Content uploaded by Joakim Malm
Author content
All content in this area was uploaded by Joakim Malm on Nov 24, 2015
Content may be subject to copyright.
This article was downloaded by: [Lund University Libraries]
On: 05 March 2015, At: 07:16
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Studies in Higher Education
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/cshe20
Supplemental instruction for improving
first year results in engineering studies
Joakim Malm a , Leif Bryngfors a & Lise-Lotte Mörner a
a Center for Supplemental Instruction, School of Engineering ,
Lund University , P.O. Box 118, S-22100 , Lund , Sweden
Published online: 24 May 2011.
To cite this article: Joakim Malm , Leif Bryngfors & Lise-Lotte Mörner (2012) Supplemental
instruction for improving first year results in engineering studies, Studies in Higher Education, 37:6,
655-666, DOI: 10.1080/03075079.2010.535610
To link to this article: http://dx.doi.org/10.1080/03075079.2010.535610
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or
arising out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
Studies in Higher Education
ISSN 0307-5079 print/ISSN 1470-174X online
© 2012 Society for Research into Higher Education
http://www.tandfonline.com
Supplemental instruction for improving first-year results in
engineering studies
Joakim Malm*, Leif Bryngfors and Lise-Lotte Mörner
Center for Supplemental Instruction, School of Engineering, Lund University, P.O. Box 118, S-
22100 Lund, Sweden
Taylor and Francis
CSHE_A_535610.sgm10.1080/03075079.2010.535610Studies in Higher Education0307-5079 (print)/1470-174X (online)Article2011Society for Research into Higher Education0000000002011Dr JoakimMalmJoakim.Malm@kansli.lth.se
Many studies have been made on the impact of supplemental instruction in
supported courses, with most showing significantly better examination results
for students attending supplemental instruction in comparison to those who do
not. However, remarkably little attention has been devoted to following up
whether the benefits of supplemental instruction reach beyond the course it
supports. The present study focuses on the influence of supplemental instruction
on the overall academic performance during the first year, for undergraduate
engineering students at a Swedish university. The results show that students
with average and high supplemental instruction attendance do significantly
better than students not attending in terms of overall first-year credit
performance. Students with low, average and high prior academic achievement
all benefit from attending supplemental instruction sessions. The data also
suggests that the transferable effects of study strategies and skills to non-
supplemental instruction courses are substantial for attendees, leading to better
results in these courses.
Keywords: supplemental instruction; engineering studies; first-year experience
Introduction
Many university programmes experience difficulty in retaining students. At the
School of Engineering at Lund University, Sweden, the percentage of matriculating
students who successfully complete their MSc engineering degree is about 60% (the
MSc engineering degree is the first university degree for the majority of the students
– a BSc degree is optional). Many of the students who drop out do so during the first
year. Some 79% of the students who began their engineering studies in 2009 returned
in the autumn of 2010. Another challenge is timely graduation. Few of the successful
students take their degree within the stipulated five years, which is due in part to diffi-
culties adjusting to studies at university level. What difficulties do new students face
when entering the School of Engineering? There are many, not all of which are related
to their studies:
●The courses are at a higher level and faster pace, thus requiring more in-depth
understanding.
●The number of study hours per week, including scheduled hours, increases by
roughly 30–40% at university.
*Corresponding author: Email: joakim.malm@kansli.lth.se
Vol. 37, No. 6, September 2012, 655–666
http://dx.doi.org/10.1080/03075079.2010.535610
Downloaded by [Lund University Libraries] at 07:16 05 March 2015
●Students’ responsibility for their studies increases. There is less feedback and
guidance from teachers in relation to their studies. Many students are dependent
on state grants for their subsistence, and need to be successful in their studies to
continue to receive these grants.
●For many students the social network disappears. They move away, often far
from family and friends, to begin a new life in a different town.
Consequently, it is a tough challenge that awaits new students at the School of Engi-
neering, and therefore it is not particularly strange that many drop out. For a couple
of years now, supplemental instruction has been used on a broader scale to facilitate
the new students’ start. Supplemental instruction was developed in 1973, at the
University of Missouri in Kansas City, to increase student success in ‘difficult’
courses (Hurley, Jacobs, and Gilbert 2006). Supplemental instruction as a concept has
since spread widely and is used at more than 1500 university colleges and universities
in nearly 30 countries (Martin 2008). Supplemental instruction is not just a method but
an attitude to learning, where inner motivation and curiosity are the driving forces, and
where the main emphasis is on self-governing and collective learning (Olstedt 2005).
