Content uploaded by Nicole Rebelo
Author content
All content in this area was uploaded by Nicole Rebelo
Content may be subject to copyright.
1
Pupils’ Attitudes and Motivation toward learning and school – Study
of exploratory models on the effects of socio-demographics,
personal attributes and school characteristics
Adelinda Araújo Candeias1
(Centre of Research in Education and Psychology, University of Évora)
\
Nicole Rebelo2
(Centre of Research in Education and Psychology, University of Évora)
Manuela Oliveira3
(Centre of Research in Applied Math, University of Évora)
Patrícia Mendes4
(Centre of Research in Education and Psychology, University of Évora)
This study investigates the relationship between pupils’ gender, contextual
background, age and school failure and attitudes toward school, learning,
competence and motivation, while explicitly modelling the multilevel structure
of a large data set from an educational context.
The aim of the present study was to investigate exploratory models about
the effects of factors – context, socio-cultural environment, individual variables
– that may interfere with attitudes toward school, and pupils’ perceptions about
learning, competence and motivation.
Based on a study with Portuguese youngsters (N=778) from regular
classes, whom we applied the QATS – Questionnaire of Attitudes Toward
School (Candeias, 1996, 2009), we use regression trees algorithm to predict
student’ attitudes and motivation toward learning and school. Predictors
include socio-demographics, personal attributes and some specific
characteristics related to school.
The results reveal the importance of understanding attitudes toward school,
toward learning, toward competence and toward motivation as being affected
from different combinations of factors. Attitudes should be understood as
based on the natures of the subject. Thus the improvement of attitudes and
motivation toward school should involve pupils, teachers, parents and
community.
Key words: Attitudes toward school; Learning; Competence; Motivation;
School failure, Regression trees.
1 Professor, University of Évora, aac@uevora.pt
2 Master Degree Student, University of Évora
3 Professor, University of Évora; mmo@uevora.pt
4 Master Degree Student, University of Évora
2
Introduction
Nowadays, it is increasingly common to find, in our schools, unmotivated
pupils who experience repeated academic failure, what leads, in many cases, to
dropping out of school. What is required from teachers, psychologists and
schools principals are new ways to combat these phenomena and increase
student interest in school, and in addition improve their academic performance.
In this work, we try a new way of looking at the academic performance
and motivational levels of pupils, starting from the proposals of the ecological
model (Bronfenbrenner, 1989) and multidisciplinary approaches (Machado,
Matias, & Leal, 2005). Taking a look at the relationships between pupils’
academic performance and their motivation levels, we used a model of pupils’
interaction, as proposed by Ren and Arnold (2003). Here, cultural, economic,
social and family systems and attitudes, pupils’ perceptions of competence and
pupils’ motivation interact in different levels, modelling the way pupils
perceive themselves and the environment that surrounds them.
With this model, we may analyse the characteristics of the contexts and
sub-contexts and their interaction with the characteristics of pupils and their
different levels of responsiveness. According to this model, it is "a system of
nested, interdependent, dynamic structures ranging from the proximal,
consisting of immediate face-to-face settings, to the most distal, comprising
broader social contexts such as classes and cultures" (Bronfenbrenner, 1993 , p.
4). How pupils behave in school, how they relate to peers and teachers, the
investment they make in academic and extracurricular activities are aspects
largely influenced by the attitudes they have toward school. Moreover,
student’s attitudes toward school are deeply influenced by their cultural
background, the kind and quality of family relationships, family and pears
support, previous school performance, that means, pupils’ positive attitudes
and behaviours play an important role in their academic success (Akey, 2006).
The construct attitude toward school was first defined by Lewy (1986) as
being the subject's behaviours, their feelings expression regarding to affection
and judgments, favourable or unfavourable, for the school and school
experiences. The affective characteristics of it may be an important explanatory
element of quality-education and investment of individual actors in the
different dimensions that make up the school (Santiago, 1994). This construct
is intrinsically related to other constructs, such as pupils’ perceptions and
interest in learning, their competence (perceived and as a result of an academic
achievement) and motivation.
