity scale (cf . Ainsworthet al, 1978); at 12, 18, and 30 months, the quality of parental support was assessedusing five rating scales developed by Erickson, Sroufe, and Egeland (1985).

Ainsworth's Sensitivity Ratings at 6, 9, and 12 Months

At these ages, parents' sensitive responsiveness to their children's signals was takento be an indicator of the quality of parental support Sensitive responsiveness wasobserved in a 10-minute play session videotaped at home. The parent was askedto play with die child, using a standard set of toys. Sensitive responsiveness was ratedfrom videotape by trained raters using Ainsworth's 9-point Sensitivity Scale (cf.Ainsworth et al., 1978). The inter-rater reliability was high (Pearson correlations of.86, .84, and .86 at the three consecutive ages)

The Erickson et ah (1985) Ratings at 12, 18, and 30 Months

From 12 months onwards, the parent-child interactions were videotaped in aninstruction situation at the laboratory. The parent was asked to help the childperform a series of tasks; for example, putting together a jigsaw puzzle or buildinga tower with cubes. At each of the three ages, the parents' behavior was rated usingfive 7-point rating scales (cf. Erickson et al., 1985) intended to capture differentaspects of the support provided to children: supportive presence, respect for diechild's autonomy, adequacy of structure and limit setting, quality of instruction, andhostility.

Trained raters independently scored all of the interactions. The inter-raterreliabilities, computed on 18 sessions, were .96, .95, .95, .93, and 1.00, respectively,for the above mentioned scales. At all of the three ages, the five scale scoresrepresenting different aspects of parental support proved to be highly consistent(Cronbach's a of .83, .84, and .82, at 12, 18, and 30 months, respectively). Wedierefore decided to sum the five scores with the score for hostility reversed tocreate a single score for the quality of parental support during the instructionalsituation.

At the age of 12 months, the quality of parental support was assessed both at home(using the Ainsworth scale) and at the lab (using the Erickson et al scales). Thisallowed us to examine the consistency of parental support across situations andcheck the concurrent validity of the rating scales we used. The correlation betweenthe Ainsworth's Sensitivity score at home and the composite score on the Ericksonet al scales at the lab was significant (r(75) = .52, p < .01). In subsequent analyses,we used the Ainsworth Sensitivity score to indicate quality of parental support at6 and 9 months and the composite score on the Erickson et al. scales as a measureof parental support at 12, 18, and 30 months.

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a D

avis

] at

23:

50 2

1 N

ovem

ber

2014

LONGITUDINAL PATTERNS 5

Children's Competence Motivation

Competence motivation was assessed at the ages of 12 and 30 months using variousprocedures and measures.

Assessment at 12 Months

According to White (1959) and others (cf. MacTurk, Morgan & Jennings, 1995),the competence motivation of infants is reflected by their exploratory behavior. Wetherefore took children's exploratory competence at 12 months of age to be anindicator of their competence motivation. The children's exploratory behavior ina 12-minute structured exploration task videotaped at home was scored using asystem designed by Van den Boom (1994). The scoring system includes four behavioralcategories indicating increasing levels of exploration: (1) simple single exploration:global inspection or touching of one single object or part of an object, (2) detailedsingle exploration: inspection in detail or manipulation with interest in the workingof one single object or part of an object, (3) simple relational exploration:manipulation of one object or part of an object, combined with manipulation orinspection of another object or part of an object, and (4) detailed relationalexploration: relating two different objects or parts of objects through differentmanipulations or examination of the relationship between one object or part ofan object and two or more other objects or parts of objects. Each exploratorybehavior of the child was scored as occurring at one of these levels. The inter-raterreliabilities (Pearson correlations) were satisfactory .82, .80, .85, and .88 for the foursuccessive levels of exploration. On the basis of these codings, the mean level ofexploration for each child was computed and taken to be an indicator of the child'scompetence motivation at 12 months.

Assessment at 30 Months

At the age of 30 months, four measures commonly used to assess competencemotivation at this age (MacTurk et al, 1995) were obtained and combined into onecomposite score.