Supplemental instruction complements the regular studies of a course. The idea
behind it is that learning a subject is enhanced by an exchange of thoughts and ideas
between students. The supplemental instruction sessions are guided by a senior
student. This senior student takes the role of a facilitator and aids clarification of tough
questions within the subject by asking questions, initiating work in small groups, and
coordinating the presentation of conclusions. This senior student attends a training
course in how to be a supplemental instruction leader, and is provided with tools to
use during the sessions.
Supplemental instruction was introduced in 1994 at the School of Engineering at
Lund University as an experimental project to increase student retention and improve
performance in ‘difficult’ courses (Bruzell-Nilsson and Bryngfors 1996). Recently the
overall objective has changed slightly, and the aim today is at bridging the gap
between secondary school and university, in addition to supporting students in ‘diffi-
cult’ initial compulsory courses. The underlying objectives (besides increasing under-
standing of course content) of the supplemental instruction programme include:
●letting the first-year students take greater responsibility for their own learning in
an informal collaborative learning environment;
●stimulating critical thinking and helping students clarify ideas through discussion;
●moving students away from considering teachers the sole sources of knowledge
and understanding;
●being in a network of students with whom you feel comfortable studying and
interacting socially;
●getting acquainted with a senior ‘model’ student’s view on successful study
strategies;
●developing study skills like note-taking, problem solving and preparation for
sitting tests.
Since the supplemental instruction programme uses collaborative techniques as a
basis, it also promotes social interaction in an academic environment, which, accord-
ing to Tinto (2010), is a recipe to increase retention. So, at least in theory, the supple-
mental instruction programme would increase both student results and student
J. Malm et al.
656
Downloaded by [Lund University Libraries] at 07:16 05 March 2015
retention. The objective of the present study is to evaluate the programme with respect
to both of these areas, during the first undergraduate year. This is done by addressing
the following research questions:
(1) To what extent does supplemental instruction increase retention during the
first year?
(2) How does supplemental instruction affect student results overall during the
first year in terms of total course credit production?
(3) Does supplemental instruction affect more than the supported course in terms
of student results? If so, how pronounced are these transfer effects?
(4) Are there differences in how supplemental instruction affects students with
different prior academic achievement?
What information on these questions is already available from other investigations
elsewhere? Not that much actually. Most attention has been paid to question one.
Blanc, DeBuhr, and Martin (1983) and Arendale (2001) considered the re-enrolment
rates among students following a semester where supplemental instruction was
provided, and found that attendees re-enrolled at a statistically significant higher
rate compared to non-attendees. Blat et al. (2001) found similar results when
comparing retention rates during promotion from second year to third and third to
final years for supplemental instruction students and non-supplemental instruction
students.
In an investigation related to question two, Bowles, McCoy, and Bates (2008)
found that students participating in supplemental instruction sessions during their first
year increased their chances of a timely graduation four years later. A preliminary
study by Malm, Bryngfors, and Mörner (2010) at the School of Engineering at Lund
University (based on a different and more limited student group than the present
study) indicated that the total course credit production during the first year was
considerably higher among supplemental instruction participants than those who did
not attend sessions. However, the relation between the degree of attendance and first-
year course credit production was not investigated.
Research question three, regarding transfer effects from supplemental instruction
supported courses, was partly addressed by Gattis (2000, 118), who found ‘effects of
fall chemistry supplemental instruction attendance on spring chemistry grade, point-
ing to long-term retention of information discussed in the supplemental instruction
sessions’.
The last research question, on the impact of supplemental instruction on students
with different prior academic achievement, has received attention in a couple of stud-
ies. Ogden et al. (2003) followed students during the two years after attending a
course in political science supported by supplemental instruction. They found that
conditional students (on learning support programmes and/or with English as a
second language entry status) participating in supplemental instruction had signifi-
cantly higher short- and long-term outcomes compared to conditional non-partici-
pants. These results are in agreement with an earlier study conducted by Ramirez
(1997). Here, students in supplemental instruction supported courses from different
faculties were followed for a period of four years beyond these courses. Ramirez
found that the greatest beneficiaries for both the immediate and the longer term were
at-risk students, having significantly greater retention rates than the at-risk non-
supplemental instruction users.