As it appends with other constructs and also with attitudes, there are
differences regarding to gender. Girls seem to have more positive attitudes,
while boys are less motivated and have more negative attitudes toward school
(Van Houtte, 2004). In general, results show that girls do not require more time
to study, engage less in cases of misconduct and behaviour, have less
absenteeism, and also have more expectations about future and are more
enthusiastic about further studies (Van Houtte, 2004). On the other hand, boys
3
are less committed to studies and give up more easily (Barber, 1996,
Warrington et al., 2000, cit. Van Houtte, 2004), once their representation of
popular teen does not require to have good grades and being committed to
school (Francis, 2000, Warrington et al., 2000, cit. Van Houtte, 2004). There is
also evidence that cognitive variables such as ability-related and expectancy
beliefs, general attitudes toward school, and attitudes toward specific academic
subjects are related to academic performance and these can differ across gender
and racial groups (e.g., Ekstrom, 1994; House, 1997; Wigfield & Eccles, 1999,
cit. in Linnehan, 2001).
When we talk about the relation between attitude toward school and socio-
economic level, the results of previous studies had shown that boys with higher
levels appear to be more satisfied with school. Pupils from lower
socioeconomic status who have less access to school resources and computers
express more negative attitudes toward school.
The socio-economic level also regards to the way in which families take
part in their children academic life. In this aspect, previous studies show that
family contexts which are less exciting and involved in their children's
education are manifested in less positive attitudes toward school, less resilience
levels (Abreu, Veiga, Ferreira & Antunes, 2006) and have higher probability of
dropping out of school, as soon as they feel less support from their family and
community (Rumberger, 2001), and tend to believe that having studies and
completing school are not important to have a job or having a career
Regarding to the relation between attitudes toward school and academic
achievement, we are able to say that previous school performance experienced
by pupils have influence in the attitudes they show toward school, learning and
commitment to school. The research developed until now reveal that pupils
with lower performance and higher rate of school failure have more negative
attitudes. But when schools are able to provide interesting activities for their
pupils and the way those activities are engaged, and even the participation of
pupils and their families in school decisions, it will have influence on how
pupils feel at school and how they react to school life. That is, schools that are
more engaging arouse more positive attitudes (Abreu, Veiga, Antunes &
Ferreira, 2006). Pupils who perceive more support from adults who live with
them at school and from colleagues have more positive attitudes and academic
values and feel more satisfaction with school (Akey, 2006). For example,
Linnehan (2001) found differences between racial groups and parental
educational background in pupils’ beliefs and attitudes toward grades. In the
same study, he found that parental educational level was associated with more
favourable attitudes toward college (except for the Asian group). Moreover,
this study stressed the relationship between the college-related attitudes and
beliefs with pupils’ performance, even among urban pupils. In short, the results
of existing studies on attitudes toward school are the most commonly found: a)
Contents of attitudes: learning, competence and motivation); b) Context
variables: the kind of environment: rural/urban, socio-economic level and
4
parents’ instruction level; and c) Personal and developmental variables (gender
and age). Previous Portuguese studies (Candeias & Rebelo, 2010) also reveal
significant effects of gender, contextual background, school failure and
development on attitude toward school in general, but not in the case of the
perception of competence, as if socio-cultural experience didn't interfere
So we propose, with the present study, to develop an exploratory model to
explain and understand attitudes toward school, learning, competence and
motivation with pupils from compulsory education during a set period of
development – adolescence.
The present study
The aim of the present study was to investigate exploratory models about
the effects of factors – context, socio-cultural environment, individual variables
– that may interfere with attitudes toward school, and pupils’ perceptions on
learning, competence and motivation.
Research question
There exists different models about the effects of factors as school
experience, socio-cultural context, developmental characteristics that may
interfere with different kinds of attitudes, namely, Attitude Toward School in
general, Attitude Toward School Learning, Self-perception of Competence and
Self-perceived Motivation. May then pupils’ attitude and perception toward
different subjects (school, learning, competence, motivation) be influenced by
different kind of factors?
METHOD
Participants
From the whole group of participants, 778 youngsters, 593 are from the
northern region and 185 from the southern region of Portugal. 594 are from
rural contexts and 184 from urban contexts. Participants are from regular
school (75 from 7th grade, 56 from 8th grade and 647 from 9th grade), and 282
experienced, at least once, a failed year (36%) and 496 (64%) have academic
success (never experienced a fail); 429 (55%) are girls and 349 (45%) are boys,
with the average age of 14,57 years (SD = 1,16).
Procedure
The test application took place in a single fifty-minute session on teaching
time and in the presence of the researcher. Answers to the questionnaires were
provided on a total voluntary basis, under permission from parents, teachers
and school principal.
