ArticlePDF Available

The academic value of internships: Benefits across disciplines and student backgrounds

Authors:

Abstract and Figures

While student benefits from internship experience have been frequently documented in research, the emphasis has been on internship effects on employment and career indicators. This work is concerned with effects on academic outcomes and focuses on the robustness of such effects across academic disciplines as well as for different achievement levels of students, student gender, and ethnicity. We present findings from a longitudinal sample (n > 15,000) that covers an extensive range of subjects and disciplines for large Undergraduate cohorts. Main effects and interactions for student background characteristics were investigated showing stable academic benefits for advantaged and disadvantaged students. Further, using ordinal logistic multi-level modelling, we explored the impact on the probability of attaining a higher degree classification for different student scenarios, thus illustrating the practical significance of these internship effects. Effects are less likely to stem from maturation or self-selection. Findings are therefore discussed against a background of motivational approaches suitable to integrate both direct and indirect paths from internship experience to academic outcomes to career indicators.
Content may be subject to copyright.
The Academic Value of Internships
1
Running head: ACADEMIC VALUE OF INTERNSHIPS
The Academic Value of Internships: Benefits Across Disciplines and Student Backgrounds
Jens F. Binder
Thom Baguley
Chris Crook
Felicity Miller
Nottingham Trent University, UK
PRE-PUBLICATION VERSION
PUBLISHED IN CONTEMPORARY EDUCATIONAL PSYCHOLOGY, 41, 73-82 (2015)
doi:10.1016/j.cedpsych.2014.12.001
The Academic Value of Internships
2
Highlights
Effect sizes for student internships on subsequent academic performance are estimated
Estimates control for prior performance, gender and ethnicity and include several
student cohorts
Positive effects are found across a large number of subjects and disciplines
Independent of student characteristics chances for top class degrees are doubled with
an internship
Author Note
Jens F. Binder, Thom Baguley and Chris Crook, Division of Psychology, School of
Social Sciences, Nottingham Trent University, UK; Felicity Miller, Department of Student
Employability and Enterprise, Nottingham Trent University, UK.
This research was funded by a Nottingham Trent University SPUR scholarship
bursary awarded to the authors.
Correspondence concerning this article should be addressed to Jens Binder, Division
of Psychology, Nottingham Trent University, Burton Street, Nottingham, NG1 4BU, UK.
Email: jens.binder@ntu.ac.uk. Phone: +44 115 8482416.
The Academic Value of Internships
3
Abstract
While student benefits from internship experience have been frequently documented in
research, the emphasis has been on internship effects on employment and career indicators.
This work is concerned with effects on academic outcomes and focuses on the robustness of
such effects across academic disciplines as well as for different achievement levels of
students, student gender, and ethnicity. We present findings from a longitudinal sample (n >
15,000) that covers an extensive range of subjects and disciplines for large Undergraduate
cohorts. Main effects and interactions for student background characteristics were
investigated showing stable academic benefits for advantaged and disadvantaged students.
Further, using ordinal logistic multi-level modelling, we explored the impact on the
probability of attaining a higher degree classification for different student scenarios, thus
illustrating the practical significance of these internship effects. Effects are less likely to stem
from maturation or self-selection. Findings are therefore discussed against a background of
motivational approaches suitable to integrate both direct and indirect paths from internship
experience to academic outcomes to career indicators.
Keywords: Internship, academic performance, student achievement, ordinal logistic
regression, placement
The Academic Value of Internships
4
The Academic Value of Internships: Benefits Across Disciplines and Student
Backgrounds
Internships as voluntary, temporary work placements, often undertaken by students at
the university and college levels, have been hailed as win-win situations for both employers
and internees (Coco, 2000). Employers do not have to commit to actual employment, and
internees can further their (future) career. Clearly, internships feature prominently when it
comes to the employability of graduates from higher education (e.g., Gault, Redington, &
Schlager, 2000), and in recent years universities across the Western world have increasingly
acknowledged the importance of career-furthering measures (Bridgstock, 2009; Smith,
McKnight, & Naylor, 2000). Yet, the exact benefits of internships, and how these are brought
about, remain a matter of ongoing debate (Narayan, Olk, & Fukami, 2010). The present work
aims to add to this debate by focusing on the academic value of internships and their direct
effects on study outcomes.
There is little doubt that internships can have a direct and positive effect on a number
of career indicators, at least under the right circumstances (for recent reviews, see Knouse &
Fontenot, 2008; Narayan et al., 2010). Studies specifically relating to business education and
training have shown that, compared to no such experience, internships are associated with
greater perceived attractiveness of job applicants to recruiters (Taylor, 1988), with graduates
obtaining a job more quickly and more easily (Knouse, Tanner, & Harris, 1999), and with
higher salary levels as well as increased job satisfaction (Gault et al., 2000).
In contrast, less emphasis has been put on internship outcomes within higher
education. In a recent synthesis of the existing literature, Narayan et al. (2010) have drawn up
an integrative model of internship effectiveness, which addresses academic preparedness as
an antecedent variable, but omits any academic benefits from student outcomes. This is
supplemented by students’ own perceptions, who have been shown to attribute substantial
social and career-related value to internships, but no academic value that would feed back into
The Academic Value of Internships
5
their studies (Cannon & Arnold, 1998; Cook, Parker, & Pettijohn, 2004). Empirical studies
conducted in educational contexts so far suggest, however, that an internship experience
directly impacts on final grades and degree classes (Gomez, Lush, & Clements, 2004;
Mandilaras, 2004; Mansfield, 2011; Rawlings, White, & Stephens, 2005; Reddy & Moores,
2006).
The first aim of the current work, therefore, is to integrate effects and processes that
relate to higher education with literature on organizational settings and career development
concerns.1 The second aim is to empirically demonstrate such educational effects. In doing so,
our study will go beyond prior research through an in-depth investigation of student
demographics, variation across academic disciplines and controls for self-selection.
For a thorough investigation of the internship-university link, a number of student
demographics are considered that are well known to be associated with study outcomes,
namely gender, ethnicity and scholastic aptitude. Some studies (e.g., Gomez et al., 2004) have
controlled for some of these factors in order to obtain a good estimate of effect size. In the
present work, we are more concerned with interactions between internship experience and
background characteristics in order to estimate internship effects for different student groups.
This can then indicate, for example, whether internships are more or less effective for students
from disadvantaged backgrounds. Another area of inquiry concerns the stability of effects
across different academic disciplines. Studies so far have thrown a spotlight on individual
degree courses (Gomez et al., 2004; Mandilaras, 2004; Mansfield, 2011; Rawlings et al.,
2005; Reddy & Moores, 2006), and most studies concerned with career indicators have been
conducted within business schools (Narayan et al., 2010). These findings, however, cannot
capture the full variation in terms of gender composition, ethnic diversity and scholastic
aptitude to be found within a full-scale university, let alone the variation in standards and
learning climates on different courses. Lastly, we are concerned with the problem of self-
selection. Where research has focused on optional internships, clear a priori differences have
The Academic Value of Internships
6
been documented between students with and without internship experience. As in all field
settings, this weakens any argument that assigns a causal role to internships in raising
academic performance.
In the following, we will elaborate on these points by discussing, firstly, the current
status of internships in higher education and, secondly, the factors that need to be considered
for documenting a general, positive effect of internships on academic outcomes. We will then
go on to present findings from a large, longitudinal student sample, spanning a broad range of
academic subjects to reliably estimate an internship effect while controlling for past academic
achievements and student demographics.
Internships and Higher Education
It is to be expected that internships will increasingly turn into a core interest for the
higher education sector. In recent years, non-academic graduate attributes such as career
management skills have become more attractive for universities to sport, mostly in order to
meet the demands of prospective employers (Bridgstock, 2009). A number of internationally
important university rankings, such as the Financial Times University Ranking in the US and
The Times Good University Guide in the UK, include indicators of postgraduate career
success (Clarke, 2007). All of this increased interest, then, is driven by an employability
agenda. If internships have a direct causal effect on career indicators, universities are well
advised to invest in internships alongside traditional, academic teaching and training.
But what about effects on academic achievement? Anecdotal evidence suggests that
excellent students come with excellent references, including internship experience. Past
research has shown that brighter students, those with better grades, are more likely to get into
an internship (Knouse et al., 1999; Knouse & Fontenot, 2008; Taylor, 1988). We find the
opposite causal direction, however, to be of much higher practical relevance. Improved
academic performance owing to internships could have a substantial indirect effect on
students’ value on the job market, given that study outcomes are routinely treated as central
The Academic Value of Internships
7
predictors of employment (see Smith et al., 2000). Roth and Clarke (1998), in their meta-
analysis, found an overall correlation of .20 between academic performance (grades) and
starting salary, as well as correlations from .20 to .30 between grades and current salary.
