Parallel Enrollments: Associations Between College Student Satisfaction and Performance

Differentiation of International Programs, Parallel Inscriptions, and Implications

Enrollment into a degree program in Austria is possible prior to the beginning of an academic year (October) as well as prior to the beginning of the summer term (March) for programs without specific requirements (Federal Ministry of Education, Science and Research, 2002). Beyond the official inscription periods, no time-related limitations exist for enrolling into study programs. It is possible to inscribe into degree programs at the same time or in a time-displaced manner. Having more than one valid enrollment in different programs simultaneously or changing the main focus of one's studies by enrolling into a new program and dropping out from another are both possible. The term “parallel” is used for studies of students being enrolled into more than one program at once with overlapping timeframes during which those programs have valid inscriptions. In general, other countries allow parallel enrollments. For instance, in Germany students can enroll in two study programs, but may be hindered by the numerus clausus and differing matriculation laws in the federated states (e.g., State Parliament of Baden-Württemberg, 2005; State Parliament of Berlin, 2011). Austria may be unique in its freedom of enrolling into an unlimited number of study programs with little bureaucracy country-wide.

This definition of studying two or more degree programs simultaneously needs to be distinguished from similar terms with different meanings. “Dual enrollment” is a term used for college programs specifically set up for advanced high school students. It allows them to enroll in high school and college courses simultaneously (Barnett & Stamm, 2010). This type of program has been shown to facilitate the transition between these two stages and to increase the odds of completing a postsecondary credential (Ison, 2022). Double degree, joint or combined degree programs usually combine two bachelor's degrees into one program. They are studied concomitantly over a longer period of time compared to single degree programs (e.g., 5 years). Students have the benefit of graduating in two fields in a shorter period as if they did both degrees consecutively (Russell et al., 2008) and may do so at two different universities and countries (Dukhanov et al., 2014). Both types of programs are similar to parallel studies that there is more than one focus for the learners. However, the major difference is that they are considered one program, underlying the rules and curricula of such. Parallel enrollments are centered on two or more study programs, distinct from each other and having little to no simplification and ease of the students’ workloads until graduation. Transferring passed exams from one program to the others is possible (Federal Ministry of Education, Science and Research, 2002), but the number of exams that can be skipped within a curriculum by transfer depends on the similarity of the fields of studies and the approval of the curriculum managers.

Student Performance in Parallel Studies

Previous research on the Austrian university system focuses on governance and management (De Boer et al., 2007; Sperrer et al., 2016), third mission (Trippl et al., 2012), regional development (Trippl et al., 2015), and other higher level topics. There is research on academic success (Bartok et al., 2021), but there is a gap in specific topics such as parallel inscriptions. Since the Austrian university system is organized differently from other countries, especially in this regard, there is no research that could be generalized to the situation of Austria.

In 2018 and 2022 student performance and workload became a critical part of the system with the introduction of two laws, regulating the financing of the universities and matriculation. First, students who reach 16 ECTS (European Credit Transfer System) credits or more within an academic year were made part of the base for the public funding of Austrian universities. Per definition, each study program of a given student reaching this threshold is considered “active” and eligible for funding. Studies below this boundary do not generate revenue for the universities. Second, a minimum requirement for studying was introduced. Defined as a need to accumulate 16 ECTS credits in the first two academic years of a study program, this threshold must be crossed by each new student in each program to be able to keep the enrollment valid. If this requirement is not reached, students are not allowed to continue and lose their admission to the degree program (Federal Ministry of Education, Science and Research, 2018, 2022a). Since these new regulations have implications for the amount of effort required from students (particularly from those pursuing more than one enrollment), they impact not only the university structures, but also students’ levels of commitment to studying. Research in the area of parallel inscriptions is needed for two major reasons: (1) Both from a governance and student perspective student workload and performance have become central components of financing and graduation. (2) Being enrolled into multiple study programs means that these laws apply to each one separately. Students must cross 16 ECTS credits in every academic year per program so that the university gets the funding and 16 ECTS credits in their first 2 years to stay in the degree programs. Therefore, simultaneous enrollments constitute a challenge to both students and universities. In respect to getting a job after graduation, receiving two degrees after graduation from parallel enrolled programs can be a desirable outcome for students worldwide. By exploring dimensions that promote student performance—and success—in this subgroup of the general student population, policy makers, and universities may benefit from Austria's unique system.

One of these dimensions can be student satisfaction, as it was previously shown to have important relations to student performance and success (Duque, 2014). Both satisfaction with the university environment, for example, programs and services, and formal structures such as registration effectiveness have a positive impact on students’ performance (Karemera et al., 2003; Oja, 2011). It can be an important control mechanism for policy makers and the steering of a university, as it can be influenced by university governance. Additionally, teaching quality of lecturers and learning facilities also promote satisfaction and can be affected by top-down decisions (Martono et al., 2020).