First, children's exploratory competence was assessed in an 8-minute structuredexploration task using a modified version of Banta's curiosity box in the lab (Banta,1970). The children's exploratory behavior was coded using the same fourobservational categories as in the exploration task at 12 months. The inter-raterreliabilities (Pearson correlations) were satisfactory .74, .85, .85, and .84 for the foursuccessive levels of exploration. Following the same procedure used at 12 months,the mean level of exploration for each child was computed and taken to be anindicator of exploratory competence at 30 months.

A second score for competence motivation at 30 months was based on thechildren's behavior during a problem-solving task also videotaped in the lab. Duringthis task, mother and child were asked to catch fish using a magnet attached toa rope with two handles, one for the child and one for the parent. Children'scompetence motivation during problem solvingvtos assessed using two 7-point rating scales(Erickson et al., 1985), one for their persistence, the other for the degree of

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a D

avis

] at

23:

50 2

1 N

ovem

ber

2014

6 H.TH. MEIJ et al.

enthusiasm displayed dviring the task. The inter-rater reliability, based on 20 randomlyselected subjects, was found to be .99 for persistence and .98 for enthusiasm(Pearson correlations). The scores for persistence and enthusiasm was significantlyinter-related (Pearson's r = .86), and were summed to obtain a single score forcompetence motivation during problem solving.

Two additional scores for children's competence motivation at 30 months werederived from their parents' answers to two questionnaires concerned with thechildren's curiosity and striving for competence during daily activities (Purmann, 1988).The internal consistencies.of the questionnaires were acceptable in our sample, withCronbach's a of .72 (37 items) and .78 (36 items), respectively.

The four measures of competence motivation, i.e., exploratory competence,competence motivation during problem solving, curiosity, and striving forcompetence, were positively interrelated, with Pearson correlations between .26 and.77. We computed a single composite score for children's competence motivation bysumming the standardized scores for the four measures.

RESULTS

Stability of Parental Support and Children's Competence Motivation

As shown in Table 1, the stability coefficients for parents' support for their childrenbetween the ages of 6 and 30 months were moderate to high, with a range from.37 to .67. It is remarkable that the height of the correlations between adjacentmeasurement points (6 to 9 months, 9 to 12 months, 12 to 18 months, and 18 to30 months) is about the same while the length of the intervals between themeasurements increases considerably. Moreover, the stability of parental supportfor the 12-month period between 18 and 30 months proves to be higher than forthe 12-month period between 6 and 18 months (Pearson correlations of .66 and.45, respectively). These results suggest that the quality of parental support becomesincreasingly stable during the first few years of life. The stability of the children'scompetence motivation from 12 to 30 months was also found to be significant (r(77)= .22, p < .05) but much lower than the stability of parental support

Children's Competence Motivation in Relation to Parental Support: MultipleRegression Analyses

First, we used correlational and multiple regression analyses to examine whetherchildren's competence motivation at 12 and 30 months was related to the supportthey received from their parents up to those ages. Table 2 shows that children'scompetence motivation at 12 months could not be predicted from the supportprovided by their parents at 6, 9, and 12 months. The multiple regression analysisyielded a non significant multiple correlation (R= .12; F($,il) = .33). In addition,the relation between children's competence motivation at 30 months and theirparents' support scores at 6,9,12,18, and 30 months was also weak. The multiple

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a D

avis

] at

23:

50 2

1 N

ovem

ber

2014

LONGITUDINAL PATTERNS

Table 1 Stability of Parental Support From 6 to 30 Months

.53"*

.61***.45***.67—.61—

.37*

.54*

.55*

.66*

AGE 6 ms 9 ms 12 ms 18 ms 30 ms

6 ms — .65***9ms —

12 ms18 ms30 ms

*7><.O1, —/x.001, one-tailed; N - 76-77.

regression analysis yielded a multiple correlation of .38, which failed to reachsignificance (F(5,68) = 2.34; p= .051). As shown in Table 2, competence motivationat 30 months was only significantly related to concurrently observed parentalsupport In other words, competence motivation at 30 months was not predictedby earlier parental support and only weakly predicted by concurrent parentalsupport

The results of the multiple regression analyses thus suggest that parental supporthas litde or no predictive value for children's competence motivation during thefirst few years of life. But multiple regression analyses do not tell the whole story.Although these analyses use parental support scores at various ages to predictchildren's competence motivation, they do not take into account the pattern ofsupport children experience over age. In another set of analyses we thereforeidentified individual patterns of parental support over time and examined whetherthese longitudinal patterns of support better predicted children's competencemotivation.

