ArticlePDF Available

The Validity of Educational Disadvantage Policy Indicators

Authors:
  • Semi-retired

Abstract and Figures

Many countries have implemented policies to prevent or combat educational disadvantage associated with socioeconomic factors in the students' home environment. Under such policies, educational institutions generally receive extra support from the central or local government. The support is normally based on indicators available in the home environment of the children, mostly family-structural characteristics. In the Netherlands, the core of educational disadvantage policy is the so-called weighted student funding scheme, which awards schools with disadvantaged students additional financial resources. When this scheme was developed in 1984, three indicators of disadvantage were selected, namely: parental education, occupation, and ethnicity. Analyses conducted at the time established a predictive validity estimate of 0.50, amounting to 25 percent of explained variance. Nowadays, some thirty years later, the funding scheme is based on only one indicator, namely parental education. Analyses performed on data collected in 2014 show a validity estimate of 0.20, thus accounting for no more than four percent of variance. This dramatic decrease of the indicator's predictive validity shows that the empirical basis of the Dutch weighted student funding scheme has become highly problematic. It is suggested that instead of employing family characteristics as educational disadvantage indicators, the actual performance of students based on test achievement and teacher observations may offer a more valid alternative.
Content may be subject to copyright.
Educational Policy Analysis and Strategic Research, V 12,N 2, 2017
© 2017 INASED
93
The Validity of Educational Disadvantage Policy Indicators
Geert Driesseni
Radboud University, Nijmegen, The Netherlands
Abstract
Many countries have implemented policies to prevent or combat educational disadvantage associated with
socioeconomic factors in the students’ home environment. Under such policies, educational institutions
generally receive extra support from the central or local government. The support is normally based on
indicators available in the home environment of the children, mostly family-structural characteristics. In the
Netherlands, the core of educational disadvantage policy is the so-called weighted student funding scheme,
which awards schools with disadvantaged students additional financial resources. When this scheme was
developed in 1984, three indicators of disadvantage were selected, namely: parental education, occupation,
and ethnicity. Analyses conducted at the time established a predictive validity estimate of 0.50, amounting to
25 percent of explained variance. Nowadays, some thirty years later, the funding scheme is based on only
one indicator, namely parental education. Analyses performed on data collected in 2014 show a validity
estimate of 0.20, thus accounting for no more than four percent of variance. This dramatic decrease of the
indicator’s predictive validity shows that the empirical basis of the Dutch weighted student funding scheme
has become highly problematic. It is suggested that instead of employing family characteristics as
educational disadvantage indicators, the actual performance of students based on test achievement and
teacher observations may offer a more valid alternative.
Keywords: Educational Disadvantage Policy, Weighted Student Funding, Predictive Validity, The
Netherlands
---------------------------
i Geert Driessen, Dr., Educational Researcher, Radboud University, Nijmegen, The Netherlands
Correspondence: driessenresearch@gmail.com
Educational Policy Analysis and Strategic Research, V 12,N 2, 2017
© 2017 INASED
94
Introduction
A recent review of education systems complimented the Netherlands for its excellence, which is
evidenced by a strong average performance, but at the same time cautioned her for a widening achievement
gap between students from disadvantaged socioeconomic backgrounds and more privileged students (OECD,
2016). The latter is not a unique development and can be witnessed in several Western countries (Davis-
Kean & Jager, 2014; Goodman & Burton, 2012; Goodman, Sands & Coley, 2015; Machin & McNally,
2012). This occurs despite the fact that in most countries under the umbrella of educational disadvantage
policy targeted school financing schemes and stimulation programs have been implemented specifically
designed to address these inequalities (Ballas et al., 2012; OECD, 2012; Ross, 2009; Stevens & Dworkin,
2014). Such compensatory policy instruments aim at achieving equality by unequal treatment according to
the principle of giving more to those who have less (Demeuse, Frandji, Greger & Rochex, 2012). For the
allocation of the support, a wide range of indicators are used to identify the policy’s target groups, that is the
disadvantaged students. Most indicators concern family structural characteristics, such as parents’ education,
parents’ occupation, ethnicity/race, home language, family structure, family income, and free school-meal
eligibility. Research into the reliability and, especially, the predictive validity of these proxy measures for
disadvantage is scarce, or typically not up to date. Insofar results are available, caution is warranted as to the
appropriateness of the indicators (Colpin et al., 2006; Gorard, 2012; Kounali, Robinson, Goldstein & Lauder,
2008; Ladd & Fiske, 2009).
In the present study, the focus is on the Dutch educational disadvantage policy, and specifically on
the most important instrument of this policy, the weighted student funding scheme which is used to allocate
additional financial resources to schools with disadvantaged students. In the next section, this policy and
funding scheme will be further explained. Then, the results of empirical analyses into the validity of the
educational disadvantage indicators will be presented, and some conclusions will be drawn.
The Dutch Educational Disadvantage Policy
Educational disadvantage policy in the Netherlands has been in effect since the 1970s. It aims at
preventing and combating educational disadvantage caused by social, economic and cultural factors in the
home environment of children. Its origin lies in the meritocratic ideal that educational opportunities should
be solely determined by innate abilities and that environmental factors should play no role (Meijnen, 2003).
To compensate for deficiencies, or lack of cultural capital, for children living in lower socioeconomic
milieus (Bourdieu & Passeron, 1977; Huang & Liang, 2016), a policy was initiated to award schools with
disadvantaged students extra financial support. Especially politicians from the political left (i.e. the Labour
Party) supported this policy.
Educational Policy Analysis and Strategic Research, V 12,N 2, 2017
© 2017 INASED
95
Initially, the policy focused on children of native-Dutch parents from lower socio-economic
environments. However, in the 1960s the number of children from non-Western immigrants in Dutch
education institutions started to increase dramatically. Three categories can be discerned: (1) immigrants
from former Dutch colonies (Surinam and the Netherlands Antilles); (2) so-called guest workers from
Mediterranean countries (especially Turkey and Morocco) and subsequent waves of immigration from these
countries for purposes of family reunification and family formation; and, more recently (3) asylum seekers
from Eastern Europe, Africa and the Middle East. In 2016 the Netherlands had nearly 17 million inhabitants,
and the largest non-Western immigrant groups had the following origins and numbers: Turkey (397,000),
Morocco (386,000), Surinam (349,000), and the Netherlands Antilles (151,000) (Statline, 2017). One
characteristic shared by most of these non-Western immigrants is their comparatively low level of education.
Because of their low socio-economic status and immigrant background (and inherent language and cultural
differences), the children of the non-Western immigrants soon became the main target groups of the
educational disadvantage policy.