Instrument
– Questionnaire of Attitudes toward School (QATS, Candeias, 2009):
Attitudes toward School were measured through QATS. The participants
are expected to analyse one item (a single and objective sentence in verbal
5
format) in three major dimensions: Attitude toward Learning, Self-perception
of Competence and Self-perceived Motivation. These are evaluated through a
questionnaire in which participants are requested to evaluate each sentence on
a 1-to-5 Likert scale. In this scale 1 is “I strongly disagree” and 5 is “I strongly
agree”. The test has a great construct validity, proved by confirmatory factorial
analyses, in which the multidimensionality of the questionnaire is shown
(Candeias, 2009), as its internal consistency of .942.
RESULTS AND DISCUSSION
In order to verify the previous questions, we proceed to the analysis of
factors – context, socio-cultural environment, individual variables – that may
interfere with attitudes toward school, and pupils’ perceptions about learning,
competence and motivation. To do so, we used decision tree analyses
(regression) which allow us to verify how independent variables interfere
significantly in participants’ punctuations in QATS, or attitudes toward school,
and pupils’ perceptions about learning, competence and motivation. Decision
trees are a way of analysing data to discover important relationships and
segments, allowing the analyst to identify the membership to certain groups
and formulate rules to make predictions for new cases. Each tree starts with a
simple node which contains all sample observations. As we progress in the
analysis and interpretation of the tree, the data shatter into mutually exclusive
subsets. This process is applied recursively to subsets, until the analysis is
completed. In this work we use only the CHAID (Chi-Squared Automatic
Interaction Detector, Breiman, et al., 1984, Kass 1980). The CHAID is a
statistical algorithm of tree growth to examine the relationships between a set
of independent variables and a dependent variable. It provides a summary
diagram, displaying categories and values of the independent variables that
impact more the dependent variable. For the purposes of segmentation, thus the
CHAID provides information on the combination of characteristics that have a
higher probability of a particular response from the dependent variable. This
method is widely used in multidimensional studies. It is very successful in
situations where the explanatory variables are a mixture of nominal, ordinal
and continuous variables. Furthermore, the model has other advantages. It
easily adapts to missing data; it is invariant to transformations of variables (e.g.
logarithmization), it does not need to comply with conditions of applicability of
the model, as in parametric models. CHAID shows categorical data into
mutually exclusive subsets that best distinguish between different categories of
the dependent variables (Using a chi-square test, p=0.005). First, the predictor
categories are merged into compound categories so that there are significant
differences among the composite categories but not within each one. After this,
each independent variable is analysed and the one that explains most of the
variation in the response variable is used to divide the data into subsets based
on the values of their merged categories (Lindahl & Winship, 1994).
6
We present the analyses of exploratory models using CHAID for Attitude
toward School, Attitude toward School Learning, Self-perception of
Competence and Self-perception of Motivation.
I. Attitude toward School
Analysis based on decision tree picks school as main predictor variable
(Figure 1). These results confirm as model of predictors of attitude toward
school. The variables that take greater relative normalized importance are on a
scale 0 to 100, respectively: school variables (school (30,6%) and school level
(28,1%)), development (age (19,4%)), and parents’ level of instruction (18%).
Other variables as school failure, socio-economic level, gender and
environment (residence) only explain 3,9% of variability (Table 1 and Table
2).
II. Attitude toward Learning
As in the case of attitude toward school, analysis of decision tree picks
school as main predictor variable (Figure 2) of attitude toward learning. Such
results confirm as model of predictors of attitude toward learning: school
variables (school (32,1%) and school level (29,7%)), development (age
(17,2%)) and parents’ level of instruction 17,7%. Other variables (namely,
socio-economic level, school failure, gender and environment (residence)),
only give a contribution of 3,3% of the explanation for the model (Table 3 and
Table 4).
III. Self-perception of Competence
Analysis based on decision tree picks school as main predictor variable
(Figure 3). These results confirm as model of predictors of attitude toward
school: school variables (school (35,5%), school failure (18%) and school level
(5,4%)), development (age (33%)), and gender (5,1%). Other variables as
parents’ level of instruction, socio-economic level and environment
(residence), explains 3 % of the model (Table 5 and Table 6).
IV. Self-perception of Motivation.
Analysis based on decision tree picks school as main predictor variable
(Figure 4). These results confirm as model of predictors of self perception of
motivation: school variables (school (42,6%), school level (20,2%),
development (age (17,3%)) and parents’ level of instruction (15%)).Other
variables as gender, socio-economic level, school failure and environment
(residence) explains 4,9 % of the model (Table 7 and Table 8).