Expectations

Due to the uniqueness of the Austrian university system, there has been no previous research on student performance from students with multiple inscriptions under similar conditions. Understanding how having the freedom of studying more than one degree program simultaneously affects student performance and how student satisfaction plays a role in this context can be an important addition to the current literature of influences on student performance. Knowing about its effects may aid policy makers and curriculum designers around the world in their decisions. This is a first approach to studying the effects of multiple enrollments on student performance as a base for future studies. The aim of this study is to exploratively compare associations between satisfaction with one's studies and major performance indicators of students with an enrollment into one degree program to those with two or more enrollments. Insights into satisfaction as predictors for performance should be derived to create a body of evidence for future research projects and recommendations on allowing students to study more than one program simultaneously. Descriptive statistics have been applied to the university's administrative database as well as official statistics from online databases have been used to broaden the understanding of the student population and student behavior in Austria in regards to parallel enrollments.

Data Background

The data was collected as part of a larger project concerned with the effects of working students and parallel inscriptions on study performance. Students from the University of Graz (Austria, Europe) were the base for data collection. It is Austria's second biggest university with around 30,000 enrolled students per academic year. The city of Graz is known as students’ city, housing the University of Graz, the Medical University of Graz, the University of Music and Performing Arts Graz, the Technical University of Graz and some smaller universities of applied science, all in close range to each other. Therefore, parallel inscriptions in multiple studies are not only attainable due to the Austrian policy (Federal Ministry of Education, Science and Research, 2002), but also due to spatial proximity. It is representative of middle sized central European universities.

Data collection was performed at the end of June 2023 via social media and a university-wide mailing list, targeting all students being currently enrolled in at least one study program and not being delisted from this mailing service for surveys. The specific point in time was chosen since it is the last possible timeframe for data collection at the end of an academic year, prior to 3 months of summer holidays; the next academic year starts in October. At the end of June, these weeks are the hot-spots for examinations and finals in most fields of study for the current academic year, concentrating the thoughts and opinions of students on university-related things more so than at other times during the year. By making the student ID a mandatory input for filling in the questions, the answers to the questionnaire could be matched with data from the university's internal database. Focusing on the entire past academic year, performance indicators, sociodemographic variables and parallel inscriptions could be joined by using the student IDs. The data was merged mid-July 2023, when most of the exams had been finished, processed and the university's database had been synchronized. Students received no compensation for taking part in the study. The study has been approved by the ethics committee of the University of Graz (GZ 39/96/63 ex 2022/23).

Due to students being enrolled in multiple study programs, two datasets were created. One dataset was kept in the long format with each row representing one study program of one person. A person's answers on the survey were used for each study program, but performance indicators from the administrative database were unique for each row. The other dataset was created in the wide format with each row representing one person, containing multiple columns for different study programs.

Descriptive Exploration of the Student Population

Using the administrative data of the University of Graz, descriptive statistics have been calculated for all students with a valid enrollment between the academic years 2012/2013 and 2022/2023. For data stability reasons and due to changes in the university laws, data before this timeframe was not included. Between the academic years of 2012/2013 and 2022/2023, 124,047 students had a valid enrollment at the University of Graz. Around 32% enrolled in more than one program during this timeframe. Among this third of all students, the relative frequency of people starting two programs in the same semester (14%) was slightly lower than the relative amount of students starting another program after being already enrolled (16%). The largest number of parallel studies is bachelor's degree programs, making up around 68% of all students with more than one enrollment. Parallel master's programs could be found in 12% of the cases, with the rest spreading across diploma degree programs (10%), additional study programs with no formal degree (9%), and doctoral programs (2%). Regarding student activity, which is defined as 16 or more ECTS credits in the first academic year (Federal Ministry of Education, Science and Research, 2018; 2022), descriptive statistics have been calculated for the first semester in parallel enrollments. This means the semester a second program is started, in case both enrollments do not happen in the same semester. For the calculation, the activity threshold was defined as eight or more ECTS credits for one semester. The following outcomes could be obtained: 25% of students with two or more enrollments were only active in the second program, 24% only in their first program; 34% of such students were not active in both programs and 16% reached the theshold in both programs. In the target timeframe, the most common outcome was dropping out from both programs (38%). Double graduations were found in 4% of all cases. Finishing one program and exiting the other happened in 13% of all outcomes. Other students not included in these percentages had still at least one valid enrollment at the end of the academic year 2022/2023.

