Christina Stage
Umeå University, Department of Educational
Measurement
Sweden has
a long tradition of compulsory, comprehensive education. As early as in 1842
the first law was passed and signed by the king that there should be at least
one proper school with a trained teacher in each municipality in the country. Since
then there have been several school reforms which have tried to create a school
system combining quality with equality.
Primary and
secondary education are common for all children in Sweden. In Autumn 1995, 930
000 pupils attended primary and secondary school.
After nine
years of compulsory, integrated education the students can choose between
different study lines (now changing to programmes) in upper secondary school. Admittance
to, or rather placement in, upper secondary school is based on average marks
from the ninth year in lower secondary school. The choices offered are between
five different theoretical study lines/programmes, all preparing for higher
education, and about 35 different, vocationally oriented study
lines/programmes. About half the students choose one of the theoretical study
lines.
In primary
school there is no marking. In lower secondary school marks are given only in
grades eight and nine.
Up to now
the marking system in Swedish secondary schools has been norm- or
group-referenced. The reference has been made to all pupils in the country of
the same grade each year. The scale of marks has ranged from one to five, where
one has been the lowest and five the highest and three should be the average
mark in each subject. The marks have been comparable all over the country and
the comparability has been ensured by centrally constructed and administered
standardised tests in the core subjects. The results from these tests were used
to decide the average of the class, i.e. the individual teacher was told how
his/her class compared to other classes in the country. If the class average was
above or below the average of all classes in the country the teacher was
supposed to adjust the class average accordingly. The test results were not
decisive for the marks of individual pupils.
In upper
secondary school marks are given at the end of each term. The marks have, up to
now, been norm or group-referenced but the norm-groups have been all other
students studying the same subject. That means that as different subjects are
studied at different study lines the norm-groups have been different for the
different study lines.
The marking
system in Sweden is now being changed from a norm-referenced to a criterion- or
goal-referenced system. The situation is a bit confused at the moment as the
goals or criteria for different marks have not yet been finally decided.
At the
university level the marking is criterion-referenced with only three levels:
failed, passed and passed with distinction.
The SweSAT
was introduced in 1977 in connection with a reform of the universities and
colleges. It was felt that an admission test would provide a possible solution
to two basic problems (1) how to find a method of selection which could be used
for applicants without formal qualifications; and (2) how to reduce the
decisive role played by marks in the selection process. When the test was first
introduced it was, however, only made available for a relatively small group of
applicants (those who were at least 25 years old and had at least four years of
work experience). Only since 1991 has the test been used for all applicants.
From 1977
to 1989, as long as the use of the SweSAT was restricted to the above-mentioned
group the number of persons taking the test was approximately 10 000 each
year; 6 000 in the spring and 4 000 in the autumn. Since 1990 the number of
testtakers has increased dramatically to around 140 000 persons each year; 75 -
80 000 in the spring and 55 - 60 000 in the autumn.
At present
the test consists of 148 multiple choice questions distributed on six subtests.
The results are transformed to a standard scale from 0.0 to 2.0 where 2.0 is
the highest result. The test is administered twice a year, in spring and
autumn. Students are allowed to take the test as many times as they wish and
for those who have several results the best one is used for application. In
principle it is optional to take the test; in reality, however, test results
can be seen as necessary, since only applicants with top marks dare refrain
from taking the test. The content of the test is shown in table 1.
Table 1: The Swedish
Scholastic Aptitude Test
|
Subtest |
Abbreviated |
Items |
Time
(min) |
|
Vocabulary |
WORD |
30 |
15 |
|
Data
Sufficiency (Quantitative reasoning) |
DS |
20 |
45 |
|
Reading
comprehension |
READ |
24 |
60 |
|
Interpretation
of diagrams, tables and maps |
DTM |
20 |
55 |
|
General information |
GI |
30 |
25 |
|
English
reading comprehension |
ERC |
24 |
50 |
|
Total |
|
148 |
4 hrs 10
min |
Vocabulary
(WORD) measures understanding of words and concepts, and consists of items
where the task is to identify which of five presented words has the same
meaning as a given word. Both Swedish and foreign words are included in the
subtest.
Data
Sufficiency (DS) aims at measuring numerical reasoning ability. In each item a
problem is presented, and the task is to decide whether the information
presented is sufficient to allow solution of the problem. The response format
is fixed, so each item presents the same five alternatives. The subtest is
designed to put as little premium as possible on mathematical knowledge and
skills in favour of problem-solving and reasoning.
Reading
Comprehension (READ) measures Swedish reading comprehension in a wide sense. The
examinees are presented with six texts and four multiple choice questions in
relation to each text. Each text comprises about one printed page. Some items
ask about particular pieces of information but most items are designed to
require understanding of larger parts of the text or the text in its entirety.