As we could observe, data suggest similar models for attitudes toward
school and attitudes toward learning. Data suggest that attitudes toward school
and toward learning depend on the school, school level, age and parents’
instruction. This group of scholar variables, developmental variables and socio-
cultural variables explain 97% of attitudes variability toward school and
7
learning. Another important suggestion that merge from data is that attitudes
toward school and attitudes toward learning seem to depend from the same
kind of factors, suggesting the idea that school is represented essentially by its
function of learning.
Other important finding is that attitudes toward competence are explained
by school, experience of failure, age and gender and school level, at 97%. This
data seem to indicate that both, the type of school as the type of school
experience (success/failure) and the level of familiarity or experience at school,
seem to have a strong effect at self-perception of competence. It is curious to
note that psychological and developmental factors as age and gender contribute
for 10% of the explanation, which should be understood in future studies.
Finally, attitudes and self-perception of motivation seem to be explained at
95%, by school, school level, age and parents’ level of instruction. Such model
suggests that the local and contextual effect of school, as the level of
experience at school, explain 60% of the self-perception of motivation. Again,
development emerges as an important factor to explain attitudes. Finally
parents’ level of instruction comes out as an important factor to understand
attitudes and self-perception of motivation, as well as to explain attitudes
toward school and toward learning.
CONCLUSIONS
The results of this investigation come to reveal the importance to
understand attitudes toward school, toward learning, toward competence and
toward motivation as affected of different combinations of factors. School
effect seems to be the most important factor present in all the models, as the
level of experience at school. Another important and emergent factor is
contextual factors. This emerges at the models of attitude toward school,
attitude toward learning and self-perception of motivation and at the level of
parents’ instruction. Thus, it suggests that contextual and cultural experience
dictates different kind of attitudes toward school, learning and motivation.
Scholar, social and familiar experiences influence pupils’ construction of
meaning about school and learning and state the orientation of its motivation
and interest toward school. So, equal opportunities to all pupils access to
economic, social and cultural resources are crucial, because family and
community shall provide tools that improve the value of school and learning to
prepare future project of life, as suggested in previous studies, for example,
Abreu et al., (2006) and Kuperminc et al., (2008).
Academic failure determines pupils attitudes about themselves and their
competence, developing the feeling of being less competent than others
participants and also more negative expectancies toward future projects and
relationships with others, as other studies had already shown (Candeias, 1997).
The experience of development seems to represent a crucial factor present in
attitudes toward school and learning, as self-perceived motivation and
competence, as a consequence of its failure experience. Pupils’ growing up,
8
became then more critics about school and academic contents, as demonstrated
by previous studies (for example, as shown by Urdan & Schoenfelder, 2006).
In short, if we keep in mind that for the large majority of pupils success in
school depends on their attitudes, the support they perceive (from family,
school, community), their motivation to study and keep on going, even when
they find barriers or difficulties. So we must be alert to all the questions
analysed in the present study and the influence they have, not only in attitudes
toward school, but also the influence they exercise one to each other,
suggesting that improvement of attitudes and motivation toward school should
involve pupils, teachers, parents and community, as ecological and
multidimensional models have suggested (for example, Bronfenbrenner, 1989;
Candeias et al., 2010; Machado, Matias, & Leal, 2005).
We assume and propose that pupils’ attitudes should be understood as
based on the natures of the subject of attitude, because, as we could see in this
study, different models of factors explain different attitudes toward
school/learning, toward competence and toward motivation.
Finally, this study supports the compulsory need to focus on pupils'
attitudes. This should be imperative to consider, before taking decisions on the
school project, in order to improve the integrative construction and
development of curricular and extra-curricular activities that promote ways of
teaching and learning closer to student's interests and preferences. Thus
attitudes toward school and toward learning, attitudes toward learning and
motivational levels are important factors to understand the relationship
between student competences and academic environment, and between those
and student academic performance.
REFERENCES
Abreu, M., Veiga, F. H., Antunes, J., & Ferreira, A. (2006). Atitudes em
relação à escola e a si próprio em alunos de contextos familiares
diferenciados. In Tavares et al. Orgs. Activação do Desenvolvimento
Psicológico: Actas do Simpósio Internacional (pp. 194-
200). Aveiro:
Universidade de Aveiro.
Akey, T. M. (2006).
School Context, Student Attitudes and Behavior, and
Academic Achievement: An Exploratory Analysis. New York: MDRC.
