Transition from high school to university: a person-centered approach to academic achievement

Abstract

Although a vast body of studies regarding the variables related to students’ achievement exists, only a handful has investigated how these variables combine and interact together. Such an investigation might make it possible to more accurately illustrate the heterogeneity of students enrolling in university and assess the impact of this diversity on academic achievement. The current study focused on the person-centered approach and investigated the possibility of determining significant subgroups of freshmen in the very first week of the academic year in the Belgian educational context. Using k-mean clustering, 2178 freshmen were classified according to their past performance, socioeconomic status, study choice process, and academic self-efficacy beliefs at the beginning of the first year at university. Analyses of variance were also conducted to analyze the relationship between these student clusters and academic achievement outcomes. Six distinct profiles of students were identified representing different combinations of achievement predictors. Results revealed different odds of success among these profiles. The implications of our approach to the understanding of the heterogeneity of freshmen and the first-year achievement process are discussed.

Full text

Transition from high school to university:
a person-centered approach to academic achievement
Mikaël De Clercq
1
&Benoît Galand
2
&Mariane Frenay
2
Received: 3 August 2015 /Revised: 13 November 2015 /Accepted: 10 March 2016 /
Published online: 23 March 2016
#Instituto Superior de Psicologia Aplicada, Lisboa, Portugal and Springer Science+Business Media Dordrecht
2016
Abstract Although a vast body of studies regarding the variables related to students’achieve-
ment exists, only a handful has investigated how these variables combine and interact together.
Such an investigation might make it possible to more accurately illustrate the heterogeneity of
students enrolling in university and assess the impact of this diversity on academic achieve-
ment. The current study focused on the person-centered approach and investigated the
possibility of determining significant subgroups of freshmen in the very first week of the
academic year in the Belgian educational context. Using k-mean clustering, 2178 freshmen
were classified according to their past performance, socioeconomic status, study choice
process, and academic self-efficacy beliefs at the beginning of the first year at
university. Analyses of variance were also conducted to analyze the relationship
between these student clusters and academic achievement outcomes. Six distinct
profiles of students were identified representing different combinations of achievement
predictors. Results revealed different odds of success among these profiles. The implications of
our approach to the understanding of the heterogeneity of freshmen and the first-year achieve-
ment process are discussed.
Keywords Highereducation.A cademic achie vement .Collegestudent.Entrancepatterns.Self-
efficacy
Eur J Psychol Educ (2017) 32:39–59
DOI 10.1007/s10212-016-0298-5
*Mikaël De Clercq
mikael.declercq@uclouvain.be; http://www.uclouvain.be/mikael.declercq
Benoît Galand
benoit.galand@uclouvain.be; http://www.uclouvain.be/benoit.galand
Mariane Frenay
mariane.frenay@uclouvain.be; http://www.uclouvain.be/mariane.frenay
1
Research Institute of Psychological Sciences, Faculty of Psychology and Education, Université
Catholique de Louvain, 10 Place Cardinal Mercier, 1348 Louvain-La-Neuve, Belgium
2
Department of Psychology, Université Catholique de Louvain, 10 Place Cardinal Mercier,
1348 Louvain-La-Neuve, Belgium

Introduction
The freshman year is a challenging transition accompanied by major changes in students’
educational environment such as unfamiliar academic tasks, new social networks, and height-
ened academic competition (Perry et al. 2001). In Belgium, 35 % of freshmen fail the freshman
year and 25 % leave higher education (Droesbeke et al. 2008). The same pattern of results is
reported across Europe, disclosing the global issue of success rates among first-year students at
university (OECD 2013). Considering educational policies’imperatives of expanding the
diversity and number of students enrolling and succeeding at university, academic achieve-
ment can be conceived as an on-going current concern (Gale and Parker 2012). An in-depth
understanding of achievement is thus needed to cope with the increased numbers of freshmen
from diverse backgrounds entering universities.
Over the years, many approaches to student achievement during the first year at university
have been proposed in the literature (Pekrun et al. 2002; Robbins et al. 2006). To date, a large
number of diverse variables, e.g., student’s socioeconomic status (Arulampalam et al. 2004),
attendance (Dollinger et al. 2008), study skills (Fenollar et al. 2007), self-regulated learning
(Minnaert and Janssen 1999), self-efficacy beliefs (Phan 2009), and social support (Fass and
Tubman 2002), have been depicted as having a direct impact on achievement. These findings
lend credence to a conception of academic achievement as a complex multifactorial process.
However, most of the studies have investigated the impact of these factors independently
without taking their interrelationships into account. Recent surveys have shed light on the
inadequacy of such single-factor analyses for understanding the complex process of academic
achievement (Allen et al. 2010; Busato et al. 2000; De Clercq et al. 2013;EcclesandWigfield
2002;Pintrich2003). These surveys have pointed out a gap in the current literature which
consists in testing the impact of a variable on academic achievement without considering that
this impact resides within students’global functioning. According to these authors, a student’s
achievement is contingent on the way that individual characteristics combine. It can therefore
be assumed that the impact of a variable on achievement will vary depending on how it
combines with a student’s other characteristics. This study therefore aimed at screening the
possibility of developing an inclusive approach to freshman achievement that might deal with
the interplay between some important achievement factors.
In such cases, a person-centered approach such as cluster analysis can provide a more
detailed examination than a variable-centered approach (Dumais 2005; Phinney et al. 2005;
Vansteenkiste et al. 2009; Hayenga and Corpus (2010,p.372);“a person-centered approach
focuses on particular combinations of variables as they exist within individuals rather than
taking each variable itself as the focal point.”The basic assumption of this framework is that
human behavior is the result of dynamic interplay between several variables and can be
accurately understood by investigating these complex patterns of factors (Bergman et al.
2003;Magnusson1999). This framework also postulates that it is possible to identify distinct
typical subgroups with specific patterns in the population (Roeser and Peck 2003).
Several authors have justified such assumptions in the context of higher education (Fenollar
et al. 2007;Heikkiläetal.2011, emphasizing the heterogeneity of students enrolling in the first
year at university and underlining the challenge for educational literature to reflect this
diversity in a more comprehensive way. Numerous higher education theories have also
postulated that students bring to university a set of important characteristics that could
constitute the shaping of forces and weaknesses to adapt to the new academic context (for a
review, see Pascarella and Terenzini 2005). For example, Tinto’stheory of student departure
40 M. De Clercq et al.

(Tinto 1982,1997) emphasized the impact of pre-entry attributes (e.g., parents’education,
student’saspirations…) on students’persistence process. Astin’s theory of involvement (Astin
1999) highlighted that inputs (family background, academic and social pre-college experi-
ence…) could intervene in the development of a student’s involvement in the college expe-
rience. Pascarella’s model of change hypothesized that a student’s background characteristics
and pre-college traits could have a direct impact on his/her environment perceptions and effort
(Pascarella and Terenzini 2005). Weidman’s model of undergraduate socialization postulated
that students’socialization could differ depending on entrance socioeconomic status, past
performance, aspiration, and motivation (Padgett et al. 2010). Finally, Price’s4Pmodelof
learning shed light on the impact of “presage”variables (initial motivation, ability…)onhow
students perceive learning contexts and initiate approaches to learning (Price 2013).
Based on the work of these authors and their theories, our study focused on cluster
analysis and investigated the possibility of determining significant subgroups of
freshmen in the very first week of the academic year in the Belgian tertiary educa-
tional system. This procedure was based on the assumption that each student enters
university with a specific and complex profile which entails specific adaptation to the
academic world. This perspective has been overlooked in the current literature where
distinctions between entrance characteristics and process variables are often blurred
(Pascarella and Terenzini 2005). The theories mentioned above have insisted on the utility of
such a distinction for understanding the differences between freshmen relative to their responses
to the same academic settings and experiences. Authors have also stressed the necessity of
promoting achievement as soon as possible, because the first months are crucial to adaptation to
the academic world (Neuville et al. 2013; Sauvé et al. 2007).
Following this line of thought, a more systematic conception of these profiles was endorsed
by focusing on four factors in particular (past performance, socioeconomic status, self-efficacy
beliefs, and study choice) documented as important achievement predictors and identified as
particularly relevant for an “open access”
1
system. The four selected factors also meet several
entrance categories endorsed by the above-mentioned theories. Recent studies have argued that
both background and psychosocial factors need to be collected in order to offer a more
comprehensive approach to students’entrance characteristics (Dollinger et al. 2008;Allen
et al. 2010; Richardson et al. 2012).
Past performance
International findings claim that past academic performances—represented by high school
grade point average or a standardized achievement test score—are the most powerful predic-
tors of achievement at university level (Díaz et al. 2001; Hackett et al. 1992; Perry et al. 2001;
Vandamme et al. 2005). Much energy has been devoted to examining past performance, and
research and theories attempting to accurately grasp the impact of this performance on
academic achievement have emerged in educational literature (Richardson et al. 2012).
Dollinger et al. (2008) concluded that past performance and abilities could explain about
37 % of the academic achievement of college students. Allen et al. (2008)foundthathigh
school grades typically explained about 28 % of freshmen’s grade point average. This factor is
1
The Belgian tertiary educational system is qualified as an “open access”system because there is no standard-
ized testing of students or admission criteria and the successful completion of comprehensive secondary school
allows any student to enter any university program without passing an entrance test (except in engineering).
A person-centered approach to academic achievement 41