Studies in Higher Education
657
Downloaded by [Lund University Libraries] at 07:16 05 March 2015
The present study could, therefore, provide some valuable new information in the
areas the research questions address. In the study, data from 10 MSc engineering
programmes during the academic year 2009/2010 was used.
The supplemental instruction programme at the School of Engineering,
Lund University
The academic year at the School of Engineering is divided into four quarters (autumn
and spring semesters of two quarters each). Each quarter consists of seven weeks of
scheduled classes and one week of examinations. A full workload for a student is
usually to take two courses each quarter. The supplemental instruction programme is
normally attached to compulsory courses which suffer comparatively high failure
rates during the first two to three quarters of the first year. The aim of the supplemen-
tal instruction programme is to help the new students to adjust to their university stud-
ies and get a good start. It is optional for the student to attend supplemental instruction
sessions.
For the 10 engineering programmes considered in the present study, all but one have
supplemental instruction attached to one common course in the first quarter – Calculus
in One Variable. The remaining programme has supplemental instruction attached to
a course in Physics. In the second quarter six programmes continue with supplemental
instruction attached to the calculus course (Calculus in One Variable is a comparatively
large course and runs over either two or three quarters). Three other programmes use
supplemental instruction in connection with a course in Linear Algebra, while one
programme has a course in Physics supported by supplemental instruction. In the first
quarter of the spring semester the number of programmes that have supplemental
instruction is reduced to eight, four with supplemental instruction still attached to the
calculus course. The other four have supplemental instruction in Linear Algebra,
Introductory Chemistry, General Chemistry and Mechanics courses respectively.
In each quarter, two-hour supplemental instruction sessions are offered once a
week to each student during weeks two to seven (thus the maximum number of
sessions a student can attend is six for each quarter). The sessions are scheduled
during normal school hours (8 a.m. to 5 p.m.), so should not conflict with other ‘out-
of-school’ activities. Therefore every student should be able to attend supplemental
instruction sessions regularly (particularly since there are no tuition fees in Sweden
and state grants to students are generous enough to allow them to study full-time:
students with family responsibilities receive higher grants to allow them to also study
full-time). Based on the attendance from the previous year, 34 supplemental instruc-
tion leaders were employed to allow for an average group size of about 10 students
per session (experience has shown that groups of this size tend to work best).
However, the supplemental instruction leaders were prepared to handle groups of up
to 25–30 students so that attendance peaks and/or increased general attendance could
be dealt with.
Data used in the study
In the study we have used data on the credits taken by each student during the first
year in the engineering programme to compare the student success of supplemental
instruction participants to those not attending. Sweden follows the European Credit
Transfer System (ECTS), which gives 60 credits for a full year of completed studies.
J. Malm et al.
658
Downloaded by [Lund University Libraries] at 07:16 05 March 2015
The engineering programmes at the School of Engineering typically have a study year
comprising eight courses worth on average 7.5 credits each. In most courses credits
are given for activities such as assignments and laboratory work in addition to
examinations. Therefore, the sum credit for the first year used in this study can differ
substantially between students, dependent on how well their studies have gone. In the
study we have also used data on credits taken in the course Calculus in One Variable
in order to isolate the ‘transfer effects’ to other courses. The calculus course is worth
in total 15 credits.
In order to examine whether there are differences in prior academic achievement
between supplemental instruction participants and non-participants, we have used the
grades from Swedish secondary school, both the total grade average and the grade
average for mathematics. For this to be comprehensible some insight into the Swedish
secondary school system is required: generally it spans three years, and is composed
of some 20–25 courses comprising programmes with different orientations (natural
science, economics, humanities, etc.). Each student obtains a grade for each course.
Besides fail, the grades are pass, good and excellent. When applying to be admitted to
university, one does so on the basis of the average grade in all courses (with compen-
sation for different course sizes). Here pass is given the numerical value 10, good is
given 15 and excellent 20. This means that the secondary school average pass grade
is a numerical value somewhere between 10.0 and 20.0. This value for each student
included in this study is used as a measure of his or her previous knowledge. Besides
this full grade average in secondary school, we have also used the average grade in
the five mathematics courses, as a measure of previous knowledge of special impor-
tance for engineering students (applicants to the School of Engineering have to pass
at least five courses in mathematics in contrast to those applying to other faculties of
the university, who do not require mathematics to the same extent).