Retrieved December 23, 2009,
from http://www.mdrc.org/publications/419/full.pdf
Antunes, J. & Veiga, F. H. (2004). Aspectos motivacionais dos alunos e do
ambiente da aula: Variáveis do contexto escolar. Comunicação
apresentada nas II Jornadas de Psicologia do Instituto Piaget. Almada:
Campus Universitário, Instituto Superior de Estudos Interculturais e
Transdisciplinares, 6 & 7 May 2004.
Breiman, L., Friedman, J. H., Olshen, R. A., and Stone, C. J. 1984.
Classification and Regression Trees. New York: Chapman & Hall.
Bronfenbrenner, U. (1989). Ecological systems theory. In, R. Vasta (Ed.), Six
9
theories of development (pp. 187-249). Grennwich, CT: JAI Press.
Bronfenbrenner, U. (1993). The ecology of cognitive development: Research
models and fugitive findings. In R. H. Wozniak & K. W. Fisher (Eds.),
Development in Context: Acting and thinking in specific environments
(pp. 3-44). Hillsdale, NJ: Erlbaum.
Candeias, A. A. (1997). Attitudes toward school - An exploratory study with
pupils from basic education. Évora: Publications “University of Évora”
(Collection of Social and Human Sciences, Vol. 7) (ISBN: 972 9313 78
4).
Candeias, A. A. (2007). Prova cognitiva de inteligência social
. Lisboa:
CEGOC.
Candeias, A.A. (2009). Estudo Psicométrico do Questionário de Atitudes Face
à Escola - QAFE - com uma amostra de alunos do 3º ciclo do Ensino
Básico. Revista Educação: Temas e Problemas, 7
, (submitted and
accepted).
Cand
eias, A. A. & Rebelo, N. (2010). Student’ Attitudes Toward School,
Learning, Competence and Motivation –
the effects of gender,
contextual background, school failure and development. Investigación
en Convivencia Escolar. (pp. 547-55). GEU Editorial (ISBN: 978-84-
9915-122-9).
Kass, G. V. 1980. An exploratory technique for investigating large quantities
of categorical data. Applied Statistics 29(2), 119-127.
Kuperminc, G. P.; Darnell, A. J. & Alvarez-
Jimenez, A. (2008). Parent
involvement in the academic ad
justment of Latino middle and high
school youth: Teacher expectations and school belonging as mediators.
Journal of Adolescence, 31, 469–483.
Lewy, A. (1986). School Attitudes: General, In
H. Husén & T. Postlethwaite
(Ed.), The International Encyclopedia of Education, 4408-4411. Oxford:
Pergamon Press.
Lindahl, E. W. & Winship, C. (1994). A longit model with interactions for
predicting major gift donors. Research in Higher Education, 35(6), 729-
743.
Linnehan, F. (2001). Examining racial and family educatio
nal background
differences in high school student beliefs and attitudes toward academic
performance. Social Psychology of Education, 5, 31–48.
Machado, L., Matias, A. R. & Leal, S. (2005). Desigualdades sociais e
diferenças culturais: Os resultados escolar
es dos filhos de imigrantes
africanos. Análise Social, 176, 695-714.
Renn, K. & Arnold, K. (2003). Reconceptualization research on college student
peer culture. The Journal of Higher Education, 74, 3,261-291.
Rumberger, R. W. (2001). Who Drops Out of School and Why. In
Understanding Dropouts: Statistics, Strategies, and High-
Stakes Testing.
A. Beatty, U. Neiser, W. Trent, and J. Heubert (eds.). Washington, D.C.:
National Academy Press.
10
Santiago, R. A. (1994). Representações da escola nos alunos. Comunicação
apresentada no VI Seminário -
A Componente de Psicologia na
Formação de Professores: Ser Aluno. Universidade de Évora.
Schunk, D. H. & Meece, J. L. (2005). Self-
efficacy development in
adolescences. In Self-Efficacy Beliefs of Adolescents
. Information Age
Publishing, 71–96.
Urdan, T. & Schoenfelder, E. (2006). Classroom effects on student motivation:
Goal structures, social relationships, and competence beliefs.
Journal of
School Psychology, 44, 331-349.
Van Houtte, M. (2004), Why boys achieve less at school than girls: the
difference between boys’ and girls’ academic culture. Educational
Studies, 30, 2, 159-173.