of utmost importance in open access systems which have no standardized testing of students
(De Clercq et al. 2013).
Several studies have concluded that the impact of past academic performance on freshmen’s
academic achievement reflects the academic preparation of students for the academic world
(Allen et al. 2008; Robbins et al. 2006). Past achievement allows easier adaptation to the new
complex environment (Díaz et al. 2001) and promotes the establishment of a stable belief in
one’s ability to succeed (Elias and MacDonald 2007). Other authors have highlighted that past
performance is an indicator of students’cognitive abilities (Willingham et al. 2002). Past
performance is therefore considered by each aforementioned theory as an important entrance
characteristic. It is, for instance, included in pre-entry attributes of Tinto’s theory of departure
and in pre-college traits of Pascarella’s model of change.
Socioeconomic status
Socioeconomic status (SES)—reflected by the social, cultural, and economic resources avail-
able to students (e.g., parents’education and income)—has also been widely studied. Many
studies found that students from lower socioeconomic backgrounds had lower achievement
(Battle and Pastrana 2007; Caldas and Bankston 1997; Robbins et al. 2004). A meta-analysis
by Sirin (2005a, b) reviewed the literature on academic achievement from 1990 to 2000 and
highlighted a moderate-to-strong relationship between SES and academic achievement.
Three main explanations for this have emerged from the literature. According to
Willingham et al. (2002, p.18), “the social advantages implied by higher SES…would
presumably act over a lifetime on the development of general cognitive skill in and out of
school.”Other authors have claimed that the impact of SES on academic achievement could
reflect a school effect (Sackett et al. 2009;ZwickandGreen2007). Freshmen from under-
privileged backgrounds have to cope with limited school resources and have no access to an
adequate education. They are therefore ill-prepared to undertake academic studies. Finally,
sociological theories have argued that the impact of SES on academic achievement may
indicate that educational institutions reproduce the social order through sorting mechanisms.
These authors suggest that underprivileged students have few opportunities to build social and
cultural capital and are therefore ill prepared to adapt to the academic world (Linnehan et al.
2011;Mills2008). Other authors have highlighted that students from underprivileged back-
grounds often have lower confidence in their abilities to succeed (Phinney et al. 2005)and
underachieve in high school (Zwick and Green 2007). Like past performance, SES is an
integral part of background variables depicted in the above theories. It can be related to family
background in Astin’s model of involvement and to parents’education in Tinto’stheoryof
departure.
Self-efficacy beliefs
Another line of research on psychosocial factors argues that academic self-efficacy beliefs are
of the utmost importance for student achievement in higher education. This concept was
originally developed by Bandura (1997) in the social cognitive theory. Several authors claim
that confidence in one’s abilities and chances of success is strongly correlated to performance
and fosters many other variables such as mastery goal orientation, intrinsic motivation, self-
regulated learning, effort ratings, emotional competences, social integration, intention to
persist, and deep-processing study strategies (Adeyemo 2007; Bong and Skaalvik 2003;
42 M. De Clercq et al.

Fenollar et al. 2007; Torres and Solberg 2001). Elias and Macdonald (2007) highlighted that
academic self-efficacy accounted for about 22 % of variance in college achievement and went
beyond the variance accounted for by past performance. These results were corroborated by
Chemers et al. (2001) who undertook a study on first-year university students and found both a
direct and indirect powerful predictive effect of academic self-efficacy beliefs on academic
achievement, above and beyond the impact of past academic performance.
According to several authors, self-efficacy is paramount because it favors a positive
motivational, cognitive, affective and behavioral process that good adaptation to academic
requirements entails (Brown et al. 2008; Elias and MacDonald 2007; Kennett et al. 2009;
Pintrich 2003; Robbins et al. 2004;Zimmerman2000). However, as mentioned by Zimmerman
(2000), self-efficacy beliefs are not an omnibus trait but rather a multidimensional construct
differing on the basis of the domain of functioning. Therefore, global self-efficacy beliefs may
be less predictive in explaining differences in academic achievement than domain-specific self-
efficacy beliefs. From a theoretical point of view, academic self-efficacy beliefs are not
traditionally included as entrance characteristics even though they could be related to pre-
college experience in Astin’s theory of involvement. However, considering the major impact of
this variable on the achievement process, the variable was embedded in the analyses.
Study choice
Finally, in certain countries such as Belgium or the Netherlands where access to university is
much more open, the study choice process seems to play a key role in achievement. In such a
context, this process is often initiated much later than in more restrictive systems. It is therefore
essential to analyze the impact of the study choice process on achievement among first-year
students at university. Informed choice can be defined as the complexity of a student’sstudy
choice process (Biémar et al. 2003). Several research studies and theories such as the future
time perspective theory (Husman and Lens 1999) and educational choice implementation
(Germeijs and Verschueren 2007) have highlighted the importance of the information process.
Findings led to the conclusion that students who make an informed and thoughtful study
choice attain higher academic achievement, are more satisfied with their courses, apply more
adaptive study strategies, and are more confident in their ability to succeed (Biémar et al. 2003;
Husman and Lens 1999;Lensetal.2002). However, this field of research remains underde-
veloped in the higher education context and needs further investigation to support these initial
findings. The notion of study choice is quite close to the aspiration highlighted in the models
developed by Tinto, Pascarella, and Weidman. These models justified the consideration of
aspiration as a significant entrance characteristic.
Research questions addressed
The studies reviewed above have highlighted important predictors of academic success.
However, although they have revealed some insights into the relationship between these
predictors, none of them has investigated their complex combinations within students and
the way these relate to academic achievement. As mentioned above, the present study was
aimed at developing a person-oriented conception of entrance factors that affect students’
achievement during their first year of university in the Belgian educational system.
Consequently, the outline of our study was determined by considering the limitations of extant
literature and addressing the following two research questions:
A person-centered approach to academic achievement 43