Right from the start, there has been discussion regarding which indicators of disadvantage should be
used to award schools extra budgets. Two approaches can be distinguished, a groupwise versus an individual
approach. In the first case, support is given for all students who have one or more family structural
characteristics in common, regardless whether they actually have educational delays or not. It is assumed that
all students who meet these characteristics suffer from a comparable lack of stimulation in the home
environment, and therefore need to be compensated for these deficiencies at school. It is then crucial to select
indicators in the home that best predict educational opportunities. In the second case, support is given to
individual students who actually show educational delays. The relationship with the child’s social milieu in
this approach is indirect. Both approaches have advantages and disadvantages (Colpin et al., 2006; Jepma &
Beekhoven, 2013). In the groupwise approach, the predictive validity of the indicators is paramount. When
the validity is low, there is a high probability of false-positives and false-negatives, or on the one hand
students who wrongly receive support, or on the other hand students who wrongly do not receive support. An
advantage of the groupwise approach is that it facilitates preventive action at an early stage. Furthermore, it
is relatively easy and cheap to collect the information on the indicators. In contrast, a disadvantage of the
individual approach is that it can be costly, because children need to be tested individually. In addition,
action takes the form of remediation after it is established that a child has delays. Also, there is discussion
regarding the reliability of testing very young children. An advantage certainly for the older children is
that there will be fewer false-positives and false-negatives. After heated discussions in the Dutch Parliament
as to the pros and cons of both approaches, the groupwise approach eventually was chosen and a so-called
student weight funding scheme was developed to award schools additional financial resources for combating
educational disadvantage.
Educational Policy Analysis and Strategic Research, V 12,N 2, 2017
© 2017 INASED
96
The Student Weight Funding Scheme
The basis for the student weight funding scheme, which in essence is still functioning now, lies in
analyses performed in 1984 (Doesborgh, 1984). At the time, the predictive power of three indicators
professional level of father and educational level of father and mother of the children’s educational
attainment was estimated using a national large-scale dataset. Ethnic origin was also considered, but as there
were no comparable data available containing information on ethnicity, this indicator could not be included
in the analyses. The results showed that the educational level of the father was the best predictor with a
correlation of 0.42 and 17.6% (=0.422) of variance accounted for. Adding both other indicators resulted in
only a limited improvement of the prediction: educational level of mother 2.7% extra, and professional level
of father another 1.0% extra. It was decided to dichotomize the three indicators (low versus intermediate and
high educational and professional level). For educational level of father, this resulted in 11.9% of explained
variance, for professional level of father in 3.6% extra, and for educational level of mother in another 2.1%
extra; thus, a total of 17.6% explained variance and a multiple correlation R of 0.42.
In the course of the years, the student funding scheme has been reconsidered several times (Claassen
& Mulder, 2011; Fettelaar & Smeets, 2013). The most important changes implemented included, first,
dropping professional level as an indicator, and then also ethnicity (or more precisely: parental country of
birth). At present, there is only one indicator left, namely parental level of education: the more students with
low-educated and very low-educated parents a school caters for, the higher the extra budget the school
receives. Three categories are distinguished: a student weight of 1.2 for very low-educated parents, a weight
of 0.3 for low-educated parents, and a weight of 0.0 for parents with an intermediate or higher education; the
two previous categories are considered the disadvantaged students, the latter the non-disadvantaged students
(CFI, 2006).
In the 2013/14 school year, the total budget for the student weight funding scheme in the primary
education sector was €358 million. However, as the Early Childhood Care and Education policy is also based
on this scheme, the total sum amounted to €729 million (Algemene Rekenkamer, 2015). In that year, 89% of
the primary school student population had a weight of 0.0, 6% a weight of 0.30, and 5% a weight of 1.20
(StatLine, 2016). In the 2008/09 school year, the average budget for a 0.0 student was €4900, for a 0.3
student €6900, and for a 1.2 student €10800. Consequently, for a 1.2 student a school received more than
twice as much as for a 0.0 student (Kuhry & De Kam, 2012).
The student weight funding budget is awarded to the school boards as part of the lump sum they
receive from the Ministry of Education (De Vijlder, Verschoor, Rozema, Van Velden & Van Gansewinkel,
2012). Although the extra funding is based on individual characteristics, this budget is not earmarked, either
at the individual, or at the group level. School boards and schools are free to spend it. Therefore, it is the
question if the extra financial resources end up with the students for whom they were awarded. A previous
Educational Policy Analysis and Strategic Research, V 12,N 2, 2017
© 2017 INASED
97
study concluded that an important part of the schools indicated that they were not aware of the fact that they
received extra funding. Furthermore, only a small number of these schools deployed the money for specific
activities aimed at combating educational disadvantage. In practice, the extra means were considered as
regular budget which in most cases resulted in smaller classes (Mulder, 1996). More recent studies showed
that nothing much has changed in the intervening years (Claassen & Mulder, 2011; Ministerie van Financiën,
2017).
An important question is whether the employment of extra financial means, via the student weight
funding scheme, has resulted in achieving the central goal of the educational disadvantage policy, which is
reduction of the achievement gap caused by socioeconomic factors in the home environment of the students.
A limited number of studies have tried to answer this question. Because no random control group design was
employed, reservations were made with regard to causality. The general conclusion was that the policy has
not led to a permanent reduction of language and mathematics delays of disadvantaged students. Several
reasons for this were put forward: a continuous changing of goals, target groups and instruments; goals that
were ambiguous and contradictory; a policy characterized by input financing without output obligations; as a
consequence of deregulation and decentralization processes a limitless freedom for school boards and
schools as to how to spend their budgets; the lack of a theory on the origin of educational disadvantage and
evidence-based solutions (Algemene Rekenkamer, 2001; Karsten, 2006; Ladd & Fiske, 2010; Leuven,
Lindahl, Oosterbeek & Webbink, 2003; Mulder, 1996).
Several monitoring studies have been conducted focusing on the development of the various target
groups. A recent study concluded that large differences exist between disadvantaged and non-disadvantaged
students at the start of primary school. Ethnic minority target group students have a substantial language
delay. In the last year of primary school, this delay has diminished somewhat but is still substantial. The
relative position of the non-minority target group students (i.e. the native-Dutch low-SES students) regarding
their language skills has deteriorated. It was also concluded that there are often hardly any differences
between disadvantaged students with the 0.3 weight and students with the 1.2 weight, while at the same time,
ethnic minority students achieve substantially lower than non-minority students with the same weight
(Driessen & Merry, 2014; Herweijer, 2009).
Research Questions
Thirty years ago, the student weight funding scheme was developed. Since then, the circumstances
have changed and the funding scheme has been reorganized several times. The main question this study aims
at answering is if the scheme is still adequate. More specifically, the research questions are:
1. How strong are the correlations between family structural indicators of educational disadvantage on
the one hand, and language and mathematics achievement of young children on the other?
Educational Policy Analysis and Strategic Research, V 12,N 2, 2017
© 2017 INASED
98
2. How do these correlations relate to the correlations found thirty years ago, when the weight funding
scheme was developed?
To answer these questions, analyses were performed on recent large-scale data. In the next section,
the results of these analyses will be presented.
Method
The data for the present study come from the Dutch cohort study COOL5-18 collected in the 2013/14
school year (Driessen, Elshof, Mulder & Roeleveld, 2015). A total of 437 primary schools with 28529
students in grades 2, 5 and 8 (6-, 9- and 12-year-olds) participated in this national large-scale study. The total
sample consisted of a so-called reference sample of 340 schools, which is representative of all Dutch primary
schools, and a supplementary sample of 97 schools with many disadvantaged students. The latter sample was
added to obtain sufficient numbers of students from smaller categories of disadvantaged students.