Following official university statistics of the Austrian Federal Ministry of Education, Science and Research, over all Austrian universities, 54% of students were female and 46% male. At the University of Graz, the percentages were a bit higher for women, for example, 62% in the same academic year. Over all universities in Austria, 68% of students came from Austria, 23% from EU countries, and 9% from other countries. In Graz, 84% of students came from Austria, 12% from EU countries and 5% from others (Federal Ministry of Education, Science and Research, 2024). Official statistics of the year 2022 over all OECD countries show that the relative frequency of foreign students was 6% women and 7% men, with a range between 48% (Luxembourg) and <1% (China, Colombia, Brazil, India). The total on EU countries is 8% for women and 9% for men and in the United States it is 4% for women and 7% for men (OECD, 2022).

Sample

A sample of 227 students filled in the questionnaire, with 213 entering a valid student ID for data merging and subsequent analyses under exclusion of the rest. Of the people considered for analyses, 161 people were female (76%) and 52 male (24%). Compared to the university-wide data from 2022/2023, the distribution was more in favor of women in this study. The mean age was 25 years (SD = 6.65 years). And 177 of them had an Austrian citizenship (83%), while 35 (16%) came from other EU countries and one student from a non-EU country, approximately reflecting the ratio found in the official statistics for Austria. Compared to the EU total, Austria has a higher percentage of international students. A total of 47 students (22% of 213) taking part in the study had two or more enrollments that could be matched with data from the administrative database. The relative frequency of parallel programs in this study was a bit lower than the values obtained from all cohorts between 2012/2013 and 2022/2023. The maximum number of parallel study programs in the sample was four. Among the participants, 168 students were enrolled in one program (79%), while 36 students had two (17%), eight students had three and one person had four simultaneous inscriptions at the University of Graz. Answers to the question whether they have a main study program (plus a text detailing which one) or whether all are treated equally suggests that four students were also enrolled in studies at other universities in Graz. No performance data was available from other universities, as the databases are enclosed and university exclusive. Enrollment at other institutions could only be estimated by the answers given, it cannot be ruled out that other students did not fill in possible enrollments at other universities.

Procedure

Students who clicked on the link to the survey were forwarded to a web-interface showing the general information, informed consent, and the data privacy statement of the study. They were informed about the data merging procedure and that their privacy is guaranteed, using their student ID only for the purpose of connecting the answers from the survey to data from the administrative database of the university. Informed consent and data privacy had to be confirmed separately. If one was missing, continuation of the participation would not have been possible. After these confirmations, the student ID was collected.

Other questions related to the project regarding student jobs and student work, financing of the studies, daily activities, and interests followed. Relevant for this study, questions on how students feel about their studies at the moment of filling in the survey in regard to (dis)satisfaction with several components of student retention were asked. The survey was concluded by items on stress and coping as well as the filter questions on possible parallel enrollments. If they answered with “yes” for multiple simultaneous study programs, they received a question about which programs they considered their main studies or whether they treated all of them as equal. Students with only one program enrolled were automatically assigned the top-priority label for their study program.

Variables

The major outcome variables were retrieved from performance indicators in the administrative database. The accumulated ECTS credits (“ECTS”), the number of exams (“n exams”), the number of enrolled courses (“n courses”), the grade point average (“M grades”), and the number of negative grades (“n negative”) at the end of the target semester were used as major outcome variables. ECTS credits are the European standard for student workload with one credit referring to 25 real-time working hours (Karran, 2004; European Union, 2015). In Austria, grades range from 1 to 5, with 1 being the best. Negative grades refer to the number 5, which means that an exam is not passed. Students can repeat exams up to four times (five times in some special programs) until they are banned from taking the same exam again at their institution (Austrian Federal Ministry of Edcuation, Science and Research, 2002). The variable n exams refer to the number of exams each student has taken in the target academic year, n courses to the number of courses a student was enrolled in. These numbers do not have to match, since students decide when to take an exam out of a pool of a few possible examination dates, which is the case, for instance, for lectures. M grades are the mean of all exam grades with higher values meaning worse grade point average.

In the wide-format dataset, these variables were used separately for the two most important studies of students with more than one inscription. The importance was determined via the question asking which of their studies was treated as their main program. If they chose the option that all of their studies were treated equally, studies were ranked via the number of ECTS credits from highest to lowest, that is, per workload. Studies ranking higher than the second enrollment (three and above) were not included in the analyses, since this applied to a small sample of nine students. Descriptive statistics for the outcome variables in the whole dataset are shown in Table 1.

OPEN IN VIEWER

Table 1.

Students’ Performance Indicators as Outcome Variables for Statistical Analyses.