Table 2 Pearson and Multiple Correlations BetweenParental Support and Children's Competence Motivation

Parental support

6 ms

9 ms

12 ms

18 ms

30 ms

Competence

12 ms

.09

.08

.10

—

MR= .12

motivation

30 ms

.17

.13

.12

.18

.34**MR= .38

*/><.05, **/K.01, one-tailed; N = 75-77

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a D

avis

] at

23:

50 2

1 N

ovem

ber

2014

8 H.TH. MEIJ et al.

Patterns of Parental Support across Time in relation to Children's CompetenceMotivation

Identifying Different Patterns of Parental Support Over Time

To identify different longitudinal patterns of parental support, we chose threeparameters to characterize the individual patterns of support over time: (1) themean level of parental support over time, (2) the variability in parental supportover time, and (3) the increase vs. decrease in parental support over time. For eachindividual child, we computed the three pattern parameters from his or her parentalsupport scores at the five ages. In statistical terms, the three parameters can bedescribed as the mean of the subject's parental support scores across the five ages,the standard deviation for these five scores, and the slope of the individual regressionline of parental support on age. Given that the parental support scores were notcomparable across time, we standardized the scores before computing the parameters.

Next, the three parameters characterizing the individual patterns of parentalsupport over time were introduced into a hierarchical cluster analysis using Ward'smethod. The purpose of this analysis was to cluster subjects together into groupssuch that the patterns of parental support over time are relatively homogeneouswithin the individual groups and the general pattern for a single group clearlydiffers from that for the other groups. Cluster analysis has also been used in otherlongitudinal studies to identify individuals with similar patterns of development(Burchinal & Appelbaum, 1991; Ramey, Lee 8c Burchinal, 1989).

The analysis yielded four meaningful clusters of subjects. Solutions with morethan four clusters were also possible but rejected due to a small number of subjectsper group. The mean patterns of parental support characterizing the four clustersof subjects are depicted in Figure 1. As can be seen, Pattern 1 (n= 18) is characterizedby an increase of parental support with age. Pattern 2 (n = 13) is characterized bya decrease in parental support with age. Pattern S (n = 13) show a stable level of verylow quality parental support over time. Pattern 4 (n = 30) shows a stable high levelof parental support over time.

Differences Among the Patterns

To examine in what respects the four longitudinal patterns of support differed fromeach other, we performed three ANOVA's on the separate pattern parameters(mean level, variability, and increase vs. decrease of parental support across time).These analyses yielded a significant Pattern effect for each of the parameters. Inother words, the four patterns indeed differed with regard to the three parameterscharacterizing their pattern of parental support over time. The F-values and meanscores for each of the four patterns are shown in Table 3. Additional Scheffe testsshowed each pattern to significantly differ from some other pattern on at least onepattern parameter. For example, the increasing Pattern I differed from the decreasingPattern 2 on the slope but not the mean level or variability of parental support overage. The increasing Pattern 1 significantly differed from the stable high Pattern 4on both slope and variability and from the stable low Pattern 3 on all three patternparameters.

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a D

avis

] at

23:

50 2

1 N

ovem

ber

2014

LONGITUDINAL PATTERNS 0

Itable 3 Mean (Standard) Scores on the Three Pattern Parameters for the Four Patterns

Parameter

MeanVariabUitySlope

1Increasing

•n.b».77.

1.26b

Pattern

2Decreasing

- 2 6 . c.99.

-1.42c

3Stable Low

-1.15c-1.11b

.07.

Stable High

54b.07c

-.17.

F

13.86"'36.62*"75.98'"

"/K.001, *cclls sharing no common subscript in a row differ significantly (Schcffe test; p<.OS)

Patterns of Parental Support in Relation to Children's Competence Motivation

To examine possible differences in the competence motivation of the children inthe four clusters, we performed a 4 (Pattern) X 2 (Sex of child) MANOVA on thestandardized competence motivation scores at 12 and 30 months with age as awithin-subjects factor. This analysis yielded a significant Pattern effect (F(3,66) =5.50; p < .01), indicating that children with different patterns of parental supportindeed differ in competence motivation. No effects involving sex or age of the childwere found.