Furthermore, disadvantaged students and, especially, minority disadvantaged students tend to be
concentrated in particular schools in large cities. The over-representation of these schools in this
supplementary sample thus provides a ‘typical’ picture of the minority disadvantaged student.
In this study, the focus is on grade 2. The students in this grade took a standardized language and
mathematics test developed by CITO (the Dutch National Institute for Test Development). The results of
both tests were expressed in so-called proficiency scores. For the sake of comparability, these scores were
transferred into z-scores, with a mean of 0 and a standard deviation of 1. Information on the student weight
came from the school administrations: 88% of the students had a weighting factor of 0.0 (i.e. with
intermediate or higher educated parents), 7% had a weighting factor of 0.3 (low educated parents) and 5%
with a weighting factor of 1.2 (very low educated parents). The students parents had completed a
questionnaire with both the mother and the father answering questions on their education, country of birth,
work, religion, and language. This written questionnaire was accompanied by an instruction in Dutch,
English, Turkish and Arabic. Nevertheless, not all of the parents returned the questionnaire. Especially the
response among immigrant parents was low, 38%, compared to 65% among native-Dutch parents. Also, the
test scores of children of parents who had not returned the questionnaire were lower than those of children
whose parents had completed the questionnaire (a difference of 0.30 standard deviation). This points to
selective response. To check whether this response possibly influences the results, analyses were performed
on the original representative sample, that is including the students whose parents had not returned the
questionnaire. This sample also includes information on the parents’ education and country of birth and the
student weight factor provided by the school administrations (but not, as in the parents’ questionnaire, on
work, religion, language choice and language proficiency). The correlations between these indicators and the
language and mathematics test scores were practically identical to the ones that will be presented hereafter.
Thus, it does not appear that the selective response has led to deviating correlations. In addition, because the
Educational Policy Analysis and Strategic Research, V 12,N 2, 2017
© 2017 INASED
99
aim of this study is not to present representative data but to explore relationships, this selective response is
less problematic here (Zetterberg, 1963).
The parent questionnaire includes the following information on fathers and mothers. Family
structure discerns one-parent and two-parent families. Regarding country of birth, two categories were
discerned: The Netherlands and other Western countries (hereafter taken together as ‘The Netherlands’),
versus non-Western countries. Education distinguishes the highest level attained, and the highest level
completed with a diploma. Paid work indicates having a paid job for at least 12 hours per week, or not.
Religion has two categories, namely religious, versus not religious. Language choice discerns Dutch versus a
foreign language. Language proficiency refers to the average score of the four modalities listening, speaking,
reading, and writing with categories (1) very low; (2) low; (3) intermediate; (4) high; (5) very high.
The original sample included 5257 grade 2 students whose parents completed the questionnaire. For
4871 students in this sample both language and mathematics test scores were available and they serve as the
final sample for the analyses. Table 1 presents an overview of the indicators selected for analyses, with a
short explanation of their meaning. For some of the indicators, combinations were constructed, for instance,
for country of birth the number of parents within a family who were born in the Netherlands, and for highest
education the average and the highest level of the mother and the father.
< insert Table 1 about here >
Results
In the left panel of Table 2, descriptive statistics are presented, in the right panel the bivariate
correlations between each of the indicators and the language and mathematics test scores. Two types of
correlations are discerned, namely Pearson r and eta, or the linear correlation and the total correlation, that is
the linear plus not-linear correlation. In the case of dichotomous indicators, the eta coefficient is the same as
the r coefficient and is therefore not included in the table. The difference between eta and r gives an
impression as to how much the correlation deviates from linearity.
< insert Table 2 about here >
The bivariate correlations between the indicators and test scores show that the correlations r vary in
strength from 0.02 (language choice child-friends) to 0.26 (education: average mother + father). According
to the rule of thumb provided by (Cohen, 1988), a correlation of 0.10 is weak, a correlation of 0.30 is
moderate, and a correlation of 0.50 is strong. A first conclusion therefore is that all correlations in this table
point to less than moderate associations. The correlation between the present indicator, the student weight,
and the language and mathematics scores is not stronger than 0.20. In general, the correlations for language
are somewhat stronger than those for mathematics, but the differences are very small indeed. As such, these
Educational Policy Analysis and Strategic Research, V 12,N 2, 2017
© 2017 INASED
100
marginal differences are rather unexpected, as language is something which is learned both at home and at
school, while mathematics typically is learned at school. Mathematics proficiency, therefore, is expected to
be less dependent on family characteristics. A second conclusion is that in almost all cases, the correlations
for the mother indicators are somewhat stronger than those for the father indicators. A third conclusion is
that the correlations for the multiple (or combined) indicators (e.g., education mother plus education father)
in general are hardly any stronger than those for single indicators (such as mothers’ education, or fathers’
education). Multiple indicators therefore appear to not result in added value. A fourth conclusion is that
when the linear correlations r are compared to the total correlations eta the differences are only minimal.
This means that there are hardly any deviations from linearity. Taken together, the findings from this table
show that the importance of all of these indicators as a basis for the funding of extra financial budgets for
combating educational disadvantage is very weak.
In the analyses reported thus far, a total of 34 indicators were included, all of them separately. To get
an impression of the correlations when several indicators are analysed simultaneously, regression analyses
were performed. Because many of these indicators within the same block (e.g. country of birth) are strongly
inter-correlated (e.g. country of birth of mother with country of birth of father), a selection within each block
was made for the mothers’ indicators. The reasons for this choice are that in general these indicators are
somewhat stronger correlated with the test scores than the fathers’ indicators; that the number of missing
values for mothers is considerably lower than that for fathers because in most one-parent families, there is a
mother but not a father; and that as a result of this criterion a consistent selection was obtained. Within the
block of education, a selection was made of the highest education level attained because the correlation of
this indicator with highest education completed with a diploma was very strong (0.87). In the selection
process of the final indicators, a lower boundary for the correlation with test scores of 0.20 was employed.
This resulted in the following indicators: country of birth, highest level of education attained, and language
proficiency. In Table 3, the results of the regression analyses are presented. [1]
< insert Table 3 about here >
When we take the indicator with the highest percentage of explained variance as a starting point, the
table shows that there are differences between the prediction of the language and of the mathematics test
scores. In the case of the language test scores, language proficiency of the parents appears to explain most of
the variance, namely 6.2%. Highest level of education attained adds 2.5%, and country of birth another 1.0%.
Taken together, these three indicators explain 9.6% of the differences in language test scores. In the case of
the mathematics test scores, highest level of education explains most of the variance, namely 5.6 %.
Language proficiency of the parents adds 1.7 %, and country of birth another 0.5%. A total of 7.8% of the
variation in mathematics test scores is thus being explained by the three indicators, which is less than for the
language test scores.