Results
Supplemental instruction attendance
The supplemental instruction attendance for first-year students from the 10 engineer-
ing programmes, in each of the first three quarters of the academic year, is given in
Table 1. In the first quarter 75% of the students attended at least one supplemental
instruction session. In the second and third quarter this percentage decreased to 58%
and 53% respectively. Likewise the average attendance became successively smaller
by quarter, from 49% in the first quarter to 31% in the second quarter and finally down
to 29% in the third quarter. The average number of attendees at a session was 12
students in the first quarter (34 supplemental instruction leaders), 8 students in the
second quarter (30 leaders) and 8 students in the third quarter (20 leaders). The
percentage of students attending all six supplemental instruction sessions in a quarter
was 21%, 10% and 7% in quarters 1, 2 and 3 respectively.
Credit production during year 1
In Table 2, student data on credits taken during the first year is given as a function of
supplemental instruction attendance. Attendance is divided into four categories: high,
average, low and no attendance. The number of students is distributed quite evenly
between the four attendance groups. The percentage of students not completing the
Studies in Higher Education
659
Downloaded by [Lund University Libraries] at 07:16 05 March 2015
first year differs considerably across the categories, from 22% for the non-attendance
group down to 3% for the high attendance group. Although it is hard to draw any defi-
nite conclusion from this data due to the potential bias of self-selection, it suggests
that attendance at supplemental instruction is good for student retention. Credit data
for the students who completed the first year clearly shows that a high or average
attendance at supplemental instruction sessions leads to considerably higher credit
production during the first year. At the School of Engineering the credit production
per student during the first year is seen as an important indicator of how many will
ultimately take their MSc degree. The goal set by the Board of the Faculty aims at
75% of new undergraduates passing at least two-thirds of the credits given in the first
year of study (Lunds Tekniska Högskola 2007). As can be seen in Table 2, there are
large differences in the percentage of students achieving this goal, from 90% for the
high attendance group down to 43% for the group of students who did not participate
in supplemental instruction sessions. These results clearly suggest a great benefit for
those attending supplemental instruction sessions in regard to the total credit produc-
tion for the first year, apart from improving chances of success in the course to which
supplemental instruction is attached.
However, measures of prior academic achievement given in Table 2, the aver-
age grade in secondary school plus the average grade in mathematics in secondary
school, indicate that the supplemental instruction attendees have a higher aptitude,
which presumably leads to higher credit production. To minimise this effect we
split the students into ‘weak’, ‘average’ or ‘strong’ based on their average mathe-
matics grade from secondary school (this indicator of prior academic achievement
has been shown to be slightly better than the average grade in secondary school for
predicting student results in engineering studies at the School of Engineering: see
Malm 2009). By this procedure we neutralise the effect of differences in mathemat-
ics grades between supplemental instruction attendees and non-attendees (the differ-
ences in mathematics grades between the weak, average and strong groups were 0.1
or less).
In Table 3 the credit production and percentage of students achieving the School
of Engineering goal are given for ‘weak’, ‘average’ and ‘strong’ students as a function
of supplemental instruction attendance. In terms of students having completed the first
year, the credit production is clearly and significantly higher for high and average
supplemental instruction attendees compared to the group of students not attending,
which fully agrees with the results from Table 2. The group that benefits the most is
the group with ‘average’ prior academic achievement, showing a substantial difference
Table 1. Supplemental instruction attendance in quarters 1–3 (of 4) in first-year courses (for
10 engineering programmes) at the School of Engineering at Lund University for the academic
year 2009/2010.
Quarter
No. of students in
courses with
supplemental instruction
Average
attendance
(%)
No. of supplemental instruction sessions
(% of students attending)
012345≥ 6
1 816 49 25 10 8 11 11 14 21
2 796 31 42 12 10 9 9 8 10
3 581 29 47 12 7 11 8 8 7
J. Malm et al.
660
Downloaded by [Lund University Libraries] at 07:16 05 March 2015
of more than 18 credits between the group with high supplemental instruction atten-
dance and the group of students not attending. The group with ‘weak’ prior academic
achievement also gains substantially in terms of credit production by attending supple-
mental instruction sessions. Again the difference is considerable between the two
opposite groups in terms of supplemental instruction attendance, being more than 14
credits. The group with high grades in mathematics from secondary school also bene-
fits from attending supplemental instruction in terms of credits taken in the first year,
though less so, as the group as a whole does well and is relatively close to the full
yearly workload of 60 credits.