Further benefits of increased academic performance include a reduction in stress and
improved adjustment to new life circumstances (Chemers, Hu, & Garcia, 2001). Our
theoretical understanding, then, distinguishes between two causal paths from internships to
career indicators, one direct and one indirect by means of influencing academic performance.2
This helps to further highlight the importance of investigating links between internships and
academic outcomes.
So far, studies that have included information on both internships and study outcomes
have been struggling with resolving the inherent confound between the two variables (Gault
et al., 2000; Knouse et al., 1999; Taylor, 1988), in part due to the level of rigour in the
statistical analyses. Although several studies have hinted at academic benefits to date (Gomez
et al., 2004; Mandilaras, 2004; Mansfield, 2011; Rawlings et al., 2005; Reddy & Moores,
2006), there is little stringent evidence for a causal link between internships and study
outcomes. Most convincingly so far, Gomez and colleagues (2004) found a relationship
between internships in the second year of an undergraduate bioscience degree course and final
marks in the third year while controlling for pre-university qualifications, prior academic
achievements, and gender. On a percentage scale, the net effect of an internship experience
amounted to an increase of 4 percentage points in final marks. With these findings, however,
there is still room for substantial student self-selection since the authors report on a UK
degree system that normally allows students to choose between a degree course with or
without internship, even after they have commenced their studies (Little & Harvey, 2006). A
similar criticism applies to other research in the field. We believe this can be overcome by,
first of all, looking at a range of degree courses and comparing effects not only for
corresponding courses with and without internships, but also for courses that never provide an
The Academic Value of Internships
8
internship option and those where internship is integral to the course (and therefore, in some
sense, compulsory).
Student Background Characteristics and Internship Effects
Closely related to the issue of self-selection, as discussed above, is the question of
student background characteristics: Would we expect internships to be equally effective for
different categories of students? A prominent factor in this context is ethnicity. It is a well-
established finding that in Western, mixed-ethnic societies, most non-white students, and in
particular Blacks, show lower academic achievements (Cohen, Garcia, Apfel, & Master,
2006; Kao & Thompson, 2003; Nora & Cabrera, 1996; Arroyo & Zigler, 1995). Without
going here into any details concerning the underlying reasons for this minority disadvantage,
it is important to note that Whites are also more likely to take up an internship than Blacks
(Knouse et al., 1999), thus potentially furthering the gap. This is particularly troublesome in
those academic disciplines where internships are far from the norm and require more student
initiative in terms of set up.
Another factor that has received considerable attention in the literature on academic
outcomes and graduate careers is gender. While research on gender and academic
achievement was historically concerned with an academic disadvantage for female students
(Rudd, 1984), more recent studies have moved away from a uniformly negative view on
female educational attainment (McNabb, Pal, & Sloane, 2002), with some reporting an overall
reversal of this gender gap (Buchmann & DiPrete, 2006). Females tend to show both higher
study motivation and higher study outcomes (e.g., Harackiewicz, Barron, Tauer, & Elliot,
2002), with the exception of science subjects and related disciplines (Mellanby, Martin, &
O’Doherty, 2010; Steele 1997). Studies on internships have mostly used gender as a control
variable (Gomez et al., 2004; Rawlings et al., 2005), but Mansfield (2011) reported a reduced
benefit from internships for female students as compared to males. This emphasises the
importance of further investigating interactions between internships and gender.
The Academic Value of Internships
9
Finally, the level of academic aptitude needs to be taken into account. As stated
before, brighter students are more likely to gain access to internships (Knouse et al., 1999).
This means that studies on internships run the risk of focusing on elite support while
neglecting lower achievers. As with some other background characteristics, studies
documenting an internship effect on academic outcomes have routinely controlled for prior
achievement. An exploration of internship effects at different levels of academic achievement,
however, is still missing. In sum, we propose that taking into account central student
characteristics, ethnicity, gender, and academic aptitude, both as control variables and
potential moderators of internship effects will further increase the relevance of findings in
current debates in the higher education sector.
General and Stable Internship Effects?
Having discussed the relevance of an internship effect on academic achievement and
the main variables investigated in this study, we briefly address factors that speak for and
against a general effectiveness of internship. At first, the generality of effects may well be
questioned on the grounds of the potentially moderating role of student characteristics
identified so far. On the other hand, small-scale studies so far have found support for
internship effects in disciplines as varied as economics (Mandilaras, 2004), bioscience
(Gomez et al., 2004), surveying (Mansfield, 2011), information systems (Rawlings et al.,
2005), and psychology (Reddy & Moores, 2006). Of course, all these benefits could stem
from general maturation in students (Devlin, 1996). However, given that academic
achievement is multi-factorial (e.g., being influenced by prior knowledge, intelligence, social
support and external pressures), we find it more fruitful to speculate on overarching
motivational processes. Although this work is not directly concerned with student motivation,
the consideration of motivational constructs helps to formulate the expectation that internship
effects will be robust across different academic disciplines and at the same time will show
some variability for different types of students.
The Academic Value of Internships
10
Internship effects on motivational factors in relation to careers are well documented.
Internships have been shown to lead to a higher fit between business students’ instrumental
values and job characteristics (Pedro, 1984), to a greater crystallization of a vocational self-
concept (Brooks, Cornelius, Greenfield, & Joseph, 1995; Taylor, 1988), and to improve the
cushioning of a graduate’s reality shock (Cook et al., 2004; Taylor, 1988). These findings
point to processes independent of specific academic fields and suggest that, overall,
internships may be related to changes in intrinsic motivation (Deci, Vallerand, Pelletier, &
Ryan, 1991) as well as social-cognitive processes leading to increased self-efficacy and
interest (Lent, Brown, & Hackett, 1994) and higher-level career aspirations (Nauta, Epperson,
& Kahn, 1998). Based on these speculations, we expected to find positive internship effects
on academic outcomes, and for these to hold across disciplines. At the same time, research on
perceived barriers (Luzzo & McWhirter, 2001) indicates that such effects may vary across
student background characteristics, with particular reference to ethnic minority status and
being female, due to group-specific anticipation of inhibiting factors on the way to
educational and career goals. A systematic comparison of internship effects across different
sub-groups in the student population, something which has not been undertaken so far, is
therefore essential.
Research aims
In the following, we present findings from a large longitudinal data set on internships
and academic achievements covering the years from 2001 to 2008 for all completed
undergraduate student cohorts at one of the largest universities in the UK. Internships were all
integrated with degree courses and took place during an additional year sandwiched in
between year two and year three of a three years B.A. or B.Sc. degree (hence the commonly
known label sandwich placement in the UK). Internship duration typically falls in between
eight and eleven months, depending on university policy.
The Academic Value of Internships
11
These data enabled us to pursue several research aims that go well beyond what
existing studies could address so far. Firstly, controlling for prior academic achievement, we
wanted to estimate the magnitude, if any, of a general internship effect and to test for the
stability of such an effect across a broad range of academic disciplines. Secondly, we wanted
to compare effects for important subgroups within the student population: for males and
females, for different ethnic backgrounds, and at different levels of academic aptitude.
Thirdly, our aim was to provide a control for self-selection of students. Although it is
impossible in most field settings to rule out self-selection biases completely, our data allowed
for a comparison of voluntary and mandatory internships in various ways. Corresponding
degree courses with and without an internship could be juxtaposed, as well as courses that
never provide an internship option and courses that only exist with an integrated internship.
In a further step, we attempted to maximize the practical relevance of our analyses by
comparing different ways of scaling academic outcomes, namely degree mark and degree
class. Within the English-speaking world, there are a number of approaches to grading and
awarding a particular level of outcome to a student’s degree (Sadler, 2005). In the UK, a
degree mark is often awarded as a grade point average with a value range from 0 to 100.
Specific ranges on this scale represent different degree classes (see Yorke, Barnett, Evanson,
Haines, Jenkins, Knight, et al., 2004, for the particulars in matching class to mark). For both
academic and career purposes, degree class solely carries importance, and in recent years
strong concerns in the UK have been raised regarding mounting pressures on students to
obtain the top-most classes in order to gain access to qualified employment (Burgess Group,
2007). The Higher Education Statistics Agency (HESA) publishes data on student
achievement on the basis of degree class only (HESA, 2012).
Method
Participants
The Academic Value of Internships
12
Data were obtained from the central administration of a UK university with one of the
largest undergraduate populations in the country on all undergraduate student entries spanning
the years from 2001 to 2009. In order to be included in the analyses, all entries had to meet
the following criteria:
(a) They had to refer to a completed undergraduate degree course, either 3 years full-time
study without internship, or 4 years with an internship during the 3rd year of study. According
to university regulations, internships could last from 36 to 52 weeks with an expected typical
duration of 44 weeks. They should not indicate shorter or integrated internships, and they
should take place in a professional setting external to the university. Organisations of all sizes
in the industrial, charitable, public and service sectors were eligible to offer internships.
Internships were facilitated by the university, but the final responsibility of securing an
internship rested with the student. This means that, for each degree course, internships were
not normally supplied by one or few external organisations, but were linked to a large and
varied range of organisations and professions.