Educational Policy Analysis and Strategic Research, V 12,N 2, 2017
© 2017 INASED
101
In the analyses performed for the validation of the student weight scheme back in 1984, only
information on parents’ educational and professional level was used. This amounted to a multiple correlation
R of 0.42. Country of birth was not available at the time, but from later studies it appeared that this also was
a relevant predictor of educational disadvantage. To arrive at an indication of the size of the extra predictive
power of this indicator, a rough estimation was made with the help of the present data. For the language test
scores R for student weight plus paid work of father and paid work of mother was 0.22, which results in
4.9% of explained variance. Adding country of birth of father and country of birth of mother resulted in a R
of 0.27 and 7.4 percent of explained variance, which is about half more. For the mathematics test scores R
was 0.21 and 4.4% of explained variance, and after adding country of birth 0.24 and 5.8 % of explained
variance, which is about one third more. If we translate these results back to the situation of 1984, this means
that to the R of 0.42 and 17.6% of explained variance between one half and one third must be added for
country of birth, which thus results in a total R of about 0.50 and 25% of explained variance. As a point of
reference, the present student weight (based solely on parental educational level) has a r of 0.20 and 4% of
explained variance.
Discussion
The results of the analyses unequivocally show that the validity of the present indicator of
educational disadvantage is very limited. At the start of the Dutch educational disadvantage policy, some
thirty years ago, the multiple correlation of the three indicators was estimated at 0.50 with 25% of explained
variance; nowadays, with parental education as the only indicator left, this correlation is 0.20 with not more
than 4% of explained variance.
Two explanations of this decrease can be put forward. On the one hand, the decrease may be caused
by characteristics of the indicator(s) used, but on the other hand may also be caused by changes in society.
Regarding the latter, in a society with more equality, the children’s social and/or immigrant background may
have lower explanatory power on educational achievement than in societies with less equality. The question
is if this explanation holds for the Dutch situation, and elsewhere. In the introduction section of this article,
several studies were mentioned that proved the opposite to be true (Davis-Kean & Jager, 2014; Goodman &
Burton, 2012; Goodman, Sands & Coley, 2015; Machin & McNally, 2012; OECD, 2016). Recently, the
Dutch Inspectorate of Education in her annual report also warned that the educational gap between children
from different social backgrounds is increasing, and this not only holds for primary, but as a consequence
also for secondary education and higher education. In addition, in both primary and secondary education
achievement of immigrant children is significantly lower than that of native-Dutch children (Inspectie van
het Onderwijs, 2016). This conclusion is in line with findings from the large-scale longitudinal study by
Driessen and Merry (2014) who showed that although immigrant children have improved their educational
position in the last decades, they still lag substantially behind their native-Dutch peers. Especially, the
Educational Policy Analysis and Strategic Research, V 12,N 2, 2017
© 2017 INASED
102
position of Turkish and Moroccan children is worrisome, even more as many of them are second or third
generation. It is obvious that this group still needs extra attention. The question is how.
A far-reaching implication of the dramatic decrease in validity of the indicator is that the hundreds of
millions of Euros yearly awarded to schools is based on quicksand. As a consequence, this inevitably leads to
unacceptable numbers of false-positives and probably even worse false-negatives, or students for whom
the schools unwarranted are awarded extra budget, respectively students for whom the schools unwarranted
are not awarded extra budget. An additional problem is that many schools indicate that they are not aware of
the fact that they receive extra budgets for combating educational disadvantage because this is part of the
lump sum they receive from the Ministry of Education. In practice, the extra budgets therefore are often
spent on creating smaller classes, as a result of which in principle all students, both disadvantaged and
advantaged, may benefit from the extra budgets. This not only leads to dilution effects (the extra budget is
spent on more students than intended), but also so-called Matthew effects may occur. The latter means that
the better students, mostly the non-disadvantaged students, profit more from the extra budget than the
disadvantaged students, which will result in an even wider achievement gap; Stanovich,1986). To this should
be added that there is no evidence that creating smaller classes sec is an effective strategy for combating
educational disadvantage (Vignoles, Levacic, Walker, Machin & Reynolds, 2000).
To summarize, the analyses show that serious doubt is warranted as to the empirical foundation of
the most important instrument of the Dutch educational disadvantage policy, viz. the weighted student
funding scheme. The question then arises whether the present groupwise approach of educational
disadvantage is still justified and whether one should not look for alternatives. Until recently, the individual
approach was held off, mainly because this was assumed to be very expensive and in the case of very young
children would lead to unreliable results. However, linguists argue that nowadays, a range of adequate
language tests for young children are available (Colpin et al., 2006; Onderwijsraad, 2002; Verhoeven &
Vermeer, 2005). And most Dutch institutions for Early Childhood Education and Care targeting children
between 2 and 6 years (playgroups and kindergartens) already work with comprehensive child monitoring
schemes that often combine standardized tests and observations by staff and teachers. A recent Dutch study
shows that subjective teacher assessment adds significantly explanatory power to cognitive test scores in
predicting student ability (Feron, Schils & Ter Weel, 2015). Another option is a two-stage approach: a first
screening by teachers followed by a more elaborate testing, or a first selection on the basis of structural
family characteristics followed by (repeated) individual testing (Onderwijsraad, 2001).
Note
1. In addition to the monolevel regression analyses presented here, multilevel regression analyses were
also performed. The results were identical. Language achievement: intercept -1.46; language proficiency
Educational Policy Analysis and Strategic Research, V 12,N 2, 2017
© 2017 INASED
103
0.17; highest level of education 0.11; country of birth 0.31 (p<0.001). Mathematics achievement: intercept -
1.24; language proficiency 0.12; highest level of education 0.12; country of birth 0.25 (p<0.001).
References
Algemene Rekenkamer (2001). Bestrijding onderwijsachterstanden. Den Haag: Algemene Rekenkamer.
Algemene Rekenkamer (2015). Resultaten verantwoordingsonderzoek 2014 Ministerie van Onderwijs, Cultuur en
Wetenschap. Den Haag: Algemene Rekenkamer.
Ballas, D., Lupton, R., Kavroudakis, D., Hennig, B., Yiagopoulou, V., Dale, R., & Dorlinget, D. (2012). Mind the
gap. Education inequality across EU regions. Brussels: European Commission.
Bourdieu, P., & Passeron, J.-C. (1977). Reproduction in education, society, culture. Beverly Hills, CA: Sage.
CFI (2006). Nieuwe gewichtenregeling in het basisonderwijs. Zoetermeer: CFI.
Claassen, A., & Mulder, L. (2011). Een afgewogen weging? De effecten van de gewijzigde gewichtenregeling in
het basisonderwijs. Nijmegen: ITS.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Erlbaum.
Colpin, M., Gysen, S., Jaspaert, K., Heymans, R., Van den Branden, K., & Verhelst, M. (2006). Studie naar de
wenselijkheid en haalbaarheid van de invoering van centrale taaltoetsen in Vlaanderen in functie van
gelijke onderwijskansen. Leuven: Katholieke Universiteit.