RQ1. Can meaningful subgroups of freshmen with specific combinations of entrance
variables be identified?
RQ2. Do these subgroups differ regarding achievement?
We might hypothesize that several complex profiles could emerge from the broad panel of
students entering the university, each profile related to specific odds of success. Several
hypotheses could be drawn on the manner in which factors might combine in pattern-
centered analyses.
First, according to the empirical studies highlighting that SES, past performance, self-
efficacy beliefs, and informed choice are connected (Husman and Lens 1999; Phinney et al.
2005; Zwick and Green 2007), we could adopt a global perspective and postulate that these
variables will evolve together. As a result, one factor’s score will have an impact on the scores
of the other factors. Three subgroups might therefore be expected, respectively characterized
by low, average, and high scores on the four factors. A subgroup with low scores on all the
factors is expected to be the least successful. Conversely, a subgroup with high scores is
assumed to be the most successful.
Second, we could move toward a specific point of view. Four subgroups of students
respectively characterized by a specific weakness in one of the four factors under investigation
are expected. These four subgroups could allow us to consider the impact of each entrance
weaknesses when it is offset by other forces. According to the core impact of these factors in
extant literature (Allen et al. 2008; Elias and Macdonald 2007), it is expected that a subgroup
with particularly low past-performance or academic self-efficacy beliefs will demonstrate
significantly lower performance than the two other subgroups. We postulated that these two
weaknesses would scarcely be compensated by other entrance forces.
Method
Sample and procedure
Two thousand one hundred seventy-eight first-year university students from 29 different study
programs (psychology, law, economics, medicine, physical education, chemistry, computer
science, engineering…) in a Belgian university participated in a survey on a voluntary basis
(48.5 % of the population of first-year students with a mean age of 18.3 years). Of the study
participants, 52.9 % were women. The questionnaire was administered during a lecture session
at the beginning of the year, and the student grade point average was collected, with students’
consent, at the end of the academic year. One thousand nine hundred thirty-three students
consented to the collection of their grade point average. Independent ttest analyses were
performed to compare initial scores on entrance variables of students who gave their consent
with those who did not. No significant difference was found, which strengthens the assumption
that our sample is representative.
Measures
Socioeconomic status In order to develop an accurate measure of SES, multiple indicators
were collected and combined in an overall SES measure. Three indicators of SES were
gathered, namely parental education (mother’s and father’s highest educational level), home
44 M. De Clercq et al.

possessions (i.e., number of cars, computers, art objects, musical instruments, literature books,
poetry books, etc.), and the student’s sociocultural activities (i.e., traveling abroad, going to the
theater, visiting museums, skiing during the winter holidays, etc.; alpha= 0.67). Recent studies
have supported such a composite measure (Sackett et al. 2009;Sirin2005a, b). Exploratory
factor analysis revealed one single factor (eigenvalue higher than 1) explaining more than
40 % of the variance, with factor loadings ranging from 0.57 to 0.72. The factor score was
extracted to create an overall SES indicator.
High school grade Participants were asked to report their overall average percentage in their
final year of high school on a four-point scale (1, 60–70 %; 2, 70–80 %; 3, 80–90 %; 4, more
than 90 %).
Academic self-efficacy beliefs Academic self-efficacy beliefs were assessed through seven
items adapted from Galand (2004)(e.g.,“As long as I do my work, I’msureIcan
succeed this year”; alpha = 0.72). Items were answered on a five-point scale from
1=“totally disagree”to 5 = “totally agree.”Exploratory factor analysis revealed one
single factor (eigenvalue higher than 1) explaining more than 50 % of the variance. Factor
loadings ranged from 0.55 to 0.77.
Informed choice This variable was operationalized as the number of actions of information
undertaken by students in their study choice process. Participants were invited to
consider six actions of information (discussions with teachers, appointments with
guidance counselors, meeting professionals of the field, spontaneously gathering
information about one’s possible education and future career, reflecting upon one’s
professional choice and job prospects, thinking about the different possible higher
education programs after secondary education) and were then asked which ones they
undertook in their choice-making process. The reliability coefficient for this scale was
good (alpha = 0.74).
Outcome variables Continuous variables: Academic achievement was represented by the
average percentage scores in all courses at the end of the academic year. This measure was
collected from department records and used as an overall indicator of achievement. To be
successful in the tertiary education system of the Belgian-French community, an overall final
percentage of 60 is required. Categorical variable: A second measure was created through
categorical variables in order to provide another appraisal of academic achievement.
Achievement scores were divided into four dimensions representing the four typical
student outcomes at the end of the year. The first dimension, entitled “attrition,”
depicted students who obtained between 0 and 30 % as their final percentage: these
students were considered to have dropped out of the academic year. The second
dimension, entitled “failure,”brought together students with a final percentage ranging
from 30 to 60 %: these students actually committed to their academic year but did not
succeed. The third dimension, entitled “success,”represented students with a final
percentage ranging from 60 to 70 %: these students completed the academic year
satisfactorily but did not gain honors. The final group was made up of students whose
final percentage was over 70 % and was called the “honor roll.”Of students, 16.4 %
were classified in the attrition group, 29.5 % in the failure group, 37.1 % in the success
group, and 17 % in the honor roll group.
A person-centered approach to academic achievement 45

Analyses
In line with the assertions made by several authors (Daniels et al. 2008; Peck and Roeser 2003;
Phinney et al. 2005; Valle et al. 2008), k-means cluster analyses were run with regard to
entrance achievement factors in order to identify the most highly similar groups of students
within groups and the most highly dissimilar students between groups. According to Daniels
and colleagues (Daniels et al. 2008), such methods can enable an unbiased investigation of
natural occurring patterns among college students and a better understanding of the combined
effects of the variables of interest. Furthermore, Neuville and Bourgeois (2007)haveargued
that the k-means method is suitable for handling a large data set (>150 subjects) and can
compensate for a low initial partition of data. However, it is important to understand that the
selection and interpretation of clusters are essentially guided by researchers rather than
by analytical results. Nonetheless, to carry out our selection, we considered the
distribution of students in each cluster and the adjustment of the final cluster solution
to the entrance factor variability using multivariate analysis of variance (MANOVA).
Once the final cluster solution was set up, analyses of variance (ANOVA) and
descriptive analyses were used to compare clusters in terms of grade point average.
Finally, categorical measures of achievement were analyzed through cross-tabulation and
chi-square tests.
Results
Preliminary analysis
Table 1displays the zero-order correlations and summary statistics of each study’s variables.
The scores corresponding to skewness and kurtosis were found to be within the normal values.
As expected, high school grade, SES, academic self-efficacy beliefs, and informed choice were
positively correlated (p<.001) with academic achievement. Although correlations were small-
er than expected, these results lend credence to the predictive value of these entrance variables
for final academic achievement.
Tab le 1 Means, standard deviations, skewness, kurtosis, and correlations of the study variables
Min Max MSD Skew. Kurt. 1 2 3 4 5
1. High school
grade
1 4 1.77 0.75 0.64 −0.24 –
2. Socioeconomic
status
Factor
score
0.00 1.00 −0.52 0.24 0.14*** –
3. Academic
self-efficacy
beliefs
1 5 3.52 0.55 −0.21 0.09 0.13*** 0.12*** –
4. Informed choice 0 6 4.54 1.34 −0.83 0.26 0.07** 0.16*** 0.02 –
5. Academic
achievement
0 100 59.1 13.9 −0.80 0.32 0.37*** 0.12*** 0.11*** 0.10*** –
*p< .05;**p<.01;***p<.001
46 M. De Clercq et al.