Interpretation
of Diagrams, Tables and Maps (DTM) consists of 10 collections of tables,
diagrams and/or maps which present information about a topic, with two multiple
choice questions in relation to each collection. The degree of complexity of
the items varies from simply reading off a presented graph, to some where
information from different sources must be combined.
General
Information (GI) measures knowledge and information from many different areas. The
test is broader than traditional school achievement tests and asks about
information that a person may acquire over an extended period of time in different
contexts such as work and education, or social, cultural and political
activities.
English
Reading Comprehension (ERC) is of the same general type as the subtest READ. However,
in this subtest there is more variability as to both the texts and item formats
used. The test consists of 8 to 10 texts of different lengths. Most texts are
followed by one or more multiple choice questions with four alternatives. In
one of the texts, some words are omitted, and the examinee is supposed to
select the omitted word from four alternatives presented alongside the text.
The SweSAT
is supposed to measure acquired (developed) abilities and it makes use of the
kind of verbal and mathematical skills that develop over the years, both in and
out of school. The content of the test does not reflect any specific curriculum
although it is designed to be consistent with school based learning.
The test is
designed for selection to all different types of university programmes and
therefore it is intended to measure the students' general aptitude for studies.
Since the test is a selection test it is supposed to rank the applicants as
fairly as possible according to their expected academic success. Other
requirements on the test are:
·
The
test should be in line with the aims and content of higher education.
·
The
test must not have negative effects on the education in upper secondary school.
·
It
should be possible to score the test fast, cheaply and objectively.
·
It
should not be possible for an individual to improve his/her test result by
means of mechanical exercises or by learning special principles for problem
solving.
·
The
examinees should experience the test as meaningful and suitable.
·
The
demand for unbiased recruitment should be observed. No group should be
discriminated against because of gender or social class.
·
The
test should also be varied and cover many different content areas. It is
possible to find the answers to roughly half of the questions in the material
provided. In order to answer the remaining questions some background knowledge
is necessary.
On the
whole the test has been surprisingly well received by testtakers as well as
educational institutions. It is now accepted as a major alternative to school
marks as selection instrument and it has even been suggested as a substitute
now that the marking system is being changed.
One reason
for this acceptance of the SweSAT might be that the test was introduced
"as a second chance" and has been regarded as such. Another reason
might be that the test along with the scoring key has always been made public
as soon as the test has been administered, which means that the testtakers
have the opportunity to control (and discuss) their results on every single
item. A final reason might be that the test is a good one or at least that the
testtakers really experience it as a meaningful and suitable selection instrument
for higher education.
In Sweden
there are six universities, 16 university colleges and six specialized
institutions of higher education. The difference between the universities and
the other institutions of higher education is that graduate programmes are only
offered by the universities.
Approximately
50 000 students are admitted to higher education every year and quite a few of
the study programmes offered have many more applicants than available study
places. As a result of the high unemployment rate, the competition for study
places has been growing. Even though the government has increased the number of
study places the number of applicants has increased still more. Therefore
selection for the study places must take place and for many of the study
programmes the competition is very keen.
The
selection to higher education has changed substantially during the last three
decades. Previously the only selection instrument was marks from upper
secondary school. In 1977 the SweSAT was introduced as a selection instrument,
but only for a small group of applicants. In 1991 the selection rules were
changed again and since then all applicants can use test results as an
alternative to marks.
A
noteworthy feature of the Swedish selection system is that the applicants may
use either marks or test results, whichever is most favourable. This means
that, even though it is optional to take the test, so far, most students are
taking it. One of the main reasons for making SweSAT scores available for all
applicants was to make the average marks from upper secondary less crucial
than they had been before and to make it easier for students to be admitted to
higher education immediately after leaving upper secondary school. The SweSAT
was to give students who had not managed to get top marks, a second chance of
admittance.
Originally
selection to approximately 60 per cent of the study places was made on the
basis of the applicants' marks and selection to the remaining 40 per cent was
based on the results on the SweSAT. Since 1993 the universities and colleges
are autonomous in deciding their admission procedures and selection devices. No
major changes have taken place yet, however, and still usually 60 per cent of
the study places are allocated on basis of average marks from upper secondary
school and 40 per cent on basis of test scores.
Medical
education is provided at the six universities in Sweden and it is one of the
study programmes for which the competition is the very hardest. These study
programmes have also been those most eager to make use of the right to decide
for themselves how to select students, and therefore the systems vary quite a
bit.
At Umeå
University 61 students are admitted each term and starting this autumn the
selection is made in two stages. In stage one, the 122 applicants with the
highest scores on the SweSAT (top marks in the core subjects, i.e. the subjects
necessary to qualify for the programme, are given some extra credit) are chosen
and invited for an interview. The interviews are made by teachers/doctors at
the university and the aim is to sort out those students whose personality,
attitudes or reasons for studying medicine are less suitable for the medical
profession.