Veiga, F. H. (2005). Novos elementos acerca da escala de autoconceito “Piers-
Harris Children Self-concept Scale”. In CIEFCUL (Ed.), Itinerários:
Investigar em educação 2005
. Lisboa: Centro de Investigação da
Faculdade de Ciências da Universidade de Lisboa.
11
Figure 1 Analysis of Decision Trees (Attitude toward school) (N=778)
12
Table 1 Analysis of Gain Summary for Nodes (Attitude toward school)
(N=778)
Node
N
Percent
Mean
2
185 23,8% 64,6595
8
118 15,2% 49,2034
7
98 12,6% 46,3163
6
150 19,3% 45,6867
9
86 11,1% 43,9651
10
141 18,1% 42,3050
Growing Method: CRT
Table 2 Analysis of Independent Variable Importance (Attitude toward school)
(N=778)
Independent Variable
Importance
Normalized Importance
School/Escola
66,972 100,0%
School level/Ano de Escolaridade
46,488 69,4%
Age/Idade
27,639 41,3%
Parents’ level of Instruction/Hab. Lit.-Pais
14,700 21,9%
School Failure/Repetência
2,636 3,9%
Soc. Economic Lev./Niv. Soc.-Ec.
2,328 3,5%
Gender/Sexo
1,453 2,2%
Residence/Residência
,004 ,0%
Growing Method: CRT
13
Figure 2 Analysis of Decision Trees (Attitude toward learning) (N=778)
14
Table 3 Analysis of Gain Summary for Nodes (Attitude toward learning)
(N=778)
Node
N
Percent
Mean
2
185 23,8% 66,2324
8
118 15,2% 46,9746
7
98 12,6% 45,2959
6
150 19,3% 44,8200
9
86 11,1% 44,2907
10
141 18,1% 43,5745
Growing Method: CRT
Table 4 Analysis of Independent Variable Importance (Attitude toward
learning) (N=778)
Independent Variable
Importance
Normalized Importance
School/Escola
82,059 100,0%
School level/Ano de Escolaridade
55,719 67,9%
Age/Idade
31,360 38,2%
Parents’ level of Instruction/Hab. Lit.-Pais
17,248 21,0%
Soc. Economic lev./Niv. Soc.-Ec.
2,747 3,3%
School Failure/Repetência
,692 ,8%
Gender/Sexo
,304 ,4%
Residence/Residência
,056 ,1%
Growing Method: CRT
15
Figure 3 Analysis of Decision Trees (Self-perception of competence ) (N=778)
16
Table 5 Analysis of Gain Summary f or Nodes (Self-perception of competence)
(N=778)
Node
N
Percent
Mean
2
185 23,8% 60,2378
4
216 27,8% 51,0741
9
98 12,6% 48,0306
10
52 6,7% 44,7308
7
131 16,8% 44,2901
8
96 12,3% 40,4688
Growing Method: CRT
Table 6 Analysis of Independent Variable Importance (Attitude toward
learning) (N=778)
Independent Variable
Importance
Normalized Importance
School/Escola
32,732 100,0%
School level/Ano de Escolaridade
21,112 64,5%
Age/Idade
19,354 59,1%
Parents’ level of Instruction/Hab. Lit.-Pais
8,541 26,1%
School Failure/Repetência
7,980 24,4%
Soc. Economic Lev./Niv. Soc.-Ec.
2,091 6,4%
Gender/Sexo
2,060 6,3%
Residence/Residência
,401 1,2%
Growing Method: CRT
17
Figure 4 Analysis of Decision Trees (Self-perception of motivation ) (N=778)
18
Table 7 Analysis of Gain Summary for Nodes (Self-perception of motivation)
(N=778)
Node
N
Percent
Mean
4
268 34,4% 55,3694
8
55 7,1% 54,6000
9
114 14,7% 52,0877
10
156 20,1% 51,2692
5
131 16,8% 39,3588
6
54 6,9% 36,7593
Growing Method: CRT
Table 8 Analysis of Independent Variable Importance (Attitude toward
motivation) (N=778)
Independent Variable
Importance
Normalized Importance
School/Escola
40,718 100,0%
School level/Ano de Escolaridade
23,365 57,4%
Age/Idade
15,145 37,2%
Parents’ level of Instruction/Hab. Lit.-Pais
8,118 19,9%
Gender/Sexo
1,993 4,9%
Soc. Economic Lev./Niv. Soc.-Ec.
1,283 3,2%
School Failure/Repetência
,614 1,5%
Growing Method: CRT