Cluster analysis
As entrance factor scales have different metrics, each one was standardized through Z-
transformations before being entered into the cluster analysis. Five k-means cluster analyses
were run separately specifying two-, three-, four-, five-, and six-cluster solutions. Based on
several indices such as the number of iterations, sample sizes in each cluster, and interpret-
ability, the six-cluster solution was eventually selected as the most meaningful distinction. The
reliability of this solution was also investigated through the cross-validation procedure de-
scribed below. The final cluster centroids are described in Table 2and illustrated in Fig. 1.
Centroids can be defined as students’means in high school grade, SES, academic self-
efficacy beliefs, and informed choice in each cluster. A cautionary note is nevertheless required
in the interpretation of these scores. As each scale was standardized, a positive centroid does
not reflect a high score per se but rather, a higher score than the overall sample mean.
Centroids must thus be understood as a normative assessment and not as a criterion-
referenced assessment. For instance, a cluster with a positive centroid on informed choice
was composed of students who made more thoughtful study choices than average students
from the sample. Finally, MANOVA was computed with cluster-membership as the between-
subjects factor and the four cluster variables as dependent variables. This analysis revealed
whether clusters differed across entrance variables. The overall MANOVA was significant, and
Roy’s largest root = 2.06, F(5, 2122) = 873.29, p< .001. As shown in Table 2, the univariate
test for each cluster variable was significant and showed that cluster membership explained
more than 50 % of the variance of the four variables used to create the clusters. These results
lend credence to the cluster solution as a meaningful depiction of entrance students’distribu-
tion and support the fact that each selected entrance variable contributes to the cluster solution.
Validation of the cluster solution A cross-validation procedure was set up to assess the
replication of the six-cluster solution (Breckenridge 2000; Phinney et al. 2005). To this end, the
data set was randomly divided into two samples (sample 1, n= 1069 and sample 2, n= 1109).
Subsequently, k-means clusters—specifying a six-cluster solution—were performed separately on
samples 1 and 2 using the cluster centroid derived from the global sample. Based on the work of
Cohen (1960), the agreement between the cluster solutions for the whole sample and for the two
subsamples was average to high (kappa = 0.62 for Sample 1; kappa = 0.58 for sample 2). This
procedure thus proved that the six-cluster solution had good reliability in the data set.
Characteristics of the clusters As mentioned earlier, statistical and theoretical criteria led to
our choice of a six-cluster solution which revealed meaningful profiles highlighting specific
patterns of factors. The description and interpretation of these clusters are presented below.
1. Disadvantaged (n=280; 12.9 %). The first cluster depicted students with the most negative
profiles. These were characterized by a negative centroid on the four entrance achievement
factors. In addition to a score on academic self-efficacy beliefs that was more than one standard
deviation below the mean score, low scores on SES, high school grades, and informed choice
were also reported. A typical student from this cluster would thus originate from a lower
socioeconomic background, report weaker high school grades, have substantially lower
confidence in his/her ability to succeed, and operate a less informed study choice process.
2. Poor performer (n= 429; 19.7 %). The second cluster was labeled poor performer due to
the extremely negative centroid (more than a standard deviation from the mean) on high
A person-centered approach to academic achievement 47

Tab le 2 Cluster centroids and multivariate analysis of variance
Entrance profiles Cluster 1 :
disadvantaged
Cluster 2: poor
performer
Cluster 3:
thoughtless
Cluster 4:
underprivileged
Cluster 5:
apprehensive
Cluster 6:
advantaged
FŊ
2
n280(12.9 %) 429 (19.7 %) 276 (12.7 %) 291 (13.4 %) 417 (19.1 %) 435 (20.0 %)
High school grade −0.66 b −1.04 a −0.23 c −0.06 d 0.71 e 0.87 f 509.49*** 0.55
Socioeconomic status −0.93 b 0.48 de 0.14 c −1.33 a 0.38 d 0.61 e 451.29*** 0.52
Academic self-efficacy beliefs −1.26 a 0.08 c 0.20 c 0.54 d −0.69 b 0.91 e 462.33*** 0.52
Informed choice −0.34 b 0.50 d −1.71 a 0.02 c 0.45 d 0.38 d 458.17*** 0.52
Letters indicate post hoc comparison grouping based on Tukey’s HSD; cluster centroids with different letters differ significantly
*p< .05;**p< .01;***p<.001
48 M. De Clercq et al.

school grade. Indeed, these students reported lower high school grades than the average
student. However, they had also made more thoughtful study choices and came from more
privileged backgrounds than the average freshman.
3. Thoughtless (n= 276; 12.7 %). The third cluster was entitled the thoughtless profile.Students
in this cluster showed a highly negative centroid on informed choice, a negative centroid on
high school grade, but a positive centroid on SES and academic self-efficacy beliefs. This kind
of student came from a more privileged background and entered university with a higher
confidence in his/her ability to succeed compared to the average student, notwithstanding lower
high school grades and a poor study choice process. Students in this profile had not initiated
any real complex study choice process and were thus potentially vulnerable to attrition.
4. Underprivileged (n= 291; 13.4 %).The fourth cluster was particularly characterized by an
extremely negative centroid on socioeconomic background and was labeled the under-
privileged profile. Such students came from a more underprivileged background but had
higher academic self-efficacy beliefs.
5. Apprehensive (n=417; 19.1 %). The fifth cluster exhibited a favorable profile except for
academic self-efficacy beliefs; it was labeled the apprehensive profile. Despite positive
centroids on high school grades, SES, and informed choice, this profile had a negative
centroid on academic self-efficacy beliefs.
6. Advantaged (n= 435; 20.0 %). The final cluster depicted students with the most positive
profiles and was thus labeled the advantaged profile. In contrast to the disadvantaged profile,
these students had initiated a complex study choice, possessed great confidence in their abilities
to succeed, reported high grades in high school, and came from privileged backgrounds.
Fig. 1 Final centroids of entrance achievement factors for each profile
A person-centered approach to academic achievement 49

Clusters and academic outcomes Using a one-way ANOVA, we found significant differ-
ences between the clusters across the average percentage at the end of the year F(5,
1928) = 37.64, p<.001,ŋ
2
= 0.13. The disadvantaged profile exhibited the lowest average final
percentage of the six clusters. The poor performer, thoughtless, and underprivileged profiles
also displayed an average final percentage below the minimum required to pass the academic
year. Special attention should thus be paid to students in these clusters. By contrast, the
apprehensive and advantageous profiles showed average performances above the minimum
60 % required to pass the year in the Belgian educational context.
More precisely, post hoc comparisons across groups based on Tukey’s honestly significant
difference (HSD) test showed that the disadvantaged profile (M= 55.4; SD = 11.6) had a
significantly lower percentage than all other profiles except the poor-performer profile (see
Table 3for post hoc detailed results). Second, the poor performer (M= 57.6; SD = 10.7) had a
significantly lower percentage than the apprehensive and advantaged profiles, but not the other
three profiles. Third, the thoughtless (M= 59.1; SD = 11.3) and underprivileged profiles
(M= 59.9: SD = 10.5) showed a significantly lower percentage than the apprehensive and
advantaged profiles and a higher percentage than the disadvantaged profile. However, they
did not differ from one another and from the poor performers. Fourth, the apprehensive
(M= 63.9; SD=10.9) and advantaged profiles (M= 66.1; SD = 10.9) both had a significantly
higher percentage than the four other profiles but did not differ from one another. The
underprivileged profile was therefore classified in the low achiever level.
Descriptive statistics allowed us to clarify these results. Frequencies were investigated on four
levels of academic outcomes: attrition, failure, success, and the honor roll. Initial results revealed
that failure and attrition rates were 16.4 and 29.5 %, respectively. These percentages are less
significant than the national rates of 25 % for attrition and 35 % for failure (Droesbeke et al. 2008).
Using a chi-square test for independence and cross-tabulation (Greenwood and Nikulin
1996), the association between the six clusters and the four typical student outcomes at the end
of the year was investigated. The results described in Table 4highlight significant variations in
the proportion of academic outcomes from one profile to another (χ
2
(15) = 238.66; p< .001).
Cramer’s V coefficient revealed a moderate association (V= 0.35; p< .001) between the
clusters to which students belonged and final outcomes.
More precisely, the attrition rates were above average for disadvantaged, poor performer,
thoughtless, and underprivileged profiles. Only the apprehensive and advantageous profiles
Tab le 3 Post hoc comparison among the six entrance profiles on final percentage
Entrance profiles Mean on final
percentage
Mean differences across clusters
1 23456
1. Disadvantaged 55.45 a –
2. Poor performer 57.57 ab −2.12 –
3. Thoughtless 59.09 b −3.64* −1.52 –
4. Underprivileged 59.95 b −4.50** −2.38 −0.86 –
5. Apprehensive 63.94 c −8.49*** −6.37*** −4.85*** −3.99** –
6. Advantaged 66.09 c −10.64*** −8.52*** −7.00*** −6.14*** −2.15 –
Letters indicate post hoc comparison grouping based on Tukey’s HSD; means with different letters differ
significantly
*p< .05;**p<.01;***p<.001
50 M. De Clercq et al.