At
Linköping University 40 students are admitted to the medical programme each
term, half of those are selected by local rules. In stage one all students who
have Linköping as their first choice and have accepted to take part in the
local admittance procedure are ranked according to average marks and SweSAT
results. A number corresponding to six times the final number of admitted are
invited to Linköping to write their autobiography, motivate why they want to
study medicine and write a short essay on a given subject. After evaluation of
the outcomes 50 per cent of these applicants are interviewed by two persons -
one teacher/doctor and one layman with experience in interviewing people. After
the interviews the interviewers make a common ranking of the applicants and the
upper third is accepted.
At Uppsala
University 55 students are admitted each term, ten of which are selected after
special tests and an interview.
At
Gothenburg University 57 students are admitted each term, so far, all in the
central selection procedure. From next autumn, however, the selection will be
locally made in Gothenburg and in a two stage procedure similar to the procedure
used in Umeå.
At Lund
University 82 students are admitted each term all in the central procedure. In
Lund they will not start with local selection until autumn 1997.
At
Karolinska Institute in Stockholm 120 students are admitted to the medical
programme each term and the institute got special permission as early as 1992
to try out local admission to some of their study places. The main reason for
Karolinska Institute to try out new methods for selection to their medical programme
was that they felt that the selection procedures used for central admission,
i.e. mark averages and SweSAT scores, when used alone, failed to give
satisfactory information about the applicants' suitability for the medical
profession or their motivation for medical studies.
At
Karolinska Institute one third of the total number of admitted or 40 study
places are allocated locally. The selection is made in three stages:
Applicants
with a result of at least 1.6 on the SweSAT are invited to the Institute where
they are asked to write a short essay on one of three suggested topics, a short
autobiography and a motivation for their wishing to become a doctor. The chosen
applicants are then interviewed twice, first by a teacher/doctor at the
Institute and then by a psychologist. These interviews are semistructured and
aim at finding out the applicant's motivation, maturity, judgement and
intellectual mobility. The results of the interviews are evaluated and the
applicants who are regarded as best suited for the the studies and the
profession are chosen.
This
admittance procedure was evaluated after a trial period where (1) study
intermissions and drop out rates, (2) number of courses passed during the four
first semesters and (3) results on the preclinical examination at the end of
the fourth semester in the medical programme, were investigated. Results obtained
by the locally admitted students were compared with those obtained by students
who entered the programme as a result of central admission.
Students
who had been locally admitted - in spite of lower average marks and test scores
- performed as well as centrally admitted students. The lower limit set on the
SweSAT seems to guarantee that the students possess the intellectual capacity
necessary to meet the requirements of the theoretical parts of the programme.
Altogether,
415 students are selected for the medical study programmes each term. 140 of
these students are admitted after some special local selection procedure
usually containing two or three steps, where the SweSAT always constitute the
first step and where the last step is an interview.
Gustafsson,
J-E., Wedman, I. & Westerlund, A. (1992). The Dimensionality of the Swedish
Scholastic Aptitude Test. Scandinavian Journal of Educational Research, Vol.
36, No 1.
Hindbeck, H. Hagenfeldt,
K. & Åberg, H. (1994). Lokal antagning till läkarutbildning vid Karolinska
institutet. Stencil, Karolinska institutet.
Holmberg, C. (1992).
Antagningen till Hälsouniversitetets läkar- och sjukgymnastlinjer. Universitetet
i Linköping, LiU-PEK-R-157.
Holmberg,
C. (1995). Alternativ antagning till högskolan. Stencil, Linköping universitet.
Stage, C.
(1992). Gender Differences on Two Instruments Used for Admission to Higher
Education. To be published in Admission to Higher Education: Issues and Practice.
Selected papers from the 18th IAEA Annual Conference.
Stage, C.
(1993). Gender Differences on the SweSAT. A Review of Studies since 1975. Department
of Educational Measurement, Umeå University, EM No 7.
Stage, C.
(1993). Average School Marks and Results on the SweSAT. Department of
Educational Measurement, Umeå University, EM No 4.
Stage, C.
(1994). Use of Assessment Outcomes in Selecting Candidates for Secondary and
Tertiary Education: A Comparison. Paper presented at the 20th Annual IAEA
Conference, Wellington, New Zealand.
Wedman, I.
(1992) Selection to Higher Education in Sweden, Department of Educational
Measurement, Umeå University. EM No 1.
Wedman, I. (1994). The Swedish Scholastic Aptitude Test: Development, Use and Research. Educational Measurement: Issues and Practice. Vol. 13, No 2.