The group with low supplemental instruction attendance does not show significant
differences in credit production compared to the group not attending, with the excep-
tion of the group with prior ‘average’ academic achievement, which does significantly
better than the corresponding group of non-attendees. This is probably not unexpected
as it presumably takes a period of time attending supplemental instruction sessions to
reap the benefits in terms of increased general credit production.
These observed differences in credit production become enhanced when compar-
ing how different attendance groups cope with the School of Engineering goal of a
minimum of 40 credits during the first year. For the groups with ‘weak’ and ‘average’
prior academic achievement there are huge differences in the percentage of all first-
year students who pass 40 credits or more, a difference of about 55–60% between the
group with high supplemental instruction attendance and the group of non-attendees.
Moreover, in this case, the difference within the group of ‘strong’ students is not as
pronounced due to a generally high achievement in credit production. But the general
tendency is the same – if you attend supplemental instruction sessions with an average
or high frequency, you will have a better chance of reaching the School of Engineering
goal. A final interesting note is that ‘weak’ students with high supplemental instruction
Table 2. Student data on credits taken during the first year as a function of supplemental
instruction attendance.
Attendance (no. of supplemental instruction
sessions)
None
(0)
Low
(1–5)
Average
(6–10)
High
(≥ 11)
Number of first-year students 165 254 221 198
Average grade in secondary school 17.0 17.4 17.7*** 18.0***
Average grade in mathematics in secondary
school
16.9 17.1 17.4* 17.7***
Percentage of students not completing the first
year
22% 15%* 8%*** 3%***
Average number of credits taken during the first
year for students completing the first year
42.1 45.3 51.9*** 55.6***
Percentage of all first-year students achieving
the School of Engineering goal of at least 40
credits during the first year
43% 57%* 78%*** 90%***
The data is taken from 10 MSc engineering programmes during the academic year 2009/2010. Statistically
significant differences using an independent t-test (two-sided distribution) or chi-square test with p < .05,
p < .01 and p < .001 compared to the group of students not attending supplemental instruction are indicated
with *, ** and ***.
Studies in Higher Education
661
Downloaded by [Lund University Libraries] at 07:16 05 March 2015
attendance do better regarding the School of Engineering goal compared to ‘strong’
students with no attendance, and do almost as well regarding credit production during
the first year.
The above results clearly indicate that, on average, supplemental instruction
attendees benefit greatly in terms of credits taken during the first year, and, further,
that the more students attend supplemental instruction, the more credits they take. But
one may ask how much is due to the likelihood that supplemental instruction attendees
perform better in the course that is supported, and how much is a transferable effect
to other courses, presumably due to better study strategies and skills obtained through
supplemental instruction, and the potential increase in self-esteem and confidence by
participation in supplemental instruction sessions? Such an investigation is not easily
undertaken for all 10 engineering programmes considered here, since most
programmes have supplemental instruction in more than one course. However, four
programmes have supplemental instruction limited to one course only, Calculus in
One Variable, making it much easier to divide the observed increase in first-year
credit production for supplemental instruction attendees into a course-related part and
a transfer effect-related part.
The credits taken in Calculus in One Variable, and totally for the first year, are
given in Table 4. As in Table 2, the number of students is distributed quite evenly
between the four attendance groups. The percentage of students not completing the
Table 3. Student data on credits taken during the first year as a function of supplemental
instruction attendance and mathematics grades from secondary school.
Supplemental
instruction attendance
(no. of sessions)
Average number of
credits taken during the
first year for students
completing the first year
Percentage of all first-year
students achieving the School
of Engineering goal of at least
40 credits during the first year
‘Weak’ students (group with 10.0–15.0 averages in mathematics grades in secondary school)
None (0) 35.8 25% (13 of 53)
Low (1–5) 36.4 36% (26 of 72)
Average (6–10) 43.5** 62% (34 of 55)***
High (≥ 11) 50.0*** 81% (30 of 37)***
‘Average’ students (group with 15.1–18.0 averages in mathematics grades in secondary school)
None (0) 37.2 32% (13 of 41)
Low (1–5) 45.9** 61% (47 of 77)**
Average (6–10) 51.6*** 76% (41 of 54)***
High (≥ 11) 55.4*** 91% (62 of 68)***
‘Strong’ students (group with 18.1–20.0 averages in mathematics grades in secondary school)
None (0) 51.9 66% (39 of 59)
Low (1–5) 52.6 73% (70 of 96)
Average (6–10) 56.5* 88% (96 of 109)**
High (≥ 11) 58.2** 94% (87 of 93)***
Statistically significant differences using an independent t-test (two-sided distribution) or chi-square test
with p < .05, p < .01 and p < .001 compared to the group of students not attending supplemental instruction
are marked with *, ** and ***. (The number of students is slightly less than in Table 2 as not all students
had mathematics grades obtained in a Swedish secondary school.)