delivered percentages below average. Attrition was especially high for the disadvantaged
profile in which 25.1 % of students dropped out before the end of the first year. Failure rates
were also above average in the disadvantaged, poor performer, thoughtless, and underprivi-
leged profiles. Once again, the failure rate for the disadvantaged profile was particularly high:
more than 40 % of students failed. Conversely, the apprehensive and advantaged profiles
presented failure rates below average. Considering attrition and failure together, more than
65 % of students classified in the disadvantaged profile were unable to cope with the academic
requirements of their first year at university. More surprisingly, only the disadvantaged profile
was found to have a success rate below average. All the other profiles showed a success rate of
approximately 40 %. The main distinction relating to the success rate lay in the honor roll
category where the success rate for the apprehensive and advantaged profiles was more than
twice that of the other profiles. Moreover, the success rate for these two profiles was higher
than 60 % when considering success and honor roll outcomes together.
Discussion
Our goal in this study was to further clarify the possibility of developing an inclusive approach
to freshman achievement that might address the interplay between some important entrance
achievement factors in the Belgian tertiary education context. More precisely, the study
attempted to test the feasibility of identifying meaningful subgroups and analyzed the extent
to which being a member of a specific subgroup impacted on student academic success.
University entrance based on a person-centered approach
The first research question was: “Can meaningful subgroups of freshmen with specific
combinations of entrance variables be identified?”The results provided empirical evidence
Tab le 4 Crosstabs cluster membership × achievement outcomes
Four-level academic outcomes
Attrition Failure Success Honor roll Total
1. Disadvantaged 61 105 61 16 243
25.1 % 43.2 % 25.1 % 6.6 % 100.0 %
2. Poor performer 76 138 153 23 390
19.5 % 35.4 % 39.2 % 5.9 % 100.0 %
3. Thoughtless 49 77 97 26 249
19.7 % 30.9 % 39.0 % 10.4 % 100.0 %
4. Underprivileged 57 84 99 27 267
21.3 % 31.5 % 37.1 % 10.1 % 100.0 %
5. Apprehensive 40 91 148 101 380
10.5 % 23.9 % 38.9 % 26.6 % 100.0 %
6. Advantaged 34 75 159 136 404
8.4 % 18.6 % 39.4 % 33.7 % 100.0 %
Total 317 570 717 329 1933
16.4 % 29.5 % 37.1 % 17 % 100 %
A person-centered approach to academic achievement 51

that various profiles can be identified at the beginning of the year, each constituting a unique
combination of predictors. This finding is in line with several higher education theories which
postulate that students enter university with a set of specific characteristics (Pascarella and
Terenzini 2005). More precisely, in addition to entirely favorable and adverse profiles, cluster
analysis highlighted complex profiles including a combination of high and low scores on the
four entrance variables. These results support two of our assumptions concerning the type of
profile that might emerge from analyses.
First, the assumption that clusters with specific weaknesses could emerge from the analysis
was corroborated by the identification of four subgroups of students with particularly low
scores in one of the four variables investigated. This finding provides rationales to consider the
contribution of each specific line of research in the academic achievement issue. In this line of
thought, research and theories about specific achievement predictors might endorse comple-
mentary perspectives of achievement, each perspective being a good fit for a specific profile of
freshmen. The vast body of knowledge that has attempted to explain the impact of past
performance on achievement (Allen et al. 2008; Elias and MacDonald 2007; Robbins et al.
2006) could be particularly relevant in understanding the adaptation of the poor-performer
profile. Authors who have undertaken in-depth investigations of the actual impact of SES on
achievement (Linnehan et al. 2011;Mills2008; Sackett et al. 2009; Zwick and Green 2007)
have been able to clearly depict specific difficulties with which the underprivileged profile has
to struggle. Bandura’s social cognitive theory (Bandura 1997) could pinpoint specific proximal
variables that could foster achievement in the apprehensive profile. Finally, future time
perspective and educational choice implementation theories (Germeijs and Verschueren
2007; Husman and Lens 1999) may be of major interest for an understanding of the
achievement process of the thoughtless profile. A substantial lens through which to understand
the thoughtless profile might also be found in Berzonsky’s identity process style which
postulates that students in this profile are more diffuse-avoidant-oriented individuals
(Berzonsky and Kuk 2000). However, further investigation is required to corroborate this
assumption.
Second, the depiction of clusters with low and high scores on the four variables partially
confirmed our expectation of clusters composed of variables evolving together. This finding
substantiates the importance of the inclusive approach of achievement (Allen et al. 2010;De
Clercq et al. 2013). We can postulate that students endorsing cumulative strengths or weak-
nesses show specific adaptation to the academic context that differs from students with a
specific weakness. Consequently, an in-depth investigation of the combined impact of achieve-
ment predictors should be an on-going concern in attempts to understand achievement from
disadvantaged and advantaged profiles.
From clusters to final outcomes
The second research question was as follows: “Do different subgroups differ regarding
achievement?”To respond to this question, we highlighted significant differences between
profiles with regard to overall percentages and the proportion of successful students. The
analyses also demonstrated that students in the disadvantaged profile are particularly vulner-
able to attrition and failure and that those in the advantaged profile are presumably better
prepared to face the first year at university. Focusing on the categorical measure of achieve-
ment, results revealed that more than 65 % of students classified in the disadvantaged profile
failed whereas more than 70 % in the advantaged profile passed the year.
52 M. De Clercq et al.

The results also highlighted that the apprehensive profile emerged as a high achiever’s
profile. Thus—notwithstanding the literature that argues that academic self-efficacy beliefs are
a key factor necessary for academic achievement—students with lower confidence in their
abilities to succeed than the average freshman might be good achievers, as this weak point can
be offset by other achievement factors. This finding illustrates the utility of evaluating the
combination of factors to develop a precise conceptualization of academic achievement. The
actual effect of a variable on achievement may depend on how it combines with other student
characteristics. It also partially qualifies the plethora of studies on the strong impact of self-
efficacy beliefs on achievement (Chemers et al. 2001).
It is also worth noting that preliminary results revealed a significant relationship between
the four variables and academic achievement. While high school grades had a moderate
relationship with achievement, SES, informed choice, and self-efficacy beliefs had a weak
positive relationship with final performance. The relatively low impact of these variables was
quite surprising, especially for academic self-efficacy beliefs which are often reported as strong
predictors of achievement. Two main explanations are possible. First as mentioned above, self-
efficacy beliefs differ according to the domain of functioning. Therefore, global confidence in
one’s ability to succeed must essentially be considered to be at the apex of academic self-
hierarchy and to have an indirect impact on academic achievement by initiating the develop-
ment of domain-specific academic self-efficacy beliefs (Zimmerman 2000). Elias and
MacDonald (2007) provide the second explanation. They argue that self-efficacy beliefs are
not good predictors of performance in new academic contexts such as the first year at
university. According to them, when facing a new learning context, past performance is the
main predictor of achievement and demonstrates students’global learning abilities. This line of
thought concurs with our findings.
With regard to our results, achievement differs depending on different subgroups. More
precisely, the combination of risk factors could have a severe negative impact on a student’s
achievement, whereas the impact of an isolated weakness could be offset by a student’s
strengths. Entrance characteristics could thus have a cumulative impact on academic achieve-
ment. However, a more detailed analysis of the results revealed that the scope of this finding
requires further assessment. First, entrance profiles explained only about 13 % of final grade
variance. Second, every cluster included both successful and unsuccessful students. It thus
appears necessary to relativize the predictive value of a student’s entrance characteristics and
consider the impact of the achievement process during the academic year. Following this idea,
students with cumulative entrance weaknesses (disadvantaged profiles) might therefore suc-
ceed if their adaptation during the year is particularly successful.
Practical implications
Taken together, these results can pinpoint—from the very beginning of the year—those
students who are at risk of failure and thus need specific attention. From a pragmatic view,
the early identification of profiles could be useful to prevent drop-out and failure. Such a
person-centered approach could provide tools to capture and consider students’specific
differences in achievement and retention interventions. This perspective could lead to a shift
from the undifferentiated promotion of achievement to the establishment of tailor-made
interventions to fit the specific needs of students according to their profiles. For example,
students from the apprehensive profile might participate in psychosocial self-affirmation
interventions that target their beliefs and feelings about the university in order to promote
A person-centered approach to academic achievement 53