Davis-Kean, P., & Jager, J. (2014). Trajectories of achievement within race/ethnicity: ‘Catching up’ in
achievement across time. The Journal of Educational Research, 107(3), 197-208.
Demeuse, M., Frandji, D., Greger, D., & Rochex, J.-Y. (Eds.) (2012). Education policies and inequalities in
Europe. New York, NY: Palgrave Macmillan.
De Vijlder, F., Verschoor, M., Rozema, M., Van Velden, J., & Gansewinkel, H. van (2012). Interne
middelenverdeling in het primair onderwijs. Arnhem/Nijmegen: HAN Press.
Doesborgh, J. (1984). De secundaire analyse. Nijmegen: ITS.
Driessen, G., & Merry, M. (2014). Trends in educational disadvantage in Dutch primary school. Educational
Review, 66(3), 276-292
Driessen, G., Elshof, D., Mulder, L., & Roeleveld, J. (2015). Cohortonderzoek COOL5-18. Technisch rapport
basisonderwijs, derde meting 2013/14. Nijmegen: ITS/ Amsterdam: Kohnstamm Instituut.
Feron, E., Schils, T., & Ter Weel, B. (2015). Does the teacher beat the test? The additional value of teacher
assessment in predicting student ability. Den Haag: CPB.
Fettelaar, D., & Smeets, E. (2013). Mogelijke indicatoren van schoolgewichten. Onderzoek naar de voorspellende
waarde. Nijmegen: ITS.
Goodman, R., & Burton, D. (2012). What is the nature of the achievement gap, why does it persist and are
government goals sufficient to create social justice in the education system? Education 3-13:
International Journal of Primary, Elementary and Early Years Education, 40(5), 500-514.
Educational Policy Analysis and Strategic Research, V 12,N 2, 2017
© 2017 INASED
104
Goodman, M., Sands, A., & Coley, R. (2015). America’s skills challenge: Millennials and the future. Princeton,
NJ: ETS.
Gorard, S. (2012). Who is eligible for free school meals? Characterising free school meals as a measure of
disadvantage in England. British Educational Research Journal, 38(6), 1003-1017.
Herweijer, L. (2009). Making up the gap. Migrant education in the Netherlands. Den Haag: SCP.
Inspectie van het Onderwijs (2016). De staat van het onderwijs. Onderwijsverslag 2014/2015. Utrecht: Inspectie
van het Onderwijs.
Jepma, IJ., & Beekhoven, S. (2013). Naar een nieuw bekostigingsarrangement voor het
onderwijsachterstandenbeleid. Utrecht: Sardes.
Huang, H., & Liang, G. (2016). Parental cultural capital and student school performance in mathematics and
science across nations. The Journal of Educational Research, 109(3), 286-295.
Karsten, S. (2006). Policies for disadvantaged children under scrutiny: The Dutch policy compared with policies
in France, England, Flanders and the USA. Comparative Education, 42(2), 261-281.
Kounali, D., Robinson, T., Goldstein, H., & Lauder, H. (2008). The probity of free school meals as a proxy
measure for disadvantage. Bristol: Bristol University.
Kuhry, B., & Kam, F. de (red.) (2012). Waar voor ons belastinggeld? Prijs en kwaliteit van publieke diensten. Den
Haag: Sociaal en Cultureel Planbureau.
Ladd, H., & Fiske, E. (2009). Weighted student funding for primary schools: An analyses of the Dutch
experience. Durham. NC: Duke University.
Leuven, E., Lindahl, M., Oosterbeek, H., & Webbink, D. (2003). The effect of extra funding for disadvantaged
students on achievement. Amsterdam: Universiteit van Amsterdam.
Machin, S., & McNally, S. (2012). The evaluation of English education policies. National Institute Economic
Review, 219(1), 15-25.
Meijnen, W. (2003). Onderwijsachterstanden: Een historische schets. In W. Meijnen (Ed.).
Onderwijsachterstanden in basisscholen (pp. 9-29). Antwerpen/Apeldoorn: Garant.
Ministerie van Financiën (2017). Onderwijsachterstandenbeleid, een duwtje in de rug? Interdepartementaal
beleidsonderzoek naar het onderwijsachterstandenbeleid. Den Haag: Ministerie van Financiën.
Mulder, L. (1996). Meer voorrang, minder achterstand? Het Onderwijsvoorrangsbeleid getoetst. Nijmegen: ITS.
OECD (2012). Equity and quality in education. Supporting disadvantaged students and schools. Paris: OECD.
OECD (2016). Netherlands 2016: Foundations for the future. Paris: OECD.
Onderwijsraad (2001). Wat ’t zwaarst weegt. Een nieuwe aanpak voor het Onderwijsachterstandenbeleid. Advies.
Den Haag: Onderwijsraad.
Onderwijsraad (2002). Over leerlinggewichten en schoolgewichten. Advies. Den Haag: Onderwijsraad.
Ross, A. (2009). Educational policies that address social inequality. Brussels: European Commission.
Stanovich, K. (1986). Matthew effects in reading: Some consequences of individual differences in the acquisition
of literacy. Reading Research Quarterly, 21(4), 360-407.
Educational Policy Analysis and Strategic Research, V 12,N 2, 2017
© 2017 INASED
105
StatLine (2016). Leerlingen in (speciaal) basisonderwijs; achterstand, herkomst, woonregio. Retrieved from the
Web June 17, 2016. http://statline.cbs.nl/Statweb/publication/?DM=SLNL&PA=83295NED&D1=1-
5&D2=a&D3=0-2&D4=0&D5=2&HDR=G4
StatLine (2017). Bevolking; generatie, geslacht, leeftijd en herkomstgroepering, 1 januari 2015. Retrieved from
the Web July 6, 2017. http://statline.cbs.nl/Statweb/publication/?DM=SLNL&PA=37296ned&D1=0-
51&D2=0,10,20,30,40,50,(l-1)-l&VW=T
Stevens, P., & Dworkin, A. (Eds.) (2014). The Palgrave handbook of race and ethnic inequalities in education.
New York, NY: Palgrave Macmillan.
Verhoeven, L., & Vermeer, A, (2005). Het ongelijk van Netelenbos? Toetsing van kleuters en hun prestaties op de
Cito Eindtoets Basisonderwijs. Toegepaste Taalwetenschap in Artikelen, 74, 123-133.
Vignoles, A., Levacic, R., Walker, J., Machin, S., & Reynolds, D. (2000). The relationship between resource
allocation and pupil attainment: A review. London: Centre for the Economics of Education.
Zetterberg, H. (1963). On theory and verification in sociology. Totowa, NJ: The Bedminster Press.