J. Malm et al.
662
Downloaded by [Lund University Libraries] at 07:16 05 March 2015
first year again differs considerably, from 24% for the non-attendance group down to
0% for the high-attendance group. The gains in credits taken in the course Calculus in
One Variable (worth in total 15 credits and consisting of either two or three modules)
are considerable if you attend supplemental instruction. For example, the students
with high attendance take on an average 4.8 credits more in the course as compared
to non-attendees. The benefits of being a student with an average or high supplemental
instruction attendance when it comes to total credits taken in the first year is clear, and
compares well to the figures in Table 2.
It is most likely that the results for the student group with high supplemental
instruction attendance are overestimated due to a slightly higher prior academic
achievement for the group (not statistically significant, however), which presumably
leads to better results. However, this effect is marginal considering the limited effect
when ‘neutralising’ the influence of prior academic achievement between Tables 2
and 3. If we then proceed to reduce the total credits taken by the credit gain due to
supplemental instruction in the calculus course, the benefits of being a supplemental
instruction attendee are still substantial (although only the higher rate of attendance
is statistically significant). The relative first-year credit ‘gain’ for supplemental
instruction attendees is also considerable for courses not supported by supplemental
instruction. These findings clearly suggest that the transferable effect to other
courses of improved study strategies and skills, combined with a probable increase
in self-esteem and confidence, leads to a significant overall improvement in student
results.
Table 4. Student data on credits taken in Calculus in One Variable (worth 15 credits) and in
total during the first year as a function of supplemental instruction attendance.
Attendance (no. of supplemental
instruction sessions)
None
(0)
Low
(1-5)
Average
(6-10)
High
(≥ 11)
Number of first-year students 75 109 86 78
Average grade in mathematics in secondary school 16.1 16.0 16.2 16.9
Percentage of students not completing the first
year
24% 16% 6%* 0%***
Average number of credits taken in the course
Calculus in One Variable for students completing
the first year
8.4 8.1 11.4** 13.2***
Difference in credits taken in the course Calculus in
One Variable compared to the group not attending
supplemental instruction
–−0.3 3.0 4.8
Average number of credits taken during the first year
for students completing the first year
41.7 40.8 49.8** 53.7***
Average number of credits taken during the first year
reduced by credit ‘gain’ in the course Calculus in
One Variable due to supplemental instruction
41.7 41.1 46.8 48.9**
The data is taken from the four MSc engineering programmes where supplemental instruction is attached
only to the calculus course. Statistically significant differences using an independent t-test (two-sided
distribution) or chi-square test with p < .05, p < .01 and p < .001 compared to the group of students not
attending supplemental instruction are indicated with *, ** and ***.
Studies in Higher Education
663
Downloaded by [Lund University Libraries] at 07:16 05 March 2015
Discussion
What are the consequences of the above results? First of all, participation in supple-
mental instruction sessions leads to fewer engineering students dropping out during
the first year, which is consistent with previous findings (Arendale 2001; Blanc,
DeBuhr, and Martin 1983; Blat et al. 2001). If we, based on the retention percentages
presented here, compare scenarios with and without a supplemental instruction
programme at the School of Engineering, the number of students dropping out during
the first year is almost halved by using the present supplemental instruction
programme set-up. Furthermore, the supplemental instruction attendees are consider-
ably more effective in taking course credits during the first year, which probably has
a pronounced impact on timely graduation. In the results presented above, frequent
supplemental instruction attendees take one-third more course credits during the first
year than students who do not attend supplemental instruction, which corresponds to
almost a quarter of a year of completed studies. Also, a more successful first year will
most likely further reduce attrition. For supplemental instruction attendees, a large
portion of the higher course credit production can be attributed to transfer effects from
supplemental instruction supported courses to other courses, in terms of better study
strategies and skills, self-esteem and an established network of study partners. These
transfer effects are also likely to benefit supplemental instruction participants in
courses after the first year, making them more efficient in terms of timely graduation,
which, as mentioned, has been suggested in the study by Bowles, McCoy, and Bates
(2008).