Variable	n	Minimum	Maximum	M	SD
Full dataset
ECTS	192	0	57	15.44	9.76
n exams	213	0	17	4.40	3.48
n courses	213	0	21	4.76	3.81
M grades	190	1	5	2.04	0.87
n negative	213	0	3	0.24	0.61
One study program
ECTS	149	0	52	15.02	9.27
n exams	163	0	17	4.72	3.54
n courses	163	0	17	4.87	3.70
M grades	148	1	5	2.05	0.90
n negative	163	0	3	0.27	0.66
Parallel enrollments: Prioritized program
ECTS	43	1	57	16.87	11.31
n exams	50	0	15	3.34	3.07
n courses	50	0	21	4.40	4.15
M grades	42	1	4	1.98	0.77
n negative	50	0	2	0.16	0.42
Parallel enrollments: Nonprioritized program
ECTS	25	0	22	7.56	5.55
n exams	47	0	6	1.04	1.33
n courses	47	0	6	1.19	1.51
M grades	25	1	5	2.07	1.32
n negative	47	0	1	0.13	0.34

The predictor variables were obtained from six items of the question: “Adjust the following sliders for each item between 0% (not at all) and 100% (exactly) that it reflects how you feel at the moment.” Each of them came with a slider between 0% (not at all) and 100% (exactly) that was meant to represent students’ personal feelings about all of their studies at the given moment of filling in the questionnaires. They items were students’ “program satisfaction,” “performance satisfaction in their studies,” “course satisfaction,” “the wish to continue studying,” “the wish to graduate,” and “the wish to drop out.” Since students with one enrollment up to several simultaneous enrollments were expected to be in the sample, there was no specification of these questions depending on the number of study programs a student is enrolled in. Descriptive statistics for the predictor variables are listed in Table 2.

OPEN IN VIEWER

Table 2.

Descriptive Statistics for Predictor Variables of the Statistical Analyses.

Variable	n	Minimum	Maximum	M	SD
Predictors
Program satisfaction	213	0	100	66.67	23.72
Performance satisfaction	213	1	100	68.30	24.46
Course satisfaction	213	0	100	58.54	23.07
Wish to continue	213	0	100	82.57	26.14
Wish to graduate	213	0	100	92.75	18.79
Wish to drop out	213	0	100	14.04	24.60

Apparatus

The online questionnaire was set up using LimeSurvey®. Data was retrieved from the administrative Oracle® SQL database of the university. Querying data, preparation, and visualization were done via R (R Core Team, 2022), using the RODBC (Ripley & Lapsley, 2022), and ggplot2 packages (Wickham et al., 2023). Regression analyses were done in IBM SPSS® 29. Power analyses were done using G*Power (Faul et al., 2007).

Statistical Analyses

Multiple Pearson-correlations on the full dataset were conducted between the major study variables and depicted as a correlation diagram. All regression analyses were applied to the dataset on three different levels: (1) unfiltered, (2) only students with one enrollment, (3) students with two or more enrolled studies. The first level means the whole dataset; all students regardless of the number of enrollments are included. Analyses are done on the prioritized program for students with two or more enrollments. On the second level, the dataset was filtered and calculations are done for students with one enrollment. The third levels is centered on students with two or more enrollments and analyses are done both for the prioritized and nonprioritized programs. Multiple one-factorial analysis of variance (ANOVA) models were applied to the long-format dataset. Post hoc tests were done using Bonferroni adjustment. Multiple linear regression models were used to analyze the associations between outcome variables and predictors. The method used was “enter”; listwise exclusion was selected for missing values. Model requirements for ANOVA and Regression models were fulfilled for most of the models (e.g., Levene's test, variance inflation factor). The according statistical parameters are provided. Due to sample size, similarity of the utilized models and their robustness to violation, nonparametrical alternatives were not considered. Additionally, a correlation diagram for all variables using the unfiltered dataset as a base is included in the appendix.

Main Results

Multiple correlation analyses were conducted between the variables used in the regression models. The results are depicted in Figure 1.

OPEN IN VIEWER

Figure 1.

Correlation matrix for all variables.

Multiple multivariate linear regression models have been calculated on the wide-format dataset. They show whether the predictor variables can explain variance in student performance. All models were set up for each analysis level. Tables 3 to 6 depict the results of the models, with VIF being the Variance Inflaction Factor.

OPEN IN VIEWER

Table 3.

Multivariate Regression Models Predicting Student Performance Indicators of Students in the Whole Dataset, for Their Prioritized Study Program.