Pairwise contrasts of the patterns showed the children with Pattern 2 (decreasingsupport) to differ significantly in competence motivation from the children widiPattern 1 (increasing support), (i^l.29) = 14.48; p < .001) and the children withPattern 4 (stable high level of support), (F{1,41) = 15.51; /x.OOl). The resultspresented in Table 4 (see also Figure 1) show children who experienced a decreasein parental support over time (Pattern 2) to have the lowest competence motivation,particularly at 30 months. It is rather surprising that these children tend to scoreeven lower on competence motivation than children widi Pattern 3 (stable low levelof support). In sum, the prognosis for the development of competence motivationis worst for children who receive declining parental support, even worse than forchildren who experience low but stable support over time. It is interesting to notethat we did not find the opposite: children with Pattern 1 (increasing support) didnot score better than children widi Pattern 4 (stable high level of support). Oneexplanation for this finding may be that a decline in parental support has an almost

Table 4 Mean Standardized Competence Motivation Scores at 12 and 30 Months for the FourPatterns

Competencemotivation

12 ms30 ms

IfIncreasing

.23 ,

.24,

2*Decreasing

- . 4 4 .-91b

Pattern

3A

Stable Low

- . 4 3 ,- . 1 1 *

Stable High

.19,

.22.

F

2.455.17"

"p<.0\, *cclls sharing no common subscript in a row differ significantly (Scheffe test; /»<.05)

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a D

avis

] at

23:

50 2

1 N

ovem

ber

2014

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a D

avis

] at

23:

50 2

1 N

ovem

ber

2014

LONGITUDINAL PATTERNS 11

immediate effect on children's competence motivation, while an increase in parentalsupport may have a more gradual or delayed effect It is also possible, however, thatthe quality of parental support beyond a certain level does not substantially contributeto the development of children's competence motivation. If this should be the case,it may also offer an explanation for the relatively low correlations between competencemotivation and parental support observed in our sample (see Table 2).

To explore this issue further, we inspected the correlations between the meanlevel of parental support collapsed across age and the child's competence motivationat 30 months in subsamples of parents with above and below average scores onparental support For the entire sample, the correlation between the mean levelof support and the child's competence motivation at 30 months was significant butrelatively low (r(74) = .25, p< .05). Within the subsample of parents providing belowan average level of support, the correlation was r(36) = .35, p < .05. However, inthe subsample of parents providing an above average level of support, the correlationwas low and non-significant r(38) = .15. This suggests that it is not high qualityparental support that positively affects children's competence motivation but, rather,a lack of quality support that negatively affects competence motivation, hi otherwords, being supportive beyond a certain level does not appear to further thedevelopment of competence motivation.

DISCUSSION

In accordance with other studies (cf. Pianta, Sroufe & Egeland, 1989), we foundparental support during the first years of a child's life to be moderately stable.Although we expected some degree of stability, the level of the stability was remarkablein light of the characteristics of our sample, which consisted of lower-class familieswith first-born children. Lower-class families (even in The Netherlands, where socialclass differences are relatively small), are expected to experience relatively highlevels of life stress, which can create unstable parental support In a sample of first-bom children, moreover, one can expect lower levels of stability for parental supportprecisely because the parents have less experience with child rearing and maytherefore be more insecure and more inclined to change their child-rearing strategies.

In contrast to the relative stability of the parental support the stability of thechildren's competence motivation was found to be low. These results are consistentwith previous research findings from van den Boom and Kramers (1991), who foundno significant relations between children's exploratory competence at the age of9 months and the ages of 18 and 24 months but significant stability in parentalsupport between these ages. Barret Morgan and Maslin-Cole (1993) also found lowstability for task persistence between 12 and 25 months and suggest that the lowstability may be due to a developmental transition in the competence motivationof children during the second half of the second year of life. In general, competencemotivation has been found to be only moderately stable during infancy andtoddlerhood (Maslin-Cole et aL, 1993; MacTurk et al., 1995).