Educational Policy Analysis and Strategic Research, V 12,N 2, 2017
© 2017 INASED
106
Table 1. Overview of the Indicators
Indicator
Information on
Values
Family structure
mother + father
1=one parent; 2=two parent
Country of birth (grand)parents NL
mother
0=non-Wes tern; 1=NL*
father
mother’s mother
mother’s father
father’s mother
father’s father
mother + father: number
0=0 NL; 1=1 NL (mixed); 2=2 NL
Highest level attended
mother
1=primary school ... 6=university
father
mother + father: average
mother + father: highest
Highest diploma
mother
1=no diploma ... 7=university
father
mother + father: average
mother + father: highest
Paid work
mother
0=no work; 1=work
father
mother + father: number
0=0 work; 1=1 work; 2=2 work
Religion
mother
0=no religion; 1=religious
father
mother + father: number
0=0 religious; 1=1 religious; 2=2
religious
Language choice NL
child with mother
0=no NL; 1=NL
child with father
child with siblings
child with friends
mother with father
family: number
0=in no area ... 5=in 5 areas
Language proficiency NL
mother
1=very low; 2 =low; 3=intermediate;
4=high; 5=very high
father
mother + father: average
mother + father: highest
Student weight factor
mother + father
1=0; 2=0.30; 3=1.20
*NL = The Netherlands
Educational Policy Analysis and Strategic Research, V 12,N 2, 2017
© 2017 INASED
107
Table 2. Indicators and correlations with language and mathematics achievement: means, standard
deviations, numbers of respondents, correlations r and eta
Indicator
Information on
%/M
SD
N
Language
r
eta
r
eta
Two-parent family
m + f*
92%
4843
0,06
0,04
Country of birth
mother
86%
4745
0.23
0.18
father
86%
4519
0.21
0.19
mother’s mother
82%
4773
0.24
0.20
mother’s father
83%
4749
0.24
0.20
father’s mother
83%
4671
0.22
0.20
father’s father
82%
4655
0.22
0.20
m + f: number
1.72
0.65
4812
0.24
0.24
0.21
0.21
Highest education
mother
3.94
1.35
4804
0.24
0.25
0.24
0.25
father
3.88
1.38
4515
0.23
0.24
0.22
0.24
m + f: average
3.89
1.23
4832
0.26
0.28
0.26
0.27
m + f: highest
4.25
1.31
4832
0.24
0.25
0.24
0.25
Highest diploma
mother
4.68
1.66
4686
0.23
0.24
0.22
0.23
father
4.55
1.75
4479
0.21
0.22
0.20
0.22
m + f: average
4.59
1.53
4826
0.25
0.26
0.24
0.26
m + f: highest
5.05
1.55
4826
0.24
0.24
0.23
0.24
Paid work
mother
70%
4772
0.12
0.12
father
92%
4491
0.11
0.10
m + f: number
1.59
0.63
4834
0.15
0.16
0.14
0.15
Religion
mother
58%
4788
-0.12
-0.10
father
55%
4492
-0.12
-0.12
m + f: number
1.13
0.94
4816
-0.13
0.14
-0.12
0.13
Language choice
child with mother
90%
0.31
4716
0.14
0.12
child with father
90%
0.31
4452
0.13
0.09
child with siblings
94%
0.24
4324
0.07
0.05
child with friends
97%
0.18
4635
0.02
0.02
m with f
81%
0.39
4549
0.19
0.14
family: number
90%
4817
0.15
0.20
0.11
0.15
Language proficiency
mother
4.62
0.67
4780
0.25
0.26
0.21
0.22
father
4.63
0.64
4509
0.22
0.23
0.18
0.19
m + f: average
4.62
0.61
4799
0.26
0.28
0.22
0.24
m + f: highest
4.72
0.54
4799
0.23
0.24
0.20
0.21
Student weight factor
m + f
1.17
0.49
4747
-0.21
0.21
-0.20
0.20
*m = mother; f = father
All correlations p < 0.001, except for Language: Language choice child-friends p = 0.185, and for Mathematics: Family
structure p = 0.002; Language choice child-siblings p = 0.001; Language choice child-friends p = 0.195.
Educational Policy Analysis and Strategic Research, V 12,N 2, 2017
© 2017 INASED
108
Table 3. Results regression analyses language and mathematics achievement and selected mother indicators:
unstandardized coefficients (B), standardized coefficients (Beta) and percentages (additionally) explained
variance (% R2)
Language
Mathematics
Full model
Stepwise
model
Full model
Stepwise
model
B
Beta
% R2
B
Beta
% R2
Constant
-1.65
Constant
-1.42
Language proficiency
0.19
0.13
6.2
Highest education
0.14
0.18
5.6
Highest education
0.12
0.16
+2.5
Language proficiency
0.15
0.10
+1.7
Country of birth
0.34
0.12
+1.0
Country of birth
0.25
0.09
+0.5
Total
9.6
Total
7.8
All effects p < 0.001.
Educational Policy Analysis and Strategic Research, V 12,N 2, 2017
© 2017 INASED
109
Miscellany
Scope of the EPASAD
Journal of Educational Policy Analysis and Strategic Research (EPASAD) is a peer reviewed interactive
electronic journal sponsored by the International Association of Educators and in part by the Graduate School of
Library and Information Science at the University of Illinois at Urbana-Champaign. EPASAD is a core partner of
the Community Informatics Initiative and a major user/developer of the Community Inquiry Laboratories.
EPASAD takes an interdisciplinary approach to its general aim of promoting an open and continuing dialogue
about the current educational issues and future conceptions of educational theory and practice in an international
context.
Editorial/Review Process
All submissions will be reviewed initially by the editors for appropriateness to EPASR. If the editor considers the
manuscript to be appropriate, it will then be sent for anonymous review. Final decision will be made by the editors
based on the reviewers’ recommendations. All process -submission, review, and revision- is carried out by e-
Journal Management System. The submissions should be written using MS-WORD or compatible word
processors. EPASR has a policy of screening for plagiarism.
Manuscript Submission Guidelines
All manuscripts should be prepared in accordance with the form and style as outlined in the American
Psychological Association Publication Manual (6th ed.). Manuscripts should be double-spaced, including
references, notes, abstracts, quotations, and tables. The title page should include, for each author, name,
institutional affiliation, mailing address, telephone number, e-mail address and a brief biographical statement. The
title page should be followed by an abstract of 100 to 150 words. Tables and references should follow APA style
and be double-spaced.Manuscript must be typed in Times New Roman font size 12-point, with 1" margins all
round.Normally, manuscripts should not exceed 30 pages (double-spaced), including tables, figures, and
references. Manuscripts should not be simultaneously submitted to another journal, nor should they have been
published elsewhere in considerably similar form or with considerably similar content.
EPASAD Co-Sponsors & Membership Information
International Association of Educators is open to all educators including undergraduate and graduate students at a
college of education who have an interest in communicating with other educators from different countries and
Educational Policy Analysis and Strategic Research, V 12,N 2, 2017
© 2017 INASED
110
nationalities. All candidates of membership must submit a membership application form to the executive
committee. E-mail address for requesting a membership form and submission is: members@inased.org
*There are two kinds of members - voting members and nonvoting members. Only the members who pay their
dues before the election call are called Voting Members and can vote in all elections and meetings and be
candidate for Executive Committee in the elections. Other members are called Nonvoting Members.
*Dues will be determined and assessed at the first week of April of each year by the Executive Committee.