(b) They had to allow for consistent mapping of students and internships to a specific course.
This did not hold in a small proportion of cases where students had changed courses during
this period.
(d) In case of internships, only those students who had also completed their internship were
considered. A small number of students were excluded for this reason (n = 65).
The above criteria resulted in a sample of 15,732 complete student entries covering five
cohorts (having commenced studies in 2001-2005). All further information is provided for
this select set of students only. Students were on average 19.4 years old (SD = 3.2) at the start
of their studies. Females were in a slight majority (52.7%), and a large majority (81.5%) were
classed as having a “white” ethnic background. Internships were completed by 4,024 (25.6%)
of all students.
The Academic Value of Internships
13
A summary of student numbers, gender ratio, ethnicity ratio, and internship ratio by subject
area is provided in Table 1.
Measures
In addition to gender (male/female), ethnicity (white/non-white) and internship
(yes/no), the following variables were available and used in the analyses:
Course affiliation. In order to model variation explained at the course, rather than
student, level, course affiliation was included as a categorical variable. Students were
distributed over a total of 186 degree courses spanning almost the full range of higher
education subjects. The Joint Academic Coding System used by the Higher Education
Council for England (HEFCE, 2010) showed entries for 16 out of 19 categories, ranging in
student frequency from 210 to 3,243. The three categories that were not represented were
Medicine and Dentistry (though other subjects related to medicine are represented),
Technologies (which does not cover subjects classed as engineering, which are represented),
and Non-European Languages. Accumulated course sizes over all student cohorts ranged
from 1 to 1,488 with a mean size of 85 (SD = 153.61, Median = 24).
Choice. To indicate whether students could effectively choose between academically
equivalent 3-year and 4-year courses, a dichotomous variable was computed, coded as 0 for
no choice and 1 for choice. Out of the 186 courses, 35 existed in both the 3-year version
without internship and the 4-year version with internship. 44 came with mandatory
internships (i.e., they were not offered in any other way by the university). 107 did not include
an internship and existed only in the 3-year version.
Prior achievement. Prior academic achievement was measured as the average of year
1 and 2 marks, r(8,876) = .67, 95% CI [.66, .68]. This index could potentially range from 0 to
100 (with higher values indicating better achievement).
Final marks. As with prior achievement, final marks could potentially range from 0
to 100 (with higher values indicating better final achievement).
The Academic Value of Internships
14
Degree class. Arguably more important than the final mark is the degree
classification, which was coded as an ordinal variable. In the UK, degree outcomes are
usually grouped into five categories: “Fail/Other” (indicating no honours degree classification
according to the regulations of the awarding body), and “Third class”, “Second, class, lower
division”, “Second, class, upper division” and “First classhonours degrees (the latter
indicating a degree with distinction, commonly achieved with an overall mark of 70% or
higher). The honours degree classifications are usually abbreviated to “Third”, “2:2”, “2:1”
and “First”. It should be noted that regulations for accumulating credit, awarding grades or
marks, and criteria for determining classifications vary considerably between awarding
bodies.
Data-analytical procedures
Multiple regression models of varying complexity were used to predict final marks
and degree class from internship experience while controlling for other student characteristics.
Given the nested data structure, students within degree courses, multilevel modelling (see,
e.g., Snijders & Bosker, 2012) was the method of choice, and particular emphasis was given
to two-level models in which variation at the course level, rather than the student level, is
explicitly modelled and controlled for. All regression models were computed in R (R
Development Core Team, 2012) using the lme4 package (Bates, Maechler, & Bolker, 2012)
for models involving final marks and the ordinal package (Christensen, 2012) for models
involving degree class.
Results
After the preliminary analyses, we will first focus on a comparison between traditional
single-level regression models that disregard course affiliation and two-level multilevel
models that consider course-level variability. We then proceed to investigate the practical
significance of an internship effect by looking at the probabilities for attaining a final degree
The Academic Value of Internships
15
class in different student scenarios. Lastly, the stability of internship effects is examined
across courses that do or do not give choice over internships to students.
Preliminary analyses
The mean for prior achievement was M = 57.1 (SD = 6.7), indicative of a 2:2 class.
Mean marks improved from year 2 to the final year with a mean of 58.6 (SD = 8.3), also
indicative of a 2:2. This difference was statistically significant, M = 1.59, 95% CI [1.41,
1.77]. Overall degree classification showed 1.7% in the Fail/Other class, 7.6% with a Third,
43.6% with a 2:2, 41.3% with a 2:1 and 5.8% with a First.
Bivariate correlations between internship, final marks, prior achievement, gender and
ethnicity are presented in Table 2. As one would expect, a significant relationship between
final marks and prior achievement was found: r(15,730) = .62; 95% CI [.61, .63]. In contrast,
all correlations involving internship were relatively modest, none exceeding a magnitude of
.14.
Comparing models with and without course-level variability
A comparison of single-level models (i.e., disregarding variability at the level of study
courses) and two-level multilevel models (including such variability as a random effect)
showed distinct differences. First, final year mark was regressed on prior achievement,
gender, ethnicity, internship and all two-way interactions involving internship using the
standard multiple regression approach (with prior achievement centred). In this model, all
effects, with the exception of the internship x ethnicity interaction, emerged as statistically
significant. A summary is provided in Table 3. Undertaking an internship led to an average
increase of 2.7 percentage points in final year marks. Interactions with gender and with prior
achievement indicated that this effect was reduced for female students and lower for higher
achieving students. The same set of predictor variables was used in a two-level model that
included course-level variability as a random effect (see Table 3).3 A model comparison
showed that 9.3% of the total variance in final year marks was located at the course level.
The Academic Value of Internships
16
According to a likelihood ratio test course-level effects were highly significant: χ2 = 723.10
(df = 1), p < .001. In contrast to the single-level model, the influence of gender was markedly
reduced (b = 0.38 vs. 1.26). While prior achievement, b = 0.766, 95% CI [0.750, 0.789],
gender, b = 0.381, 95% CI [0.129, 0.678], and ethnicity, b = 1.251, 95% CI [0.922, 1.538],
remained significant predictors with similar effects, there was no interaction between
internship and gender. Undertaking an internship led to an average increase of 3.4 percentage
points in final year marks, b= 3.409, 95% CI [2.725, 3.968], and this effect decreased as prior
achievement increased, b = -0.151, 95% CI [-0.188, -0.117]. In sum, internships had a
positive and reliable effect on final year marks. The multi-level approach showed that this
was independent of gender and held over a large range of academic courses.
A similar pattern emerged when the odds of attaining a particular degree class were
predicted in a series of ordinal logistic regressions. In a single-level model containing prior
achievement, gender, ethnicity, internship and all two-way interactions involving internship,
all predictors were significant with the exception of the ethnicity x internship interaction. A
summary is provided in Table 4. A positive coefficient for internship (b = 0.349) indicated
that undertaking an internship increased the odds of attaining a higher degree classification. In
addition, as indicated by the interactions, this internship effect was reduced for female
students and lower for higher achieving students. Including course level variation as a random
effect in a two-level model again changed the overall picture (see Table 4). Although there is
no established way of partitioning the variance for ordinal multilevel models, a likelihood
ratio test confirmed that course-level effects, again, were highly significant: χ2 = 882.46 (df =
1), p < .001. Compared to the single-level model, the internship effect was substantially
stronger: b = 0.744, 95% CI [0.538, 0.950] vs. 0.349 in the single-level model. As before, this
effect decreased as prior achievement increased, b = -0.029, 95% CI [-0.041, -0.016].
However, in direct contrast to the single level model, the effect was now increased for female
students, b = 0.245, 95% CI [0.080, 0.402], whereas gender itself was no longer significant. In
The Academic Value of Internships
17
addition, prior achievement, b = 0.266, 95% CI [0.258, 0.274], and ethnicity, b = 0.337, 95%
CI [0.232, 0.441], both remained significant predictors in this model. A detailed interpretation
of coefficients will be provided in the next section. Here, it can be stated that internships
substantially increased students’ chances of improving their degree classification, over a large
range of academic courses.
In order to investigate further the striking differences between single- and two-level
models in terms of gender-related effects, we focussed on gender ratio at the course-level in a
number of follow-up analyses. It is a well-known fact that academic subjects differ widely in
gender composition, and the two-level models automatically control for these differences. For
each student in the data set, the corresponding gender ratio at the course level was computed.