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a D

avis

] at

23:

50 2

1 N

ovem

ber

2014

12 H.TH. MEIJ Bt al.

Despite the moderate to high stability of the parental support in our sample, wefound different patterns of parental support over time. The first pattern wascharacterized by a significant increase in parental support over time; the secondpattern was characterized by a significant decrease in parental support over time.In the remaining two patterns, the parental support was stable and either high orlow. The observed increase in parental support may be associated with the fact thatour sample consisted of first-born children. Parents of first-born children have lesschild-rearing experience and may therefore gradually develop these skills. Theobserved decrease in parental support may be related to a different characteristicof the sample. That is, a decrease in the quality of parental support may be causedby relatively high levels of life stress among this subgroup of the sample and maynot be found to such extent in a higher-or middle-class samples (Vaughn, Egeland,Waters & Sroufe, 1979).

In general, higher levels of parental support were related to higher levels ofcompetence motivation. However, for Pattern 1, with an increase in parental support,we found no increase in the children's competence motivation over time. Furtherinspection of the data shows the relation between the quality of parental supportand children's competence motivation to decrease when the quality of parentalsupport goes beyond a particular level. Children whose parents were extremelysupportive at later ages (Pattern 1) did not score extremely high on competencemotivation at 30 months. In fact, their scores were comparable to those for childrenreceiving a moderate level of parental support (Pattern 4). These findings can beinterpreted in terms of the "good enough environment" hypothesis (cf. Scarr 1992;1993; Baumrind, 1993). That is, for the children with the stable high and increasingsupport patterns, the quality of parental support can be considered as "goodenough" for competence motivation to develop normally in children. Comparable"threshold effects" have been found in two other studies. In a longitudinal studywith preschool-aged children and their parents, Roberts (1986), found an s-shapedrelationship between parental warmth and children's "general competence", whichis a measure with a strong motivational component. While parental warmth provedto be significantly related to children's competence, the relationship suddenlydisappeared when the support went beyond a particular level, which suggests thatadditional increases in parental warmth do not contribute to further increases inchildren's competence. In a more recent longitudinal study with children aged 6to 40 months, Landry, Smith, Miller-Loncar and Swank (1998), found a comparablethreshold for the relation between maternal "warm sensitivity" and children's socialcompetence. Once again, a significant positive relation was found between maternalsensitivity and child competence up to a point beyond which higher levels ofmaternal sensitivity no longer fostered further increases in competence. Thus, inline with the studies of Roberts (1986) and Landry et al (1998), the results of ourstudy suggest that very low levels of parental support may lead to suboptimaldevelopment of young children's competence and that above a certain "goodenough" level, additional parental support may have little further impact on children'scompetence development

The results of our study nevertheless indicate that a very low level of parentalsupport alone cannot explain the severe lack of competence motivation observed

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a D

avis

] at

23:

50 2

1 N

ovem

ber

2014

LONGITUDINAL PATTERNS 13

in certain toddlers. Patterns 2 and 3 reflecting decreasing and stable low levels ofparental support, respectively, did not differ in the mean level of parental supportprovided across age or in the final level of support observed at 30 months. Yet onlythe children with the decreasing parental support pattern showed extremely lowlevels of competence motivation at 30 months. In fact, the children with the stablelow support pattern did not end up with significantly lower competence motivationscores than the children with the stable high and increasing support patterns. Thestable low and decreasing support patterns did not differ in the mean level ofsupport the parents provided their children but did differ significantly in thevariability and decrease of parental support experienced by the children. Thissuggests that the "good enough" environment for children's competence developmentcannot merely be defined in terms of the current or mean level of parental supportover time; possible changes in the quality of support over time must also be takeninto account. In particular, our findings suggest that the loss of good quality supportmay be even more predictive of low competence motivation in children thanexperiencing a generally low quality of support from the beginning.

Acknowledgement

We would like to thank Robert B. Cairns and Jan MAM. Janssens for their helpfulcomments on earlier drafts of this paper and Juchke van Roozendaal for collectingparts of the data.

References

Ainsworth, M.D.S., Blehar, M.C., Waters, E. and Wall, S. (1978). Patterns of attachment: A psychological studyof the Strange Situation. Hillsdale, NJ: Lawrence Erlbaum Associates.

Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review,84, 191-215.