*Only members of the association can use the University of Illinois Community Inquiry Lab. In order to log into
the forum page, each member needs to get an user ID and password from the association. If you are a member,
and if you do not have an user ID and password, please send an e-mail to the secretary: secretary@inased.org .
For membership information, contact:
Hill Hall 205 D
901 South National Avenue
Springfield, MO 65897 / USA
E-mail: info@inased.org
Electronic Access to EPASAD
All issues of the Journal of Educational Policy Analysis and Strategic Research may be accessed on the
World Wide Web at: http://epasr.penpublishing.net (Note: this URL is case sensitive).
... Inicialment, aquest sistema estava pensat per compensar les desigualtats educatives de l'alumnat, i s'orientava principalment a l'alumnat holandès de famílies de classe treballadora. En paral·lel a les dinàmiques demogràfiques i migratòries, la fórmula no va trigar a incloure l'alumnat d'origen immigrant a causa dels desavantatges associats a factors socioeconòmics i culturals, com ara el domini de la llengua (Driessen, 2017). ...
... Finalment, també cal destacar que, malgrat la llarga trajectòria d'aquesta política als Països Baixos, les diferències entre els resultats de l'alumnat de diferents contextos socials no ens permeten ser optimistes (Karsten, 2006;Driessen, 2017), malgrat que tampoc no es puguin establir relacions causals sobre els efectes d'aquesta política per una manca d'un «contrafactual» en el mateix context. ...
Technical Report
Full-text available
Proposta d'un sistema de finançament per fórmula dels centres educatius orientat a l’equitat.
... Etnik ve kültürel açıdan farklı (azınlık, göçmen, geçici koruma statüsü vs.) öğrencilerin diğer öğrencilere göre eğitsel faaliyetlerde dezavantajlı olduklarını gösteren birçok araştırma mevcuttur (bkz. Driessen, 2017;Alba, Sloan ve Sperling, 2011;Heath, Rothon ve Kilpi, 2008). Günümüzde etnik köken, dil ve kültürel açıdan farklılıkları bulunan çocuklar ABD ve Avrupa başta olmak üzere çoğu yerde eğitimöğretim faaliyetlerinden yararlanmakta (Arzubiaga, Nogueron ve Sullivan, 2009;Bravo Moreno, 2009), eğitim-öğretim faaliyetlerininin yapısını da etkilemektedirler (Bravo Moreno, 2009). ...
... There are many studies showing that students of ethnic and cultural differences (minority, immigrant, temporary protection status, etc.) are disadvantaged in terms of benefiting from educational activities compared to other students (Heath, Rothon ve Kilpi, 2008;Alba, Sloan ve Sperling, 2011;Driessen, 2017). Nowdays, children with ethnic, linguistic and cultural differences benefit from educational activities in many places, especially in the USA and Europe (Arzubiaga, Nogueron, & Sullivan, 2009;Bravo Moreno, 2009), and they also affect the structure of educational activities (Bravo Moreno, 2009). ...
... The amount of extra budget is based on socioeconomic and ethnic features of the children's parents. To a large extent the schools are free to spend the money; most schools use it to create smaller classes, but there also are many schools that invest in extra parental involvement activities, especially in the lower grades (Cebeon, 2016;Driessen, 2017). According to an evaluation committee appointed by the Ministry of Education, the interventions too often focus on the children and not enough on the children's home environment and parents. ...
... Minor parts of this article have been published in G. Driessen (2017). Early childhood education intervention programs in the Netherlands. ...
Article
Full-text available
Early Childhood Education (ECE) provides compensatory educational programs both in preschools and the early grades of primary school, and for parents at home. The aim of this policy is to prevent young children from disadvantaged backgrounds starting formal schooling with significant educational delays. In many countries ECE programs are in existence for several decades now. The search in this article is for the scientific evidence-base of this policy. While the focus is on the Netherlands, the findings probably also are valid for many other countries.
... Originally targeted at student teachers with an immigrant background, the network quickly expanded to all student teachers and current educators. As of 2016, it is a registered university group that meets regularly to share intercultural experiences and organise workshops (Public Policy and Management Institute (PPMI), 2017 [170]; Universität Paderborn, 2016 [171]). Specific workshops, seminars, and activities can be found on the group's Facebook page (www.facebook.com/LebenistVielfalt/); ...
... Furthermore, the network positively influenced teacher education at the University of Paderborn to include more content on diversity training. The network is currently working to create partnerships with other universities as well as add more members from diverse subject fields as most of the members are either German or social studies teachers (Universität Paderborn, 2016 [171]). ...
... Schools receive additional funding for students at risk of educational disadvantage (OECD, 2021). Changes in the funding formula used in the Netherlands appear to address at least some of the criticisms levelled at the earlier weighted student funding scheme, which by its conclusion was based only on parental education (Driessen, 2017). ...
Article
Full-text available
Reducing the socio-economic achievement gap is a key goal of education policy internationally. Since 2017, the stated intention in Ireland is to use the Pobal HP Deprivation Index to identify schools serving high proportions of disadvantaged students. In spite of this, little published research has compared the performance of the HP index to richer survey measures from large-scale educational assessments or examined its association with educational achievement at primary or second level. This paper aims to assess the validity and fitness for purpose of the HP index for use in identifying schools serving high concentrations of socio-economically disadvantaged students. Analyses draw on Ireland’s Program for International Student Assessment (PISA) 2018 dataset matched with administrative data. Findings show a strong correlation between school-average HP and other socio-economic measures examined. However, about one-in-five schools are identified as potential ‘false positives’ or ‘false negatives’ when school-average HP is used for identification purposes. Also, school-average HP explains less variance in reading achievement than other variables considered. Conclusions recognize the benefits of the HP index but emphasize the need for ongoing examination of the most appropriate methods of identification.
... Several countries have developed a system where schools with many disadvantaged students receive extra funding (Vandevoort et al., 2020;Vignoles et al., 2000). To a certain degree the schools are free to spend these supplementary budgets, for instance on class size reduction, special (language training) programs, and individual help for students (Driessen, 2017). Lately, there is a (renewed) interest in the preschool and early school phase. ...
Chapter
Full-text available
The achievement gap of disadvantaged students has always been large, and is still widening. Even more now, during the Covid-19 pandemic. Parental involvement is seen as an important strategy for closing this gap. The ultimate objective is to expand the academic and social capacities of students, especially those of disadvantaged backgrounds determined by ethnic minority/immigrant origin and low socioeconomic status. This article focuses on possible roles of parents in education and aims at answering two questions: (1) What types of parental involvement can be discerned? and (2) What are the effects of parental activities on their children's attainment? To answer both questions, a review of the literature was conducted, and a synthesis of the results from twelve meta-analyses was performed. The review pointed to a considerable diversity in parental involvement typologies, classifi cations, roles, forms, and activities. Nevertheless, they can be ordered along the lines of just a few perspectives, namely locus (at home/at school), style (formal/ informal), action (active/passive), and actor (parent/student/school). From the synthesis of the meta-analyses it can be concluded that the average effect of involvement on attainment is small. In addition to many positive effects there are also substantial numbers of null and even negative effects. The type of involvement with the strongest effect appeared to be parents having high aspirations and expectations for their child. No differences in effects of involvement on attainment according to ethnic/immigrant and social background could be established. Prudence is called for, however, as there are many limitations to studying parental involvement in a reliable and valid way.