their confidence in their ability to succeed (Cohen et al. 2009). Moreover, characterizing
specific patterns from the very beginning of the year also provides some indications that make
it possible to intervene in the early promotion of academic achievement, before students fall
too far behind to reduce the gaps (Neuville et al. 2013).
To some extent, we might also question the relevance of the university entrance selection
process. Our findings highlighted that, in each entrance profile, a significant proportion of
students actually passed the first year. This assertion was also true for students with cumulative
entrance weaknesses who—in more restrictive systems—would probably not gain entry to
university. Consequently, this paper suggests that the open access system might be considered
a viable perspective to democratize access to higher education, in line with the aspirational
expansion of higher education set out in educational policies in many European countries
(Gale and Parker 2012).
Limitations and future perspectives
Among the limitations of this study, three should be pointed out. First, our approach to
achievement factors has not taken cognitive factors into account. However, cognitive variables
such as learning strategies and self-regulation have been shown to have a significant impact on
achievement (Fenollar et al. 2007; Minnaert and Janssen 1999;Notaetal.2004;Pintrichand
De Groot 1990; Shell and Husman 2008; Vermunt 2005; Ward and Walker 2008). An
estimation of freshmen’s cognitive strategies could thus enrich our approach to students’
entrance profiles. Second, studies have depicted informed choice as a meaningful entrance
factor in an open-access educational system such as in the Netherlands or Belgium. However,
such a variable may be of minor concern in more restrictive systems such as the English
educational system. Replicating this study in a country where access to higher education is
highly contingent on school performance and/or SAT scores may thus be interesting. Finally,
the past performance measure used in the study can be criticized. As no standardized test
scores exist in the Belgian educational context, past performance was measured using a four-
point item assessing high school final average percentages. This measure is quite global and
self-reported, and it does not consider educational track, specialization in secondary education,
and the characteristics of the high schools. Given the circumstances, the reliability of high
school grades as a reliable indicator of students’actual past performance is questionable
(Zwick and Green 2007).
The major contribution of these findings lies in the fact that they highlight the relevance of
taking a person-centered approach into account and make it possible to depict different
meaningful entrance profiles. However, many questions remain unanswered, and a comple-
mentary future issue would be to investigate the relation between these entrance profiles and
more proximal achievement variables such as attendance, social integration, and teaching
practice. We might presumably infer that freshmen within distinct profiles adapt differently to
the academic world. In other words, students’entrance characteristics could entail a specific
achievement process with specific variables. For instance, a poor-performer profile could
involve other adaptation difficulties compared to a thoughtless profile. The impact of proximal
predictors of achievement such as academic engagement, motivation, and social integration
may therefore vary from one profile to another. Several theories support such an assumption.
For example, Astin’s theory of involvement (Astin 1999) postulates that students’pre-entry
attributes will affect their engagement during the academic year. Future studies might therefore
determine whether the pattern of relationships between academic achievement and proximal
54 M. De Clercq et al.

variables is the same across entrance profiles. Further investigation of this question might be
attempted through multigroup structural equation modeling. Based on the literature, an overall
model of the achievement process at university could be tested while investigating the ways in
which it varies depending on students’entrance characteristics.
Another future perspective could be to more deeply assess atypical students in the different
profiles. On the one hand, it would be quite interesting to explore how successful students
from the disadvantaged profile cope with first-year university difficulties to achieve academic
success. On the other hand, scrutinizing students from the advantaged profile who fail or drop
out of the first year of university could also provide a more in-depth understanding of the
achievement issue. Perry et al. (2001) have emphasized the necessity of fully grasping this
“paradox of failure,”namely why bright high school students fail when they enter university.
This assessment might be undertaken by combining a person-centered approach with qualita-
tive procedures.
In conclusion, our approach has provided some clues to differentiate students who cope
with the transition from high school to university from those who do not. However, in order to
deepen our understanding of the former issue and address the latter, our findings need to be
replicated. Moreover, they must be combined with more specific analytical approaches such as
a qualitative approach or structural equation modeling. This would enable a dynamic and
comprehensive conceptualization of academic achievement to be carried out.
Acknowledgments The authors would like to thank Gentiane Boudrenghien, Sandrine Neuville, and Julia
Schmitz for their help with data collection.
References
Adeyemo, D. A. (2007). Moderating influence of emotional intelligence on the link between academic self-
efficacy and achievement of university students. Psychology and Developing Societies, 19(2), 199–213. doi:
10.1177/097133360701900204.
Allen, J., Robbins, S., Casillas, A., & Oh, I.-S. (2008). Third-year college retention and transfer: effects of
academic performance, motivation, and social connectedness. Research in Higher Education, 49(7), 647–
664. doi:10.1007/s11162-008-9098-3.
Allen, J., Robbins, S. B., & Sawyer, R. (2010). Can measuring psychosocial factors promote college success?
Applied Measurement in Education, 23(1), 1–22. doi:10.1080/08957340903423503.
Arulampalam, W., Naylor, R., & Smith, J. (2004). Undergraduate medical education. Factors affecting the
probability of first year medical student dropout inthe UK: a logistic analysis for the intake cohorts of 1980–
92. Medical Education, 38(5), 492–503.
Astin, A. W. (1999). Student involvement: a developmental theory for higher education. Journal of College
Student Development, 40(5), 518–529.
Bandura, A. (1997). Self-efficacy: the exercise of control. New-York: Freeman.
Battle, J., & Pastrana, A., Jr. (2007). The relative importance of race and socioeconomic status among Hispanic
and White students. Hispanic Journal of Behavioral Sciences, 29(1), 35–49. doi:10.1177/
0739986306294783.
Bergman, L. R., Magnusson, D., & El-Khouri, B. M. (2003). Studying individual development in an
interindividual context: a person-oriented approach. Mahwah: Lawrence Erlbaum Associates
Publishers.
Berzonsky, M. D., & Kuk, L. S. (2000). Identity status, identity processing style, and the transition to university.
JournalofAdolescentResearch,15(1), 81–98. doi:10.1177/0743558400151005.
Biémar, S., Philippe, M.-C., & Romainville, M. (2003). L’injonction au projet: paradoxale et infondée ?
Approche longitudinale du choix d’études supérieures. L’orientation scolaire et professionnelle, 32,31–51.
Bong, M., & Skaalvik, E. M. (2003). Academic self-concept and self-efficacy: how different are they really?
Educational Psychology Review, 15(1), 1–40.
A person-centered approach to academic achievement 55