Gender ratio had a mean of M = 0.527 (in favour of being female) with a standard deviation
of SD = 0.266, indicating substantial variation in the data. If course-level effects involving
gender are indeed driven by gender ratio, then controlling for ratio in the single-level models
should yield effects that are more in line with those obtained in the two-level models. This is
exactly what we found. Adding gender ratio and the internship x ratio interaction to the
single-level model predicting final year marks changed the gender coefficient from b = 1.256
to b = 0.399 (0.381 in the two-level model) and the gender x internship interaction from b = -
0.916 to b = 0.458 (0.295 in the two-level model). Similarly, for the prediction of degree
class, adding gender ratio and the internship x ratio interaction to the single-level model
changed the gender coefficient from b = 0.311 to b = 0.024 (0.019 in the two-level model) and
the gender x internship interaction from b = -0.172 to b = 0.282 (0.245 in the two-level
model).
Investigating internship effects across student scenarios
Given that interactions between internship and both prior achievement and gender
were obtained, the two-level models were used to chart predictions of final marks and final
degree class for different student scenarios. Internship effects were investigated for male and
The Academic Value of Internships
18
female students separately and, further, for students with average prior achievement and for
students below and above the average (defined here as one standard deviation below and
above the mean). Taking into account the significant role of ethnicity, this illustration is
presented for male, non-white students and for female, white students, the two groups with
lowest and highest academic achievements, respectively. Put differently, internship effects are
shown for an extreme group comparison.
Internship effects on mean final marks across student scenarios are summarised in
Figure 1. Figure 1 displays the mean with two-tiered error bars (Baguley, 2012). The outer
tier (thin lines) depicts a conventional 95% CI for each mean. The inner tier (thick lines) is
adjusted so that error bars that do not overlap are significantly different with approximately
95% confidence. In spite of the interaction effects obtained, the effect of internship is evident
across gender and ethnicity. It is more pronounced for students at below average performance,
but remains statistically significant for students with above average performance as indicated
by the clear lack of overlap for the inner tier error bars. (Note that performance levels were
determined using the distributions of prior academic achievement within male, non-white
students and within female, white students to avoid confounding with any internship effect).
In the case of degree class, similar comparisons of student scenarios were made, but
now the probability distributions with and without internship are presented. Figure 2
summarises these comparisons. Again, performance levels were determined using the
distributions of prior academic achievement within student groups. As with mean final marks,
the internship effect is substantial across all scenarios. While the probabilities of obtaining a
third class degree or of failing are always reduced with an internship, the probabilities of
obtaining an upper second (2:1) or first class degree are always increased. Generally, the
probability of first class degrees was low in these scenarios, which is not surprising given that
in the whole sample, above average performance was defined as one SD above the mean (an
overall mark of 57.08 + 6.66 = 63.74). In other words, our academically able group were
The Academic Value of Internships
19
performing at the low 2:1 level in the first two years of studying. Even so, internships can
increase the odds of attaining a first class degree substantially. As can be seen from Table 5,
the odds increased by a factor of 2 or greater post-internships for most scenarios.
Effects of choice and maturation
We also tested for effects of choice over undertaking an internship. In other words,
differences between courses with mandatory internships, those with optional internships and
those with no internships were investigated. This amounts to a test of self-selection effects
because it includes a direct comparison of 3-year and 4-year courses of the same kind. Adding
choice and the interaction between choice and internship to the two-level model predicting
final marks as specified in Table 3 did not substantially affect the reported effects, and neither
choice nor the corresponding interaction were significant. The same was found for the two-
level model predicting degree class as specified in Table 4. Adding choice and the choice x
internship interaction to this model did not substantially affect the reported effects, and
neither choice nor the corresponding interaction were statistically significant. To sum up, not
only did internships show reliable positive effects on academic outcomes over a range of
courses, these effects were also unaffected by the degree of choice that students had over
taking up or not taking up an internship.
Finally, the data allow for a tentative analysis of the effects of maturation. For this, a
small sub-set of students was identified (n = 186) on a language degree course, located within
the Humanities, who had undertaken a study placement at another university instead of a
work placement.4 Put differently, these students had completed a regular degree course within
four years with one year away from their main study environment. We compared this sub-set
with all students on three-year degree courses within the Humanities, recreating all analyses
for final marks and degree class as outlined in Tables 3 and 4 with a sample size of n = 2652.
For all models, and in contrast to previous internship effects, placement students showed
significantly lower outcomes than non-placed students. Coefficients for study placement were
The Academic Value of Internships
20
b = -3.94 (p < .05) for the two-level model predicting final marks and b = -1.53 (p < .05) for
the two-level model predicting degree class.
Discussion
Internships have been hailed as powerful career boosters, and, indeed, researchers
have generally found positive associations between internships and career-relevant variables.
Our findings show that internships also have a crucial effect on subsequent academic
outcomes. These effects hold controlling for prior academic achievement, and they hold for
both advantaged and disadvantaged students. This work goes beyond previously published
research by investigating the moderating role of student characteristics across a large range of
academic subjects and by addressing questions of student self-selection. In dealing with these
issues on an unprecedented scale and with data-analytical techniques that take adequate
account of the nested data structure, it is hoped that the present study can contribute
substantially to current debates in the higher education sector and to a more thorough
understanding of the benefits that internships provide. In the following, we will discuss our
findings in detail and then address wider issues in relation to our study.
First, the positive internship effects on academic outcomes, obtained here with a
longitudinal data set spanning all courses from a higher education institution with a large and
varied undergraduate student population, were surprisingly robust. Nearly identical patterns
were obtained from using final year mark and degree classification as outcome variables.
While the effect found for final year mark, an average increase of 3.3 percentage points, may
look small, the effect for degree classification shows that this change in marks is mirrored by
substantially higher probabilities for achieving top degree classes, and substantially lower
probabilities for incurring a low degree class. This demonstrates the practical relevance of
going on a placement in terms of immediate academic benefits.
Second, taking course-level variability into account led to markedly different patterns
of findings when compared to single-level models. In particular, internship effects were
The Academic Value of Internships
21
stronger in all two-level models, and the role of gender as a variable changed substantially.
Not only were direct effects for gender reduced in two-level models (to non-significance in
the case of degree classification), the interaction between gender and internship changed from
being significant and negative to non-significant for final year mark, and then to significant
and positive for degree classification. These changes show that course-level variability should
always be taken into account, and they imply a note of caution regarding prior research that
has focused on single degree courses. Effects obtained for a particular degree or subject may
not generalise easily to other degrees or subjects. We can therefore assume, in line with the
two-level models, that being female indicates a higher final year mark although it has no
direct relation to final degree class. Still, for the latter outcome variable, female students seem
to benefit more from undertaking an internship than male students. The most promising
candidate to explain the sensitivity of gender effects to the inclusion of course-level
variability, according to our results, is the variation in gender composition across subjects.
Third, prior achievement had a positive effect on academic outcomes in all models,
and the higher prior achievement, the more internship effects were reduced. Being classed as
a white student also had a positive effect on academic outcomes, regardless of undertaking an
internship. This, together with the effects found for gender, raises the question of who is most
likely to benefit from an internship and whether positive effects will be completely offset for
some students. Explorations of different student scenarios showed that in spite of some
variability, a positive effect for internships was maintained for all combinations of gender,
ethnicity and level of prior academic achievement. Moreover, the odds of achieving a first
class degree were most increased for students performing at an average or below average
level. Correlations between gender, ethnicity and internship experience were generally weak
(< .2), which suggests that the scenarios we explored are realistic and frequently occurred in
the sample. For instance, being male and non-white did not drastically reduce the likelihood
of undertaking an internship. In addition, the relationship between levels of prior achievement
The Academic Value of Internships
22
and undertaking an internship, although significant, was again weak. This finding, although
supportive of previous research (Knouse et al., 1999; Knouse & Fontenot, 2008; Taylor,
1988), indicates that internships were not only undertaken by brighter students, but by
students at all levels of academic achievement.
Fourth, and related to the issue of academic aptitude discussed above, our findings
show that self-selection plays a secondary role at best. Our comparison of degree courses with
mandatory and optional internships showed no differences in terms of an internship effect. Of
course, it is impossible to rule out self-selection effects entirely. Students’ choice to enrol on a
particular course with or without mandatory internship may be influenced by wider career
goals, school grades, levels of confidence and so forth. In the same manner, students may
have to decide prior to enrolment whether they wish to join a 3-year version without or a 4-
year version with internship. Still, whatever the modalities of particular courses, our analyses
demonstrate that across courses and disciplines the benefits of an internship are persistent.
Similar to arguments surrounding self-selection, findings do not suggest that
internship effects are a product of mere maturation. On the contrary, students who spent a
year studying in a different place before completing their degree showed less favourable
outcomes compared to students in the same university segment who did not undertake such a
study placement. This discrepancy could be due partly to different marking standards on a
particular degree course; there is only one course with study placements in the data. At the
same time, the discrepancy could also point towards student difficulties with fitting back in
after they have been exposed to an entirely different set of standards at another university. At
the very least, our analyses raise important questions regarding the facilitation or inhibition of
student progress by means of a sandwich year. They also make an explanation of internship
effects based on maturation less likely.