Banta, T.J. (1970). Tests for the evaluation of early childhood education: the Cincinatti Autonomy TestBattery (CATB). In J. Hellmuth (Ed.), Cognitive Studies, 1. New York: Brunner/Mazel.

Barret, K.C., Morgan, G A and Maslin-Cole, C. (1993). Three studies on the development of masterymotivation in infancy and toddlerhood. In D.J. Messer (Ed.), Mastery motivation in early childhood (pp.83-109). London: Routledge.

Baumrind, D. (1993). The average expectable environment is not good enough: A response to Scarr.Child Development, 64, 1299-1317.

Belsky, J. (1984). The determinants of parenting: A process model. Child Development, 41, 292-311.Belsky, J., Crnic, K. and Woodworth, S. (1995). Personality and parenting: Exploring the mediational

role of transient mood and daily hassles. Journal of Personality, 63, 905-930.Burchinal, M., R. and Appelbaum, M.I. (1991). Estimating individual developmental functions: Methods

and their assumptions. Child Development, 62, 23-43.Busch-Rossnagel, N.A., Knauf-Jensen, D.E. and DesRosiers, F.S. (1995). Mothers and others: The role

of the socializing environment in the development of mastery motivation. In: R.H. MacTurk and G.A.Morgan (Eds.), Mastery motivation: Origins, conceptualizations, and applications. Advances in AppliedDevelopmental Psychology, 12(pp.ll7-147). Norwood NJ: Ablex

Erickson, M.F., Sroufe, L.A. and Egeland, B. (1985). The relationship between quality of attachmentand behavior problems in preschool in a high-risk sample. In I. Bretherton and E. Waters (Eds.),Growing points of attachment theory and research. Monographs of the Society for Research in ChildDevelopment, 50, 147-166.

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a D

avis

] at

23:

50 2

1 N

ovem

ber

2014

14 H.TH. MEIJ et al.

Ford, M.E. and Thompson, RA. (1985). Perceptions of personal agency and infant attachment: Towarda life-span perspective on competence development International Journal of Behavioral Development,8, 377-406.

Frodi, A., Bridges, L. and Groinick, W. (1985). Correlates of mastery-related behavior A short-termlongitudinal study of infants in their second year. Child Development, 56, 1291-1298.

Harter. S. (1978). Effective motivation reconsidered: Towards a development model. Human Development,21, 34-64.

Harter, S. (1981). A model of mastery motivation in children. In W.A. Collins (Ed.), Minnesota Symposiaon Child Psychology, 14, Hillsdale, NJ, Erlbaum.

Heckhausen, J. (1993). The development of mastery and its perception within caretaker-child dyads.In D.J. Messer (Ed.), Mastery motivation in early childhood (pp. 55-83). London: Routledge.

Kochanska, G., Clark, L.A., Goldman, M.S. (1997). Implications of mothers' personality for theirparenting and their young children's developmental outcomes. Journal of Personality, 65, 387-421.

Landry, S.H., Smith, K.E., Miller-Loncar, C.L. and Swank, P.R (1998). The relation of change in maternalinteractive styles to the developing social competence of full-term and preterm children. ChildDevelopment, 69, 105-123.

Lewis, M. and Goldberg, S. (1969). Perceptual-cognitive development in infancy: a generalized expectancymodel as a function of mother-infant interactions. Merrill-Palmer Quarterly, 15, 81-100.

Liitkenhaus, P. (1984). Pleasure derived from mastery in three-year-olds: its function for persistenceand the influence of maternal behavior. International Journal of Behavioral Development, 2, 343-358.

MacTurk, R.H., Morgan, G.A. and Jennings, K.D. (1995). The assessment of mastery motivation ininfants and young children. In: R.H. MacTurk and G.A. Morgan (Eds.), Mastery motivation: Origins,conceptualizations, and implications. Advances in Applied Developmental Psychology, 12, (pp. 19-57). NorwoodNJ: Ablex.

Maslin-Cole, C., Bretherton, I. and Morgan, G A (1993). Toddler mastery motivation and competence:links with attachment security, maternal scaffolding and family climate. In D.J. Messer (Ed.), Masterymotivation in early childhood (pp. 205-230). London: Routledge.