Breckenridge, J. N. (2000). Validating cluster analysis: consistent replication and symmetry. Multivariate
Behavioral Research, 35(2), 261–285. doi:10.1207/s15327906mbr3502_5.
Brown, S. D., Tramayne, S., Hoxha, D., Telander, K., Fan, X., & Lent, R. W. (2008). Social cognitive predictors
of college students’academic performance and persistence: a meta-analytic path analysis. Journal of
Vocational Behavior, 72(3), 298–308.
Busato, V. V., Prins, F. J., Elshout, J. J., & Hamaker, C. (2000). Intellectual ability, learning style, personality,
achievement motivation and academic success of psychology students in higher education. Personality and
Individual Differences, 29(6), 1057–1068. doi:10.1016/s0191-8869(99)00253-6.
Caldas, S. J., & Bankston, C., III. (1997). Effect of school population socioeconomic status on individual
academic achievement. JEducRes,90(5), 269–277.
Chemers, M. M., Hu, L.-t., & Garcia, B. (2001). Academic self-efficacy and first-year college student perfor-
mance and adjustment. Journal of Educational Psychology, 93(1), 55–64.
Cohen, J. (1960). A coefficient of agreement for nominal scales. Educational and Psychological Measurement,
20,37–46.
Cohen, G. L., Garcia, J., Purdie-Vaughns, V., Apfel, N., & Brzustoski, P. (2009). Recursive processes in self-
affirmation: intervening to close the minority achievement gap. Science, 324, 400–403.
Daniels, L. M., Haynes, T. L., Stupnisky, R. H., Perry, R. P., Newall, N. E., & Pekrun, R. (2008). Individual
differences in achievement goals: a longitudinal study of cognitive, emotional, and achievement outcomes.
Contemporary Educational Psychology, 33(4), 584–608. doi:10.1016/j.cedpsych.2007.08.002.
De Clercq, M., Galand, B., Dupont, S., & Frenay, M. (2013). Achievement among first-year university students:
an integrated and contextualised approach. European Journal of Psychology of Education, 28(3), 641–662.
doi:10.1007/s10212-012-0133-6.
Díaz, R. J., Glass, C. R., Arnkoff, D. B., & Tanofsky-Kraff, M. (2001). Cognition, anxiety, and prediction
of performance in 1st-year law students. Journal of Educational Psychology, 93(2), 420–429. doi:
10.1037/0022-0663.93.2.420.
Dollinger, S. J., Matyja, A. M., & Huber, J. L. (2008). Which factors best account for academic success: those
which college students can control or those they cannot? Journal of Research in Personality, 42(4), 872–
885. doi:10.1016/j.jrp.2007.11.007.
Droesbeke, J.-J., Lecrenier, C., Tabutin, D., & Vermandele, C. (2008). Réussite ou échec à l’université:
trajectoire des étudiants en Belgique francophone. Paris: Ellipses.
Dumais, S. A. (2005). Accumulating adversity and advantage on the path to postsecondary education:
an application of a person-centered approach. Social Science Research, 34(2), 304–332. doi:10.
1016/j.ssresearch.2004.04.001.
Eccles, J. S., & Wigfield, A. (2002). Motivational beliefs, values, and goals. Annual Review of Psychology, 53, 109–132.
Elias, S. M., & MacDonald, S. (2007). Using past performance, proxy efficacy, and academic self-
efficacy to predict college performance. Journal of Applied Social Psychology, 37(11), 2518–2531.
doi:10.1111/j.1559-1816.2007.00268.x.
Fass, M. E., & Tubman, J. G. (2002). The influence of parental and peer attachment on college students’
academic achievement. Psychology in the Schools, 39(5), 561.
Fenollar, P., Romajn, S., & Cuestas, P. J. (2007). University students’academic performance: an integrative
conceptual framework and empirical analysis. British Journal of Educational Psychology, 77(4), 873–891.
Galand, B. (2004). Le rôle du contexte scolaire et de la démotivation dans l’absentéisme des élèves. Revue des
Sciences de l’éducatio n, 30,125–142.
Gale, T., & Parker, S. (2012). Navigating change: a typology of student transition in higher education. Studies in
Higher Education, 39(5), 734–753. doi:10.1080/03075079.2012.721351.
Germeijs, V., & Verschueren, K. (2007). High school students’career decision-making process:
consequences for choice implementation in higher education. Journal of Vocational Behavior,
70(2), 223–241. doi:10.1016/j.jvb.2006.10.004.
Greenwood, P., & Nikulin, M. (1996). A guide to chi-squared testing (Wiley Series in Probability and Statistics).
New York: Wiley.
Hackett, G., Betz, N. E., Casas, J. M., & Rocha-Singh, I. A. (1992). Gender, ethnicity, and social cognitive
factors predicting the academic achievement of students in engineering. Journal of Counseling Psychology,
39(4), 527–538. doi:10.1037/0022-0167.39.4.527.
Hayenga, A. O., & Corpus, J. H. (2010). Profiles of intrinsic and extrinsic motivations: a person-centered
approach to motivation and achievement in middle school. Motivation and Emotion, 34(4), 371–383.
doi:10.1007/s11031-010-9181-x.
Heikkilä, A., Niemivirta, M., Nieminen, J., & Lonka, K. (2011). Interrelations among university students’
approaches to learning, regulation of learning, and cognitive and attributional strategies: a person oriented
approach. Higher Education, 61(5), 513–529. doi:10.1007/s10734-010-9346-2.
56 M. De Clercq et al.

Husman, J., & Lens, W. (1999). The role of the future in student motivation. Educational Psychologist, 34(2),
113–125.
Kennett, D., Young, A. M., & Catanzaro, M. (2009). Variables contributing to academic success in an
intermediate statistics course: the importance of learned resourcefulness. Educational Psychology, 29(7),
815–830. doi:10.1080/01443410903305401.
Lens, W., Simons, J., & Dewitte, S. (2002). From duty to desire: the role of students’future time perspective and
instrumentality perceptions for study motivation and self-regulation. In F. Pajares & T. Urdan (Eds.),
Adolescence and education. Vol. 2. Academic motivation of adolescents. Greenwich: Information Age
Publishing.
Linnehan, F., Weer, C. H., & Stonely, P. (2011). High school guidance counselor recommendations: the role of
student race, socioeconomic status, and academic performance. Journal of Applied Social Psychology, 41(3),
536–558. doi:10.1111/j.15 59-1816.2011.00725.x.
Magnusson, D. (1999). On the individual: a person-oriented approach to developmental research. European
Psychologist, 4(4), 205–218. doi:10.1037/0003-066x.54.7.462.
Mills, C. (2008). Reproduction and transformation of inequalities in schooling: the transformative potential of the
theoretical constructs of Bourdieu. British Journal of Sociology of Education, 29(1), 79–89.
Minnaert, A., & Janssen, P. J. (1999). The additive effect of regulatory activities on top of intelligence in relation
to academic performance in higher education. Learning and Instruction, 9(1), 77–91. doi:10.1016/s0959-
4752(98)00019-x.
Neuville, S., Bourgeois, E., & Frenay, M. (2007). Task value, self-efficacy and goal orientations: impact on self-
regulated learning, choice and performance among university students. Psychologica Belgica, 47(1–2), 95–117.
Neuville, S., Frenay, M., Noël, B., & Wertz, V. (2013). Persévérer et réussir à l’université (pp. 107–(134).
Louvain-la-Neuve: Presses universitaires de Louvain.
Nota, L., Soresi, S., & Zimmerman, B. J. (2004). Self-regulation and academic achievement and resilience:
a longitudinal study. International Journal of Educational Research, 41(3), 198–215. doi:10.1016/
j.ijer.2005.07.001.
OECD. (2013). Education at a glance 2013: OECD indicators. Retrieved from http://www.oecd.org/edu/
eag2013%20%28eng%29–post-B%C3%A0T%2013%2009%202013%20%28eBook%29-XIX.pdf
Padgett, R. D., Goodman, K. M., Johnson, M. P., Saichaie, K., Umbach, P. D., & Pascarella, E. T. (2010). The
impact of college student socialization, social class, and race on need for cognition. New Directions for
Institutional Research, 2010(145), 99–111. d oi:10.1002/ir.324.
Pascarella, E., & Terenzini, P. (2005). How college affects students. 2, A third decade of research. San Francisco:
Jossey-Bass.
Peck, S. C., & Roeser, R. W. (2003). Editors’notes. Special issue: person-centered approaches to studying
development in context. New Directions for Child and Adolescent Development, 2003(101), 1–1.
doi:10.1002/cd.78.
Pekrun, R., Goetz, T., Titz, W., & Perry, R. P. (2002). Academic emotions in students’self-regulated learning and
achievement: a program of qualitative and quantitative research. Educational Psychologist, 37(2), 91–105.
Perry, R. P., Hladkyj, S., Pekrun, R. H., & Pelletier, S. T. (2001). Academic control and action control in the
achievement of college students: a longitudinal field study. Journal of Educational Psychology, 93(4), 776–
789. doi:10.1037/0022-0663.93.4.776.
Phan, H. P. (2009). Relations between goals, self-efficacy, critical thinking and deep processing strategies: A path
analysis. Educational Psychology, 29(7), 777-799. doi:10.1080/01443410903289423
Phinney, J. S., Dennis, J. M., & Gutierrez, D. M. (2005). College orientation profiles of Latino students from low
socioeconomic backgrounds: a cluster analytic approach. Hispanic Journal of Behavioral Sciences, 27(4),
387–408. doi:10.1177/0739986305280692.
Pintrich, P. R. (2003). A motivational science perspective on the role of student motivation in learning and
teaching contexts. Journal of Educational Psychology, 95(4), 667–686.
Pintrich, P. R., & De Groot, E. V. (1990). Motivational and self-regulated learning components of classroom
academic performance. Journal of Educational Psychology, 82(1), 33–40.
Price, L. (2013). Modelling factors for predicting student learning outcomes in higher education. In D. Gijbels, V.
Donche, J. T. E. Richardson, & J. D. Vermunt (Eds.), Learning patterns in higher education: dimensions and
research perspectives. New York: Taylor & Francis.
Richardson, M., Abraham, C., & Bond, R. (2012). Psychological correlates of university students’academic
performance: a systematic review and meta-analysis. Psychological Bulletin, 138(2), 353–387. doi:10.1177/
0016440206200300910.1177/001644020620030092002-01523-009.
Robbins, S. B., Lauver, K., Le, H., Davis, D., Langley, R., & Carlstrom, A. (2004). Do psychosocial and study
skill factors predict college outcomes? A meta-analysis. Psychological Bulletin, 130(2), 261–288. doi:
10.1037/0033-2909.130.2.261.
A person-centered approach to academic achievement 57