Some issues remain that must be left to future research. We are well aware that any
wider claim to generality of effects is compromised by the fact that we looked at only one
The Academic Value of Internships
23
university and one basic format for internships. The decentralised management of internships
within the institution and the range of teaching and learning methods, from practice-based to
abstract instruction, to be found across academic subjects give us confidence that future
research will essentially confirm our findings for the comparatively long internships that we
investigated. The question remains, though, whether shorter internships would be equally
effective, and, more generally, what constitutes an optimal duration. If equal benefits,
academic and otherwise, could be reaped from internships that do not require the expansion of
a three-year into a four-year degree course, this would be of immediate relevance for general
study design and the format of courses that higher education institutions want to put on offer.
If, on the other hand, year-long internships prove to be superior, a case can be made for
different types of study support for degree courses with and without internships. Our findings
essentially state that course without internship carry an academic disadvantage. In order to
determine the right kind of study support, more needs to be known about the changes in
students brought about by internships.
Our study was not designed to test hypotheses regarding the underlying psychological
processes that drive internship effects. As we noted in the beginning, motivational factors can
provide a rationale for effects that occur across different academic subjects. In particular, the
formation of overarching career goals feeding back into academic studies and a shift towards
intrinsic motivation are promising candidates. If future research can indeed assign a central
role to these factors, would we then expect the same benefits from fostering goals and
intrinsic motives without internships? Or are the underlying processes such that a non-
academic environment is needed for optimal outcomes? At the same time, we did not find
reduced effects for disadvantaged students, an expectation suggested by the motivatonal
literature on perceived barriers (Luzzo & McWhirter, 2001). It may be that internships
provide immersive experiences that simply outweigh such barriers. Clearly, linking the
The Academic Value of Internships
24
academic outcomes of internships to motivational constructs is a task that still needs to be
done.
Conclusions
We see the academic value of internships as an indirect path to career indicators and as
such, the present work carries some important messages for higher education institutions and
for those advising students on their career plans. 1) Internships typically come with benefits,
and all students across all subject areas are likely to reap these benefits. 2) There is, on the
whole, surprisingly little variation between advantaged and disadvantaged student groups.
Encouraging weaker students to take up an internship is no wasted effort. 3) When evaluating
the effectiveness of internships, institutions are well advised to consider course-level
variability, in particular gender ratio. 4) Ironically, academic benefits may be due to aspects of
the non-academic environment in which internships happen. Mere maturation (e.g., studying
for an additional year in a different academic environment) did not lead to the same positve
outcomes in our sample. 5) Given this pattern of findings, institutions should consider
whether degree courses without internships carry specific disadvantages, and if so, how these
could be addressed through specific study support.
The Academic Value of Internships
25
References
Arroyo, C.G., & Zigler, E. (1995). Racial identity, academic achievement, and the
psychological well-being of economically disadvantaged adolescents. Journal of
Personality and Social Psychology, 69, 903-914. doi: 10.1037/0022-3514.69.5.903
Baguley, T. (2012). Calculating and graphing within-subject confidence intervals for
ANOVA. Behavior Research Methods, 44, 15875. doi:10.3758/s13428-011-0123-7
Bates, D., Maechler, M., & Bolker, B. (2012). lme4: Linear mixed-effects models using S4
classes. Available at: http://CRAN.R-project.org/package=lme4 [accessed 01-04-12]
Bridgstock, R. (2009). The graduate attributes we’ve overlooked: Enhancing graduate
employability through career management skills. Higher Education Research &
Development, 28, 31-44. doi: 10.1080/07294360802444347
Brooks, L., Cornelius, A., Greenfield, E., & Joseph, R. (1995). The relation of career-related
work or internship experiences on the career development of college seniors. Journal of
Vocational Behavior, 46, 332-349. doi: 10.1006/jvbe.1995.1024
Buchmann, C., & DiPrete, T.A. (2006). The Growing Female Advantage in College
Completion: The Role of Family Background and Academic Achievement. American
Sociological Review, 71, 515-541. doi: 10.1177/000312240607100401
Burgess Group (2007). Beyond the honours degree classification: Burgess Group final report.
London: Universities UK.
Cannon, J.A., & Arnold, M.J. (1998). Student expectations of collegiate internship programs
in business: A 10-year update. Journal of Education for Business, 73, 202-205. doi:
10.1080/08832329809601630
Chemers, M.M., Hu, L., & Garcia, B.F. (2001). Academic self-efficacy and first-year college
student performance and adjustment. Journal of Educational Psychology, 93, 55-64.
doi: 10.1037/0022-0663.93.1.55
The Academic Value of Internships
26
Christensen, R.H.B. (2012). ordinal: Regression Models for Ordinal Data. Available at:
http://www.cran.r-project.org/package=ordinal/ [accessed 01-04-12]
Clarke, M. (2007). The impact of higher education rankings on student access, choice, and
opportunity. In Institute for Higher Education (Ed.), College and university ranking
systems: Global perspectives and American challenges. Washington, DC: Institute for
Higher Education Policy.
Coco, M. (2000). Internships: A try before you buy arrangement. SAM Advanced
Management Journal, 65, 41-47.
Cohen, G.L., Garcia, J., Apfel, N., & Master, A. (2006). Reducing the racial achievement gap:
A social-psychological intervention. Science, 313, 1307-1310. doi:
10.1126/science.1128317
Cook, S.J., Parker, R.S., & Pettijohn, C.E. (2004). The perceptions of interns: A longitudinal
case study. Journal of Education for Business, 79, 179-185.
Deci, E. L., Vallerand, R. J., Pelletier, L. G., & Ryan, R. M. (1991). Motivation and
education: The self-determination perspective. Educational psychologist, 26, 325-346.
Devlin, M. (1996). Older and wiser?: A comparison of the learning and study strategies of
mature age and younger teacher education students. Higher Education Research and
Development, 15, 51-60. doi: 10.1080/0729436960150104
Gault, J., Redington, J., & Schlager, T. (2000). Undergraduate business internships and career
success: Are they related? Journal of Marketing Education, 22, 45-53. doi:
10.1177/0273475300221006
Gomez, S., Lush, D., & Clements, M. (2004). Work placements enhance the academic
performance of bioscience undergraduates. Journal of Vocational Education and
Training, 56, 373-385. doi: 10.1080/13636820400200260
The Academic Value of Internships
27
Hadfield, J. D. (2010). MCMC Methods for Multi-Response Generalized Linear Mixed
Models: The MCMCglmm R Package. Journal of Statistical Software, 33, 1-22.
Available at: http://www.jstatsoft.org/v33/i02/ [accessed 03-11-13]
Harackiewicz, J.M., Barron, K.E., Tauer, J.M., & Elliot, A.J. (2002). Predicting success in
college: A longitudinal study of achievement goals and ability measures as predictors of
interest and performance from freshman year through graduation. Journal of
Educational Psychology, 94, 562-575. doi: 10.1037/0022-0663.94.3.562
HESA [Higher Education Statistics Agency] (2012). Statistics Students and qualifiers at UK
HE institutions. Available at: http://www.hesa.ac.uk/content/view/1897/239/ [accessed
01-04-12]
Kao, G., & Thompson, J.S. (2003). Racial and ethnic stratification in educational achievement
and attainment. Annual Review of Sociology, 29, 417-442. doi:
10.1146/annurev.soc.29.010202.100019
Knouse, S.B., & Fontenot, G. (2008). Benefits of the business college internship: A research
review. Journal of Employment Counseling, 45, 61-66. doi: 10.1002/j.2161-
1920.2008.tb00045.x
Knouse, S.B., Tanner, J.T., & Harris, E.W. (1999). The Relation of College Internships,
College Performance, and Subsequent Job Opportunity. Journal of Employment
Counseling, 36, 35-43. doi: 10.1002/j.2161-1920.1999.tb01007.x
Lent, R. W., Brown, S. D., & Hackett, G. (1994). Toward a unifying social cognitive theory
of career and academic interest, choice, and performance. Journal of Vocational
Behavior, 45, 79-122. doi: 10.1006/jvbe.1994.1027
Little, B., & Harvey, L. (2006). Learning through work placements and beyond. Sheffield,
UK: Centre for Research and Evaluation, Sheffield-Hallam University.
The Academic Value of Internships
28
Luzzo, D. A., & McWhirter, E. H. (2001). Sex and ethnic differences in the perception of
educational and careerrelated barriers and levels of coping efficacy. Journal of
Counseling & Development, 79(1), 61-67. doi: 10.1002/j.1556-6676.2001.tb01944.x
Mandilaras, A. (2004). Industrial placement and degree performance: Evidence from a British
higher institution. International Review of Economics Education, 3, 39-51.