In order to determine the exact benefits of supplemental instruction in terms of
how it changes the percentage of students who take their engineering degree, and
how it affects timely graduation, a longitudinal study is required. But it is quite likely
that student results during the first year have a significant influence on these
outcomes. Therefore, the results achieved by using supplemental instruction at the
start of engineering studies at the School of Engineering at Lund University seem
very encouraging.
Prior academic achievement in secondary school does matter when comparing
how many course credits a student takes over the first year of engineering studies at
the School of Engineering. Based on the average grade in mathematics in secondary
school, students with high prior academic achievement completing year one take 92%
of the given course credits during the first year, while students with average or low
prior academic achievement take 81% and 68% respectively. However, students that
did not do as well in secondary school can offset these differences by attending
supplemental instruction sessions. For instance, students with low or average prior
academic achievement, but with a high supplemental instruction attendance, do as
well as students with high prior academic achievement who did not attend supplemen-
tal instruction, in terms of course credit production over the first year. This clearly
indicates the potential of supplemental instruction as a support programme.
Interestingly, however, the supplemental instruction sessions at the School of
Engineering are not seen among students as extra sessions for ‘weaker’ students.
There is actually an over-representation of students with higher aptitudes attending
supplemental instruction. The reason for this we do not know. Perhaps students who
did do well in secondary school better recognise the need for improved study strate-
gies and skills to cope with studies at university level. The consequence, however, is
that we need to get better at informing students of the potential of supplemental
J. Malm et al.
664
Downloaded by [Lund University Libraries] at 07:16 05 March 2015
instruction to affect their study results, so that the representation of students with low
prior academic achievement improves.
Can these results from supplemental instruction be generalised beyond the engi-
neering programmes at Lund University? The answer is yes, definitely, with regard to
the portion of increased course credit production that comes from students performing
better in supplemental instruction supported courses (although the magnitude is likely
to differ from case to case, dependent on how the supplemental instruction programme
is set up, the subject, the education year, etc.) This has been shown repeatedly in liter-
ature independent of course subject. It also seems reasonable to expect that the trans-
fer effects from supplemental instruction supported courses to other courses, as
experienced in the present study, will occur in other educational areas as well.
However, one can expect variation in how great the transfer effects will be, not just
due to differences in how the supplemental instruction programme is set up and differ-
ences between universities and countries, but also due to educational programme
structure and area. Here the degree under consideration consists of compulsory
courses over the first three years. This means that the peer group and network of study
partners obtained in supplemental instruction remains the same over a long period of
time, and that conditions for transfer effects to other courses are optimised. One can
also expect that the main skill improved through supplemental instruction – problem
solving in an engineering/natural science context – is central in most other non-
supplemental instruction supported courses over the first year, which again favours
transfer effects. It may not always be the case, however, that the skills developed
through supplemental instruction in one course are so readily applicable to other
courses.
Consider the costs associated with supplemental instruction in the present case –
have they been justified from an economic standpoint? Costs for 34 supplemental
instruction leaders, venues for sessions, one supervisor and office supplies amounts to
approximately €100,000. Gains with supplemental instruction are primarily based on
a higher credit production (undergraduate and graduate studies at Swedish universities
are partly funded on student performance and expressed in credits produced per year
by each student). In the present case, with 198 students with high supplemental
instruction attendance and 221 students with average attendance producing 13 and 10
more credits respectively than non-supplemental instruction attendees, the total gain
can be estimated to be approximately €300,000. Thus, from a purely economic stand-
point, the investment is sound. For universities funded by tuition fees, the expected
financial rewards from a supplemental instruction programme are likely to be consid-
erably higher. One example from the University of Central Florida published by
Congos (2001), based on tuition fees from students not dropping out due to participa-
tion in the supplemental instruction programme, estimated that the gain obtained
exceeded costs more than tenfold. Thus, supplemental instruction would generally
appear to be a wise investment for universities experiencing retention problems.