Meij, H. (1992) Sociale ondersteuning, gehechtheidskwaliteit en vroegkinderlijke competentie-ontwikkeling. [Parentalsupport, quality of attachment and development of competence in the first years]. Nijmegen: Universityof Nijmegen (doctoral dissertation).

Pianta, R, Sroufe, L A and Egeland, B. (1989). Continuity and discontinuity in maternal sensitivity at6, 24, 42 months in a high-risk sample. Child Development, 60, 481-487.

Purmann, E. (1988). Neugier und Kompetenzstreben: Entwicklung eines Elternfragebogens. [Curiosity andStriving for competence: development of a questionnaire for parents]. Unpublished doctoraldissertation, University of Bochum, Germany.

Ramey, C.T., Lee, M.W. and Burchinal, M.R (1989). Developmental plasticity and predictability:Consequences of altered ecologies. In M.H. Bomstein and N.A. Krasnegor (Eds.), Stability andcontinuity in mental development: Behavioral and biological perspectives (pp. 217-234). Hillsdale, NJ:Erlbaum.

Ramey, C.T., Starr, R, Pallas, J., Whitten, C. and Reed, V. (1975). Nutrition, response-contingentstimulation and the maternal deprivation syndrome: results of an early intervention program. Merrill-Palmer Quarterly, 21, 45-54.

Riksen-Walraven J.M.A. (1978). Effects of caregiver behavior on habituation rate and self-efficacy ininfants. International Journal of Behavioral Development, 1, 105-130.

Riksen-Watraven, J.M.A and van Aken, MAG. (1997). Effects of two mother-infant intervention programsupon children's development at 7, 10, and 12 years. In W. Koops, J.B. Hoeksma, and D.C. van denBoom (Eds.): Development of interaction and attachment: traditional and non-traditional approaches (pp.79-93). Amsterdam: North-Holland.

Roberts, W.L. (1986). Nonlinear models of development: An example from the socialization ofcompetence. Child Development, 57, 1166-1178.

Scarr, S. (1992). Developmental theories for the 1990s: Development and individual differences. ChildDevelopment, 63, 1-19.

Scarr, S. (1993). Biological and cultural diversity. The legacy of Darwin for development Child Development,64, 1333-1353.

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a D

avis

] at

23:

50 2

1 N

ovem

ber

2014

LONGITUDINAL PATTERNS 15

Simons, R.L., Lorenz, F.O., Wu, C. and Conger, R.D. (1993). Social network and marital support asmediators and moderators of the impact of stress and depression on parental behavior. DevelopmentalPsychology, 29, 568-381.

Skinner, E.A. (1986). The origins of young children's perceived control: mother contingent andsensitive behavior. International Journal of Behavioral Development, 9, 359-382.

Teti, D.M. and Gelfand D.M. (1991). Behavioral competence among mothers of infants in the first year:the mediational role of maternal self-efficacy. Child Development, 62, 918-929.

Van den Boom, D.C. (1994). The influence of temperament and mothering on attachment andexploration: An experimental manipulation of sensitive responsiveness among lower-class motherswith irritable infants. Child Development, 65, 1449-1469.

Van den Boom, D.C. and Kramers, H. (1991). Maternal responsiveness and infant exploration inirritable infants: A longitudinal perspective. In D.N.M. De Gruijter, A.J.M. Houtsma, M.W. Van derMolen and M.N. Verbaten (Eds.), The Fifth anniversary of Psychon. Den Haag: Labyrinth Publication.

Vaughn, B., Egeland, B., Waters, E. and Sroufe, L.A. (1979). Individual differences in infant-motherattachment at 12 and 18 months: Stability and change in families under stress. Child Development, 50,971-975.

Wachs, T.D. (1987). Specificity of environmental action as manifest in environmental correlates ofinfants' mastery motivation. Developmental Psychology, 23, 782-790.

Watson, J.S. (1966). The development and generalization of continency awareness in early infancy: somehypotheses. Merrill Palmer Quarterly, 12, 123-135.

Watson, J.S. (1972). Smiling, cooing and "The Game'. Merrill Palmer Quarterly, 18, 323-340.White, R.W. (1959). Motivation reconsidered: the concept of competence. Psychological Review, 66, 297-

333.

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a D

avis

] at

23:

50 2

1 N

ovem

ber

2014