Mansfield, R. (2011). The effect of placement experience upon final-year results for
surveying degree programmes. Studies in Higher Education, 36, 939-952. doi:
10.1080/03075079.2010.486073
McNabb, R., Pal, S. & Sloane, P. (2002). Gender differences in educational attainment: the
case of university students in England and Wales. Economica, 69, 481503. doi:
10.1111/1468-0335.00295
Mellanby, J., Martin, M., & O’Doherty, J. (2010). The ‘gender gap’ in final examination
results at Oxford University. British Journal of Psychology, 91, 377-390. doi:
10.1348/000712600161880
Nauta, M. M., Epperson, D. L., & Kahn, J. H. (1998). A multiple-groups analysis of
predictors of higher level career aspirations among women in mathematics, science, and
engineering majors. Journal of Counseling Psychology, 45, 483-496. doi:
10.1037/0022-0167.45.4.483
Narayanan, V.K., Olk, P.M., & Fukami, C.V. (2010). Determinants of internship
effectiveness: An exploratory model. Academy of Management Learning & Education,
9, 61-80.
Nora, A., & Cabrera, A.F. (1996). The role of perceptions of prejudice and discrimination on
the adjustment of minority students to college. The Journal of Higher Education, 67,
119-148.
Pedro, J.D. (1984). Induction into the workplace: The impact of internships. Journal of
Vocational Behavior, 25, 80-95.
The Academic Value of Internships
29
R Development Core Team (2012). R: A language and environment for statistical computing.
Vienna: R Foundation for Statistical Computing.
Rawlings, P., White, P., & Stephens, R. (2005). Practice-based learning in information
systems: The advantages for students. Journal of Information Systems Education, 16,
455-465.
Reddy, P., & Moores, E. (2006). Measuring the benefits of a psychology placement year.
Assessment & Evaluation in Higher Education, 31, 551-567. doi:
10.1080/02602930600679555
Roth, P.L., & Clarke, R.L. (1998). Meta-analyzing the relation between grades and salary.
Journal of Vocational Behavior, 53, 386-400. doi: 10.1006/jvbe.1997.1621
Rudd, E. (1984). A comparison between the results achieved by women and men studying for
first degrees in British universities. Studies in Higher Education, 9, 4757.
Sadler, D.R. (2005). Interpretations of criteria-based assessment and grading in higher
education. Assessment & Evaluation in Higher Education, 30, 175-194. doi:
10.1080/0260293042000264262
Smith, J., McKnight, A., & Naylor, R. (2000). Graduate employability: Policy and
performance in higher education in the UK. The Economic Journal, 110, 382-411. doi:
10.1111/1468-0297.00546
Snijders, T.A., & Bosker, R.J. (2012). Multilevel analysis: An introduction to basic and
advanced multilevel modeling (2nd ed.). London: Sage.
Steele, C.M. (1997). A Threat in the air: How stereotypes shape intellectual identity and
performance. American Psychologist, 52, 613-629. doi: 10.1037/0003-066X.52.6.613
Taylor, M.S. (1988). Effects of college internships on individual participants. Journal of
Applied Psychology, 73, 393-401. doi: 10.1037/0021-9010.73.3.393
Yorke, M., Barnett, G., Evanson, P., Haines, C., Jenkins, D.Knight, P., et al. (2004). Some
effects of the award algorithm on honours degree classifications in the UK higher
The Academic Value of Internships
30
education. Assessment & Evaluation in Higher Education, 29, 401-413. doi:
10.1080/02602930310001689000
The Academic Value of Internships
31
Footnotes
1 In light of the relevant literature, there is a tendency toward the term ‘internship’ in the US,
in contrast to, for example, the UK and Australia where ‘placement’ is more commonly used.
Further, internship is more accepted than placement in the areas of organizational behaviour
and management. For our purposes, we will adopt the term internship throughout this work.
2 Further, as already stated, academic performance also facilitates access to internships, but in
the interest of clarity, we are not concerned with bi-directional causality at this stage. Later
analyses include a crucial comparison of compulsory versus voluntary internships and thus
provide a check on the assumptions implicit in this simplified model.
3Confidence intervals for the two-level modelling of final marks are HPD derived confidence
intervals obtained with Markov chain Monte Carlo methods using MCMCglmm (Hadfield,
2010).
4These 186 students, not being on regular internships as defined here, were not included
among the internship students in the main analyses.
The Academic Value of Internships
32
Table 1
Summary of Student Numbers, Gender Ratio, Ethnicity Ratio, and Internship Ratio by Subject Area
Subject area
female
non-white
internship
degree format
FT only
SW only
FT/SW
Creative arts & design
.54
.18
.30
.46
.12
.34
Business & administrative studies
.46
.18
.20
.66
.06
.25
Social studies
.57
.17
.22
.60
.09
.27
Biological sciences
.49
.18
.27
.52
.12
.32
Law
.61
.21
.23
.60
.10
.26
Education
.46
.22
.24
.56
.11
.27
Architecture, building & planning
.63
.17
.29
.52
.12
.32
Mass communication & documentation
.51
.20
.17
.68
.08
.20
Mathematical & Computer sciences
.54
.17
.23
.63
.08
.26
Combined
.55
.17
.23
.58
.07
.29
Physical sciences
.49
.28
.54
.14
.29
.47
Linguistics, classics & related subjects
.52
.19
.27
.48
.11
.33
Historical and philosophical studies
.48
.19
.26
.54
.09
.32
Engineering
.59
.15
.16
.78
.07
.07
Veterinary sciences, agriculture & related subjects
.41
.22
.44
.24
.18
.50
Subjects allied to medicine
.52
.17
.34
.46
.13
.38
European languages, literature & related subjects
.43
.16
.22
.62
.09
.25
Note. All figures other than ns represent proportions. Subject areas are defined by the Joint Academic Coding System common to the UK. FT: full-
time. SW: sandwich/with internship. Degree format excludes students not classifiable due to degree changes while studying.
The Academic Value of Internships
33
Table 2
Bivariate Correlations
1
2
3
4
5
1 Final mark
-
.62
.12
.13
.18
2 Prior achievement
-
.10
.11
.19
3 Internship
-
-.14
-.01
4 Gender
-
-.02*
5 Ethnicity
-
Note. Unless indicated otherwise, all coefficients are significant at p < .01. *: p < .05. : n.s.
The Academic Value of Internships
34
Table 3
Summary of Regression Models Predicting Final Year Marks
Single-level model
(R2 = .380***)
Two-level model
b
SE(b)
95% CI
b
SE(b)
95% CI
Intercept
56.192***
0.163
55.872, 56.512
55.987***
0.252
55.505, 56.473
Prior
Achievement
0.758***
0.010
0.739, 0.777
0.766***
0.010
0.750, 0.789
Gender
1.256***
0.124
1.013, 1.499
0.381**
0.137
0.129, 0.678
Ethnicity
1.619***
0.161
1.304, 1.934
1.251***
0.162
0.922, 1.538
Internship
2.681***
0.309
2.076, 3.287
3.409***
0.322
2.725, 3.968
I x PA
-0.168***
0.018
-0.203, -0.132
-0.151***
0.018
-0.188, -0.117
I x G
-0.916***
0.247
-1.401, -0.431
0.295
0.263
-0.259, 0.780
I x E
-0.183
0.314
-0.798, 0.433
0.204
0.310
-0.326, 0.839
Note. The two-level model treats students as nested within degree courses. I: Internship. PA:
Prior Achievement. G: Gender. E: Ethnicity. ***: p < .001. **: p < .01.
The Academic Value of Internships
35
Table 4
Summary of Ordinal Logistic Regression Models Predicting Degree Classification
Single-level model
Two-level model
b
SE(b)
95% CI
b
SE(b)
95% CI
Thresholdfail|third
-4.676***
0.081
-4.835, -4.517
-4.895***
0.106
-5.102, -4.688
Thresholdthird|2.2
-2.610***
0.057
-2.722, -2.498
-2.735***
0.088
-2.908, -2.563
Threshold2.2|2.1
0.772***
0.053
0.668, 0.876
0.863***
0.085
0.696, 1.031
Threshold2.1|first
4.445***
0.066
4.316, 4.574
4.703***
0.095
4.516, 4.890
Prior
Achievement
0.252***
0.004
0.245, 0.259
0.266***
0.004
0.258, 0.274
Gender
0.311***
0.039
0.236, 0.387
0.019
0.031
-0.043, 0.080
Ethnicity
0.463***
0.050
0.365, 0.561
0.337***
0.053
0.232, 0.441
Internship
0.349***
0.095
0.163, 0.535
0.744***
0.105
0.538, 0.950
I x PA
-0.033***
0.006
-0.045, -0.022
-0.029***
0.006
-0.041, -0.016
I x G
-0.172*
0.076
-0.322, -0.022
0.245**
0.080
0.089, 0.402
I x E
0.025
0.097
-0.165, 0.214
0.158
0.102
-0.042, 0.358
Note. The two-level model treats students as nested within degree courses. I: Internship. PA:
Prior Achievement. G: Gender. E: Ethnicity. ‘fail’, ‘third’, ‘2.2’, ‘2.1’, and ‘first’ refer to UK
degree classifications. ***: p < .001. **: p < .01.