Robbins, S. B., Allen, J., Casillas, A., Peterson, C. H., & Le, H. (2006). Unraveling the differential effects of
motivational and skills, social, and self-management measures from traditional predictors of college
outcomes. Journal of Educational Psychology, 98(3), 598–616. doi:10.1037/0022-0663.98.3.598.
Roeser, R. W., & Peck, S. C. (2003). Patterns and pathways of educational achievement across adolescence: a
holistic-developmental perspective. New Directions for Child and Adolescent Development, 101,39–62. doi:
10.1002/cd.81.
Sackett, P. R., Kuncel, N. R., Arneson, J. J., Cooper, S. R., & Waters, S. D. (2009). Does socioeconomic status
explain the relationship between admissions tests and post-secondary academic performance? Psychological
Bulletin, 135(1), 1–22. doi:10.1037/a0013978.
Sauvé, L., Debeurme, G., Martel, V., Wright, A., & Hanca, G. (2007). Soutenir la persévérance des étudiants (sur
campus et à distance) dans leur première session d’études universitaires: constats de recherche et
recommandations. Revue Internationale Des Technologies En Pédagogie Universitaire, 4(3), 58–72.
Shell, D.-F., & Husman, J. (2008). Control, motivation, affect, and strategic self-regulation in the college
classroom: a multidimensional phenomenon. Journal of Educational Psychology, 100(2), 443–459.
Sirin, S. R. (2005). Socioeconomic status and academic achievement: a meta-analytic review of research. Review
of Educational Research, 75(3), 417–453. doi:10.3102/00346543075003417.
Tinto, V. (1982). Limits of theory and practice in student attrition. Journal of Higher Education, 53, 687–700.
Tinto, V. (1997). Classrooms as communities. Exploring the educational character of student persistence. Journal
of Higher Education, 68(6), 599–623.
Torres, J. B., & Solberg, V. S. (2001). Role of self-efficacy, stress, social integration, and family support in Latino
college student persistence and health. Journal of Vocational Behavior, 59(1), 53–63.
Valle, A., Núñez, J. C., Cabanach, R. G., González-Pienda, J. A., Rodríguez, S., Rosário, P., & Muñoz-Cadavid,
M. A. (2008). Self-regulated profiles and academic achievement. Psicothema, 20(4), 724–731.
Vandamme, J.-P., Superby, J. F., & Meskens, N. (2005). Freshers’achievement: prediction methods and influent
factors. Mons, Belgique: Paper presented at the Higher Education Multijuridictionality and Globalisation
Conference.
Vansteenkiste, M., Sierens, E., Soenens, B., Luyckx, K., & Lens, W. (2009). Motivational profiles from a self-
determination perspective: the quality of motivation matters. Journal of Educational Psychology, 101(3),
671–688. doi:10.1037/a0015083.
Vermunt, J. (2005). Relations between student learning patterns and personal and contextual factors and
academic performance. Higher Education, 49(3), 205–234. doi:10.1007/s10734-004-6664-2.
Ward, P. J., & Walker, J. J. (2008). The influence of study methods and knowledge processing on academic
success and long-term recall of anatomy learning by first-year Veterinary students. Anatomical Sciences
Education, 1(2), 68–74. doi:10.1002/Ase.12.
Willingham, W. W., Pollack, J. M., & Lewis, C. (2002). Grades and test scores: accounting for observed
differences. Journal of Educational Measurement, 39(1), 1–37. doi:10.1111/j.1745- 3984. 2002.tb01133.x.
Zimmerman, B. J. (2000). Self-efficacy: an essential motive to learn. Contemporary Educational Psychology, 25,
82–91.
Zwick, R., & Green, J. G. (2007). New perspectives on the correlation of sat scores, high school grades, and
socioeconomic factors. Journal of Educational Measurement, 44(1), 23–45.
Mikaël De Clercq. Université Catholique de Louvain, Research Institute of Psychological Sciences, Faculty of
Psychology and Education, 10 Place Cardinal Mercier, B-1348 Louvain-La-Neuve, Belgium. Phone: 00 32
(0)10 47 86 27; URL: mikael.declercq@uclouvain.be, Web site: www.uclouvain.be/mikael.declercq
Currentthemesofresearch:
Mikaël De Clercq is a Ph.D student at the Université catholique de Louvain.
His research interests focus on the predictors of academic achievement among first-year university students and
on the impact of motivation on students’learning processes.
Most relevant publications in the field of Psychology of Education:
De Clercq, M., Galand, B., Dupont, S., & Frenay, M. (2013). Achievement among first-year university students:
an integrated and contextualised approach. European Journal of Psychology of Education, 28(3), 641–662.
doi: 10.1007/s10212-012-0133-6
58 M. De Clercq et al.

De Clercq, M., Galand, B., & Frenay, M. (2013). Chicken or the egg: longitudinal analysis of the causal dilemma
between goal orientation, self-regulation and cognitive processing strategies in higher education. Studies in
Educational Evaluation, 39(1), 4–13. doi: http://dx.doi.org/10.1016/j.stueduc.2012.10.003
De Clercq, M., Galand, B., & Frenay, M. (2013). Learning processes in higher education: providing new insights
into the effects of motivation and cognition on specific and global measuresof achievement. In D. Gijbels, V.
Donche, J. T. E. Richardson & J. D. Vermunt (Eds.), Learning Patterns in Higher Education: Dimensions
and research perspectives:Taylor&Francis.
Benoît Galand. Department of Psychology, Université Catholique de Louvain, 10 Place Cardinal Mercier,
? B-1348 Louvain-La-Neuve, Belgium. Phone: benoit.galand@uclouvain.be; Web site: www.uclouvain.be/
benoit.galand
Currentthemesofresearch:
Benoît Galand obtained a Ph.D. degree in psychology from the University of Louvain in 2001.
He is a professor in the department of Psychology of UCL. His research interests revolve around the effects of
instructional practices on student motivation, learning, and psycho-social adaptation. He lectures in educational
psychology, pedagogy, and research methodology to graduate and undergraduate students.
Mariane Frenay. Department of Psychology, université Catholique de Louvain, 10 Place Cardinal Mercier,
? B-1348 Louvain-La-Neuve, Belgium. Phone: mariane.frenay@uclouvain.be, Web site : http://www.
uclouvain.be/mariane.frenay
Currentthemesofresearch:
Mariane Frenay is a professor at the Université catholique de Louvain, Belgium
(Dean of the faculty of Psychology and Education) and received her PhD in Instructional Psychology in 1994.
She is an active researcher in the field of higher education teaching and learning and faculty development. She
has been a member of the UNESCO Chair of university teaching and learning at her university since 2001 and
has been teaching for several years in Ma ster’s and Doctoral programs in Psychology and Education.
A person-centered approach to academic achievement 59