In terms of first-year credit production, engineering students with average and
high supplemental instruction attendance do significantly better than students not
attending supplemental instruction. Students with low, average and high prior
academic achievement all benefit from attending supplemental instruction sessions,
but the benefits are more pronounced among the first two categories. The data also
suggests that the transferable effects of study strategies and skills to non-supplemental
instruction courses are substantial for supplemental instruction attendees, leading to
better results in these courses too.
Studies in Higher Education
665
Downloaded by [Lund University Libraries] at 07:16 05 March 2015
References
Arendale, D.R. 2001. Supplemental instruction (SI): Review of research concerning the
effectiveness of SI from the University of Missouri-Kansas City and other institutions
from across the United States. http://www.tc.umn.edu/∼arend011/SIresearchreview01.pdf
(accessed August 28, 2010).
Blanc, R.A., L.E. DeBuhr, and D.C. Martin. 1983. Breaking the attrition cycle: The effects of
supplemental instruction on undergraduate performance and attrition. Journal of Higher
Education 54: 80–90.
Blat, C., S. Myers, K. Nunnally, and P. Tolley. 2001. Successfully applying the supplemental
instruction model to sophomore-level engineering courses. Proceedings of the 2001
American Society for Engineering Education Annual Conference & Exposition, American
Society for Engineering Education.
Bowles, T.J., A.C. McCoy, and S. Bates. 2008. The effect of supplemental instruction on
timely graduation. College Student Journal 43: 853–59.
Bruzell-Nilsson, M., and L. Bryngfors. 1996. Supplemental instruction: Student success in
high-risk courses. Paper presented at the ninth International Conference on the First-Year
Experience, July 15–19, in St Andrews, Scotland.
Congos, D.H. 2001. How supplemental instruction (SI) generates revenue for colleges and
universities. Journal of College Student Retention: Research, Theory, & Practice 3: 301–9.
Gattis, K.W. 2000. Long-term knowledge gains due to supplemental instruction in college
chemistry courses. Journal of Research and Development in Education 33: 118–26.
Hurley M., G. Jacobs, and M. Gilbert. 2006. The basic SI model. In Supplemental instruction:
New visions for empowering student learning. New Directions for Teaching and Learning,
no. 106, ed. M.E. Stone and G. Jacobs, 11–22. San Francisco, CA: Wiley.
Lunds Tekniska Högskola. 2007. Strategisk plan för Lunds Tekniska Högskola, 2007-2011.
Dnr LTH 2007/549 [Strategic plan for the School of Engineering, Lund University, 2007–
2001]. http://www.lth.se/fileadmin/lth/omlth/Strategisk_plan_foer_LTH_2007-2011_final.
pdf (accessed August 28, 2010).
Malm, J. 2009. Har gymnasiebetygen någon inverkan på studieresultaten vid LTH? En analys
av studieresultaten under det första läsåret för den första årskullen i de nya 5-åriga
civilingenjörsutbildningarna [Do high school grades have an influence on student success
at the School of Engineering at Lund University? An analysis of results from the first year
in the new five-year MSc engineering educations]. Lund: KFS I Lund.
Malm, J., L. Bryngfors, and L. Mörner. 2010. Supplemental instruction (SI) at the Faculty of
Engineering (LTH), Lund University, Sweden. An evaluation of the SI program at five
LTH engineering programs, autumn 2008. Australian Journal of Peer Learning 3, no. 1:
38–50.
Martin, D. 2008. Foreword. Australian Journal of Peer Learning 1: 3–5.
Ogden, P., D. Thompson, A. Russell, and C. Simons. 2003. Supplemental instruction: Short-
and long-term impact. Journal of Developmental Education 26: 2–8.
Olstedt, E. 2005. Supplemental instruction, SI – Ett förhållningssätt till lärande. I SI Metod
och teori, 8–14 [Supplemental instruction, SI – An approach to learning]. Center for
Supplemental Instruction, School of Engineering, Lund University. http://www.si-
mentor.lth.se/SI%20Metod/SI-ett%20f%F6rh%E5llningss%E4tt.htm (accessed August 28,
2010).
Ramirez, G.M. 1997. Supplemental instruction: The long-term impact. Journal of Developmental
Education 21: 2–8.
Tinto, V. 2010. From theory to action: Exploring the institutional conditions for student retention.
In Higher education: Handbook of theory and research 25, ed. J.C. Smart, 51–89. New
York: Springer.
J. Malm et al.
666
Downloaded by [Lund University Libraries] at 07:16 05 March 2015