The Academic Value of Internships
36
Table 5
Summary of Change in Odds for Attaining a First Class Degree Depending on Internship
Experience and Prior Achievement (Below Average, Average or Above Average)
Male, non-white
Female, white
Internship
Δ odds
Internship
Δ odds
No
Yes
No
Yes
Below
average
.001
.002
× 2.8
.004
.014
× 3.6
Average
.004
.009
× 2.3
.018
.054
× 2.9
Above
average
.023
.043
× 1.9
.083
.187
× 2.3
Note. Figures under internship refer to the probability of attaining a first class degree. Below
average is defined as 1 SD below group mean (males or females) and above average as 1 SD
above group mean for prior achievement.
The Academic Value of Internships
37
Figure Captions
Figure 1. Internship Effects on Final Year Marks for Select Student Scenarios. Outer Tier
Error Bars (Thin Lines) Depict Conventional 95% CIs. Inner Tier Error Bars (Thick Lines)
Depict Adjusted CIs With Non-Overlap Indicating Statistically Significant Differences.
Figure 2. Internship Effects on the Predicted Probability of Attaining a Particular Degree
Classification for Select Student Scenarios. Solid Lines (to the Right) Indicate Internship
Experience, Broken Lines (to the Left) Indicate No Internship.
The Academic Value of Internships
38
Figure 1
The Academic Value of Internships
39
Figure 2
... In general, a student can develop various skills in a typical internship experience, including technical, analytical, leadership, managerial, organizational, ethical, creativity, adaptability, and professional [11]- [18]. In return, internships can positively impact future job satisfaction [19], [20], faster job securing [19], [21], [22], higher compensation [19], [20], [23], and more job recruiters' attention [19], [24], and in general successful future employment [25], [26]. Internships and college performance have an undeniable relation to each other. ...
... In general, a student can develop various skills in a typical internship experience, including technical, analytical, leadership, managerial, organizational, ethical, creativity, adaptability, and professional [11]- [18]. In return, internships can positively impact future job satisfaction [19], [20], faster job securing [19], [21], [22], higher compensation [19], [20], [23], and more job recruiters' attention [19], [24], and in general successful future employment [25], [26]. Internships and college performance have an undeniable relation to each other. ...
... In general, a student can develop various skills in a typical internship experience, including technical, analytical, leadership, managerial, organizational, ethical, creativity, adaptability, and professional [11]- [18]. In return, internships can positively impact future job satisfaction [19], [20], faster job securing [19], [21], [22], higher compensation [19], [20], [23], and more job recruiters' attention [19], [24], and in general successful future employment [25], [26]. Internships and college performance have an undeniable relation to each other. ...
... International Journal of Sciences: Basic and Applied Research (IJSBAR) (2020)Volume 54,No 2, ...
... . In item No.(6), weight mean equals "81.73%" while p-value equals "0.000" which is less than 0.05, International Journal of Sciences: Basic and Applied Research (IJSBAR) (2020)Volume 54,No 2, ...
... In item No. (4), weight mean equals "72.38%" while p-value equals "0.000" which is less than 0.05, which means that respondents (Improved their GPA).6.3 Academic (learning) aspectsInternational Journal of Sciences: Basic and Applied Research (IJSBAR) (2020)Volume 54,No 2, ...
Article
Full-text available
The research which titled (Impact of internships on students personal, Interpersonal, Academic, occupational and civic characteristics in Turkish Academic Institutions) aims to identifying the impact of Impact of internships on students personal, Interpersonal, Academic, occupational and civic characteristics in Turkish Academic Institutions It aims at providing quantitative data that shows that joining an internship program could affect students in personal, interpersonal, academic, occupational and civic areas This research adapted to prove by evidence how would internships affect the students, and show the extent of the impact and answering by numbers how does internship affect a student in details in every aspect. The researcher adapted a descriptive analytical approach which depends on data collection, analysis using SPSS and interpretation of the results to determine the variables mentioned. An accepted measurement tools were adapted, and modified to suit the purpose of the study. The results also proved that Participation in a student's internship have an overall impact on the five variables mentioned. The research has presented some recommendations concerning applying learning program which could be more effective if it has been taken into consideration.
... Peers and group composition have been noted to impact interest in STEM subjects, particularly in adolescents (Turner et al. 2019). STEM-and non-STEM-based internships have also been noted to have positively benefit white female students more than their white male counterparts (Binder et al. 2015). Understanding the role of internships on minority participants warrants further investigation. ...
Article
Full-text available
While Science, Technology, Engineering, and Mathematics (STEM) fields are growing, there are persistent issues in sustaining interest and engagement. Internship programs and partnerships are increasingly being deployed to combat these issues by providing students relevant experience. However, there are considerable data gaps in understanding how effective these programs are, and what factors contribute to greater engagement. To this end, this study utilizes a pre-post survey method (n = 50) to understand how the Green Teams internship program at Montclair State University affects student interest and engagement in STEM and sustainability. Data were used in a cluster analysis and then in stepwise regressions to assess changing opinions in STEM engagement, interest in the pursuit of a higher degree or career in STEM fields, and sustainability. We found that age, class year, and parent careers in STEM were significant in predicting positive changes in various attitudes toward STEM. While diversity within the programs did not prove to be significant in our analysis, we did find that more diverse groups with more STEM students generally had higher engagement with the program. These findings can be useful in informing other such programs to increase student engagement in STEM disciplines in both academic and professional settings.
... For corporations, internships lead to better-skilled graduates and reduce hiring practice funds and lower worker turnover. According to Coco (2000) and Binder, Baguley, Crook, and Miller (2015), engaging in internships, which are often offered by universities and taken by college students as volunteers and temporary works, were considered profitable situations for both employers and interns. ...
Thesis
Full-text available
Because of Yemeni’s high rate of unemployed graduates and the difficulty of graduates' transmission into the job market, this dissertation explores the role that internship programs play in setting for graduating youth and offering them employability skills, so graduates can find worthwhile employment. Graduates face many difficulties after graduation to enroll in the workforce. Recent graduates are lacking knowledge, practical skills, and experience which are mandatory requirements demanded by employers for recruitment. This thesis tries to explore the enhancements of graduates' employability caused by participating in internships and identify the required skills that are demanded by employers. Through a quantitative approach, the questionnaire used in this study consists of closed-ended questions that have been directed to both populations of the study. The first population was graduates from two Yemeni universities, Lebanese International University (LIU) and International University of Technology Twintech (IUTT), who participated in internship programs, whereas the second population was employers who provide internship programs for graduates. The findings highlighted the usefulness of internship programs for enhancing employability skills, maximizing employment, and developing graduates. The findings also showed that internship programs are crucial for recent graduates. In addition, as perceived by graduates and employers, the findings showed the most important skills that graduates should have to secure employment in the competitive labor market. Moreover, the researcher provides recommendations for stakeholders and future research.
... Marneros et al. [26] studied the skills and knowledge required by hospitality management graduates and the content of the courses that students should take to be prepared for a restaurant positions; those authors found that graduates should have catering management skills, financial management skills, problem-solving abilities, creativity and flexible thinking. Askren and James [1] and Binder et al. [27] specifically mentioned the contradiction between the rapid development of technology and business models in recent years and the excessively long education cycle in schools. When schools lack the capacity to internalize and organize knowledge into curricula, it is likely that the courses they teach are outdated and do not meet the needs of the industry. ...
Article
Full-text available
In recent years, with the flourishing of the catering economy and the trends in computer technology, restaurant operators have increasingly relied on employees with computational and information skills. Breaking through the traditional teaching method of mere lecturing, the study conducts a teaching demonstration by integrating the computational thinking concept and a Microsoft Excel computational system on the school’s E-learning platform into the teaching of a Culinary and Restaurant Management course. A non-equivalent control group pretest–posttest study with a quasi-experimental design is adopted for the assignment of experimental participants and the design of the course. The results show that a curriculum design with computational thinking significantly improves the effectiveness of students’ learning in digital technology and is especially helpful for the cultivation of the key capabilities of menu design and cost planning among restaurant management skills. The study makes the following contributions: during the Culinary and Restaurant Management course, the use of the E-learning platform and computing programs such as Microsoft Excel is associated with greater learning effectiveness than traditional teaching methods. The research results can serve as a reference for promoting an E-catering business model and a sustainable educational model in the future.
... • Experiences in England show that those corporations which build a close relationship with students, do not have recruitment problems. (Binder et al 2015) This relationship means professional opportunities -e.g. apprenticeship -and programs. ...
Article
Full-text available
There is a structural transformation going on, with effects on our everyday life, due to the rapid technological changes we experience nowadays. This transformation calls for fast and effective answers from the economic actors and leaders. Taylorism brought new methods in the organization of work, but now new ideas are needed in management. This paper analyzes the factors that have been forming the economic sector for the last ten years in Europe, and the reactions these changes have generated. We refer to both the international and the Hungarian context. The conclusion of our theoretical work is that a new, modern economic policy is the key to the higher competitiveness of enterprises.