Article
Full-text available
At present around 865.000 Muslims live in the Netherlands. In 1988 the first Islamic primary school was founded; now there are 61 with 15,000 students. Islamic education always has been a highly controversial topic in the Netherlands. The debate centers around the question whether the schools contribute to the integration of Muslim youth into Dutch society, or leads to isolation and segregation. This article’s goal is to entangle why and how the schools were established, the obstacles met in this process and the resulting heated societal debate, and the schools’ attainments in terms of cognitive and noncognitive student achievement. To arrive at these insights a review and analysis of the literature was conducted. The results show that Islamic schools academically achieve relatively well, that is, taking into account their largely socioeconomically disadvantaged student population. Also, they perform best on a pen-and-paper integration test. This does not mean, however, that especially populist and right-wing politicians are convinced now that all Muslim youngsters will accept the Dutch norms and values and will integrate into Dutch society.
Article
Full-text available
The pupil weighting system is the core financial element of the Dutch Educational Disadvantage Policy. This policy departs from the idea that children who grow up in an environment where specific “cultural capital” is lacking, face a grave risk of developing severe educational arrears right from their start in school. To compensate for this “deficit”, primary schools and other educational and welfare institutions receive additional budgets from the Ministry of Education they can use for providing extra help to the policy’s target groups. The amount of extra budget is based on socio-economic and ethnic factors in the children’s home situation. Since its implementation in 1974 the system has been changed several times. These changes are described here. For the 2019/20 school year a new system has been developed. Some points of criticism are presented; the most important is that there is no evidence that the system which has cost some 20 billion euros has been effective.
Article
Full-text available
De nieuwe gewichtenregeling die de kern vormt van het Onderwijsachterstandenbeleid en ontworpen is door het CBS bevat drie dubieuze indicatoren. Een verbetering zal zij daarom zeker niet zijn.
Article
Full-text available
Het onderwijsachterstandenbeleid gaat een nieuwe fase in. Het ministerie van OCW heeft het CBS gevraagd een nieuw model te ontwikkelen voor de toekenning van extra budget aan de scholen ter bestrijding van onderwijsachterstanden. Dit ter vervanging van de gewichtenregeling, waarbij scholen extra geld ontvangen voor laagopgeleide ouders. In deze blog wordt verschillende argumenten aangevoerd waarom die nieuwe regeling geen verbetering van de (al zwakke) oude zal zijn. Deze tekst verscheen eerder op DidactiefOnline. Geert Driessen (2018). Zo doorgaan met het onderwijsachterstandenbeleid? Blog Didactief Online, 4 januari 2018, https://didactiefonline.nl/blog/blonz/zo-doorgaan-met-het-onderwijsachterstandenbeleid
Book
Full-text available
De aanvullende bekostiging van scholen in verband met het onderwijsachterstandenbeleid is momenteel gebaseerd op informatie over de opleiding van de ouders van de leerlingen. Deze informatie is niet altijd correct. De overheid studeert daarom op de mogelijkheid om de basisgegevens voor deze bekostiging voortaan zonder tussenkomst van scholen en ouders te verzamelen. Daarbij zou gebruik moeten worden gemaakt van centraal beschikbare gegevensbestanden. Het doel van dit onderzoek is nagaan hoe goed indicatoren die in centrale bestanden beschikbaar zijn leerresultaten voorspellen. De eerste onderzoeksvraag betreft de voorspellende waarde van (combinaties van) alternatieve indicatoren van onderwijsachterstand voor de leerresultaten van leerlingen in vergelijking met het huidige leerlinggewicht. De tweede vraag heeft betrekking op eventuele verschillen tussen groep 2, 5 en 8 in voorspellende waarde van (combinaties van) indicatoren. De derde vraag betreft het al dan niet lineair zijn van het verband tussen huishoudinkomen en leerresultaten van leerlingen en de omvang van de verschillen in resultaten tussen inkomensgroepen.
Article
Full-text available
The authors' purpose was to examine the relationship between three forms of cultural capital—the embodied, the objectified, and the institutionalized—and student performance in mathematics and science. Their analysis of Trends in International Mathematics and Science Study 2011 data from 32 countries and regions revealed that parental expectation (the embodied cultural capital) had the strongest association with student performance. Parental education (the institutionalized cultural capital) and book possession (the objectified cultural capital) were also significant, with some exceptions. However, their relationship with student performance was relatively weaker. This suggested that the significant relationship between cultural capital and student performance holds in the 32 countries included in the sample, about 4 decades after Bourdieu first introduced the concept (i.e., students with higher parental cultural capital are more likely to perform better in mathematics and science).
Article
A framework for conceptualizing the development of individual differences in reading ability is presented that synthesizes a great deal of the research literature. The framework places special emphasis on the effects of reading on cognitive development and on “bootstrapping” relationships involving reading. Of key importance are the concepts of reciprocal relationships—situations where the causal connection between reading ability and the efficiency of a cognitive process is bidirectional-and organism-environment correlation—the fact that differentially advantaged organisms are exposed to nonrandom distributions of environmental quality. Hypotheses are advanced to explain how these mechanisms operate to create rich-get-richer and poor-get-poorer patterns of reading achievement. The framework is used to explicate some persisting problems in the literature on reading disability and to conceptualize remediation efforts in reading.
Article
This research investigates to what extent the subjective teacher’s assessment of children’s ability predicts children’s outcomes in the transition from primary to secondary school in terms of initial track allocation, track switching in the first three years of secondary education and subsequent test scores. We apply micro-data from the Netherlands about cognitive test scores and teacher’s assessment in primary schools and about track placement, track switching and test scores in secondary schools. Our estimates suggest that the subjective teacher’s assessment is about twice as important as the primary school cognitive test scores for initial track placement in secondary school. In addition, the teacher’s assessment is more predictive of track allocation in 9th grade compared to cognitive test scores. Next, children who switch tracks are more likely to be placed in tracks based on test scores. Also, test scores in 9th grade are predicted by the subjective teacher’s assessment, not by test scores in 6th grade.
Article
The achievement gap has long been the focus of educational research, policy, and intervention. The authors took a new approach to examining the achievement gap by examining achievement trajectories within each racial group. To identify these trajectories they used the Early Childhood Longitudinal Study–Kindergarten Cohort, which is a nationally representative sample of students in kindergarten through Grade 5. In the analyses, the authors found heterogeneity within each racial group in mathematics and reading achievement, suggesting that there are in fact achievement gaps within each race/ethnicity group. The authors also found that there are groups that catch up to the highest achieving groups by Grade 5, suggesting a positive impact of schooling on particular subgroups of children. The authors discuss the various trajectories that have been found in each racial group and the implications this has for future research on the achievement gap.