... Obviously, students amass a wealth of work experience from an internship (Okolie et al., 2021;Binder et al., 2015). Another angle is employers' benefits resulting from employing interns who are young fresh and of the first choice to work with (Kysor and Pierce, 2000;Mamaleka, 2020). ...
Article
Full-text available
Purpose As the 4th industrial revolution (4IR) unfolds, there is an increasing awareness that its implications for workforce transformation and shifts in workforce demand will profoundly impact the future of work. Specifically, the paper seeks to answer the following research questions: i) how does Students’ Industrial Work Experience Scheme (SIWES) equip young people for the real world of work, especially in the era of the third industrial revolution?; ii) does SIWES support the exposure of young people to the world of digitalization?; and iii) what are the effects of the SIWES exposure on the employability of young people? This paper aims to evaluate the University Internship system and preparation of young people for the world of work in the 4th industrial revolution. Design/methodology/approach This paper used a mixed method to unravel the objectives of this study, that is, quantitative and qualitative methods. For the former, structured questionnaires were used to elicit a response from 249 young people drawn from tertiary institutions across Lagos State, Nigeria. The latter used an in-depth interview method conducted among 45 respondents (25 employers of labor and 20 lecturers). Findings The findings reveal that: SIWES contributes meaningfully to the advancement of knowledge and capacity building among young people; SIWES exposes young people to the world of digitalization, depending on the organization where the internship takes place; and SIWES pays little attention to financial rewards and more attention to the acquisition of skills that are relevant to the world of work. The practical and policy implications of the findings are critically discussed. Originality/value This paper critically evaluates the SIWES policy amidst the growing threats of widening skills gap, greater inequality and broader polarization.
Article
Las prácticas consisten en trabajos temporales que brindan experiencia en tiempo real a los estudiantes. Estas prácticas son especialmente valoradas no solo por los estudiantes, sino también por el profesorado y las empresas. A pesar de la importancia y el crecimiento de los programas de prácticas profesionales en la educación superior, no se ha estudiado suficientemente la influencia de la Inteligencia Emocional (IE) en estos programas. Este estudio pretende abordar esta carencia por medio del análisis de la relación del efecto de la IE con en el periodo de prácticas, la empleabilidad y la satisfacción de los estudiantes. Para este propósito se diseñó un cuestionario que fue enviado de modo online a estudiantes de dos universidades españolas. Para ello, se aplicó un modelo de ecuaciones estructurales a una muestra de 240 estudiantes. Los resultados indican que la IE influye directamente en las prácticas de los estudiantes universitarios. Los resultados también nos permiten afirmar que la IE tiene una influencia indirecta en la mejora de la empleabilidad y en el nivel de satisfacción de los estudiantes con las prácticas. Estos hallazgos pueden ayudar a universidades, profesorado, facultades y a las empresas anfitrionas a mejorar el diseño de los programas de prácticas en la educación superior.
Article
The development of international cooperation along with the growing number of the participants of academic mobility are among the attributes of internationalization of education. Its positive effects are universally acknowledged, however take effort to be achieved. Based on the example of national research universities, the article examines the peculiarities of international activity and opportunities for the Russian universities to participate in the international academic mobility. Primary focus is given to the student international internships, which are considered as highly effective educational and mentoring practices. The theoretical framework for this research is comprised of a set of provisions that characterize international academic exchange as mutually beneficial interaction of different values, cultures, experience, knowledge, interests, and goals. The scientific novelty lies in introduction of the new empirical self-examination reports of the universities, use of various statistical data and rating results. The conclusion is made that international academic mobility has its own structure, pronounced geographical focus, and industry characteristics. Cooperation with individual foreign universities and enterprises, as well as with international university associations and branch associations is well developed and creates favorable environment for the advancement of international academic mobility. However, these opportunities are not used to the fullest, as testified by relatively low number of the Russian participants, prevalence of incoming flow vs outgoing, and unequal conditions for the capital and regional universities.
Article
Full-text available
This study examined the impact of academic resilience and attitude on management performance. Managerial performance is the outcome variable, whereasacademic resilienceand manager attitude are independent variables. Data was collected through a standardized survey questionnaire with items relevant to all dimensions and demographical characteristics of 120 managers of different enterprises from Peshawar. SPSS 24 is used for data analysis; Pearson'scorrelation and regression. The outcomes indicated that academic resilience and attitude positively and significantly affect managerial performance.
Article
Full-text available
The authors examined the role of achievement goals, ability, and high school performance in predicting academic success over students' college careers. First, the authors examined which variables predicted students' interest and performance in an introductory psychology course taken their first semester in college. Then, the authors followed students until they graduated to examine continued interest in psychology and performance in subsequent classes. Achievement goals, ability measures, and prior high school performance each contributed unique variance in predicting initial and long-term outcomes, but these predictors were linked to different educational outcomes. Mastery goals predicted continued interest, whereas performance-approach goals predicted performance. Ability measures and prior high school performance predicted academic performance but not interest. The findings support a multiple goals perspective.
Article
Full-text available
Despite the growing popularity of internships, surprisingly little research has investigated causes of their effectiveness. We combine the findings from these studies with insights from the personnel and knowledge transfer literatures to identify the different roles of three actors-students, university, and business-and to propose a multistage model of determinants of effectiveness. Exploratory analysis of a portion of the model on Portuguese internships data reveals the importance of considering the respective roles of the multiple actors and of the internship process in explaining student satisfaction, but not project implementation. Using our conceptual model and these initial empirical findings, we offer recommendations for actions each actor can take to enhance internship effectiveness and lead to suggestions for researchers interested in identifying determinants of internship effectiveness.
Article
Full-text available
This longitudinal study examined the role of perceptions of prejudice-discrimination on collegiate experiences and on college-related outcomes among minority and nonminority students at a public, predominately white, commuter institution. Results indicated that minorities were more prone to feel discrimination and prejudice while on campus than were whites and that these perceptions were found to affect minority students' adjustments to college and college-related outcomes.
Article
Full-text available
Maximum likelihood or restricted maximum likelihood (REML) estimates of the parameters in linear mixed-effects models can be determined using the lmer function in the lme4 package for R. As for most model-fitting functions in R, the model is described in an lmer call by a formula, in this case including both fixed- and random-effects terms. The formula and data together determine a numerical representation of the model from which the profiled deviance or the profiled REML criterion can be evaluated as a function of some of the model parameters. The appropriate criterion is optimized, using one of the constrained optimization functions in R, to provide the parameter estimates. We describe the structure of the model, the steps in evaluating the profiled deviance or REML criterion, and the structure of classes or types that represents such a model. Sufficient detail is included to allow specialization of these structures by users who wish to write functions to fit specialized linear mixed models, such as models incorporating pedigrees or smoothing splines, that are not easily expressible in the formula language used by lmer.
Article
Full-text available
This article reports the results of an investigation of the relationship between early career success and past participation in an undergraduate field internship. The study extends earlier research on the effects of formal marketing education on career success. A survey of intern and nonintern business alumni of a northeastern U.S. public university indicated significant early career advantages for undergraduates with internship experience. Advantages included less time to obtain first position, increased monetary compensation, and greater overall job satisfaction. In addition to the career benefits provided to the students, the positive implications for marketing educators, university administrators, and intern employers are also discussed.
Article
A general theory of domain identification is used to describe achievement barriers still faced by women in advanced quantitative areas and by African Americans in school. The theory assumes that sustained school success requires identification with school and its subdomains; that societal pressures on these groups (e.g., economic disadvantage, gender roles) can frustrate this identification; and that in school domains where these groups are negatively stereotyped, those who have become domain identified face the further barrier of stereotype threat, the threat that others' judgments or their own actions will negatively stereotype them in the domain. Research shows that this threat dramatically depresses the standardized test performance of women and African Americans who are in the academic vanguard of their groups (offering a new interpretation of group differences in standardized test performance), that it causes disidentification with school, and that practices that reduce this threat can reduce these negative effects.
Article
Participants (168 female and 118 male undergraduate students) completed a brief questionnaire, a measure of perceived educational and career‐related barriers, and a measure of coping efficacy. As expected, women and ethnic minorities anticipated significantly more career‐related barriers than did men and European American students, respectively. Ethnic minorities also exhibited more perceived educational barriers and lower self‐efficacy for coping with perceived career‐related barriers relative to their European American counterparts. Findings are discussed in terms of their theoretical implications and practical career counseling applications.
Article
The pattern of honours degrees awarded to women and to men by British universities in 1967, 1978 and 1979 is examined. Women gained lower percentages than the men of both first class degrees and the weakest honours degrees. Various explanations for the smaller percentage of women gaining firsts are considered. The only explanation that seems to fit all the facts is that this difference is linked to differences in the distribution of ability as measured by the scores gained in intelligence tests.