Outcome	R2	F(df1, df2)	p	Durbin Watson	Predictors	b	t	p	CI	Tolerance	VIF
ECTS	.04	1.26 (6, 185)	.278	2.21	Program satisfaction	0.08	1.52	.129	−0.02–0.18	.38	2.65
					Performance satisfaction	0.04	1.10	.272	−0.03–0.10	.85	1.18
					Course satisfaction	−0.01	−0.13	.895	−0.09–0.08	.53	1.91
					Wish to continue	−0.06	−1.33	.185	−0.15–0.03	.38	2.61
					Wish to graduate	0.04	0.73	.468	−0.07–0.16	.51	1.96
					Wish to drop out	−0.02	−0.39	.700	−0.09–0.06	.57	1.76
n exams	.02	0.77 (6, 206)	.597	1.64	Program satisfaction	0.01	0.77	.443	−0.02–0.05	.34	2.95
					Performance satisfaction	0.01	0.96	.339	−0.01–0.03	.80	1.24
					Course satisfaction	−0.02	−1.05	.295	−0.05–0.01	.49	2.05
					Wish to continue	−0.003	−0.19	.850	−0.03–0.03	.36	2.81
					Wish to graduate	0.003	0.18	.855	−0.03–0.04	.50	1.99
					Wish to drop out	−0.01	−0.90	.368	−0.04–0.01	.61	1.64
n courses	.02	0.58 (6, 206)	.744	1.91	Program satisfaction	0.003	0.15	.882	−0.04–0.04	.34	2.95
					Performance satisfaction	0.01	0.77	.444	−0.01–0.03	.80	1.24
					Course satisfaction	−0.01	−0.41	.683	−0.04–0.03	.49	2.05
					Wish to continue	0.002	0.10	.922	−0.03–0.04	.36	2.81
					Wish to graduate	0.01	0.32	.751	−0.03–0.05	.50	1.99
					Wish to drop out	−0.01	−0.80	.424	−0.04–0.02	.61	1.64
M grades	.16	5.88 (6, 183)	<.001*	1.86	Program satisfaction	0.001	0.34	.735	−0.01–0.01	.39	2.57
					Performance satisfaction	−0.01	−5.12	<.001*	−0.02 - −0.01	.86	1.16
					Course satisfaction	−0.001	−0.20	.845	−0.01–0.01	.53	1.88
					Wish to continue	−0.01	−1.69	.094	−0.01–0.001	.41	2.44
					Wish to graduate	−0.002	−0.45	.657	−0.01–0.01	.56	1.80
					Wish to drop out	−0.01	−1.49	.138	−0.01–0.002	.61	1.63
n negative	.08	2.94 (6, 206)	.009*	1.76	Program satisfaction	0.002	0.73	.469	−0.004–0.01	.34	2.95
					Performance satisfaction	−0.01	−3.66	<.001*	−0.01—−0.003	.80	1.24
					Course satisfaction	<0.001	−0.06	.953	−0.01–0.01	.49	2.05
					Wish to continue	−0.01	−1.83	.068	−0.01–0.000	.36	2.81
					Wish to graduate	0.004	1.18	.238	−0.002–0.01	.50	1.99
					Wish to drop out	−0.001	−0.38	.702	−0.01–0.003	.61	1.64

OPEN IN VIEWER

Table 4.

Multivariate Regression Models Predicting Student Performance Indicators of Students With Only One Enrollment.

Outcome	R2	F(df1, df2)	p	Durbin Watson	Predictors	b	t	p	CI	Tolerance	VIF
ECTS	.05	1.16 (6, 142)	.328	1.82	Program satisfaction	0.11	1.74	.084	−0.02–0.23	.32	3.11
					Performance satisfaction	0.05	1.46	.146	−0.02–0.12	.84	1.19
					Course satisfaction	−0.07	−1.41	.162	−0.17–0.03	.47	2.13
					Wish to continue	−0.06	−1.12	.264	−0.16–0.04	.36	2.76
					Wish to graduate	0.04	0.64	.522	−0.09–0.17	.45	2.25
					Wish to drop out	−0.01	−0.27	.791	−0.11–0.08	.50	2.02
n exams	.02	0.53 (6, 156)	.784	1.54	Program satisfaction	0.003	0.11	.910	−0.04–0.05	.29	3.50
					Performance satisfaction	0.01	1.08	.283	−0.01–0.04	.77	1.29
					Course satisfaction	−0.01	−0.59	.554	−0.05–0.03	.42	2.36
					Wish to continue	0.001	0.06	.953	−0.04–0.04	.34	2.98
					Wish to graduate	0.02	0.77	.442	−0.03–0.06	.44	2.25
					Wish to drop out	0.001	0.04	.970	−0.03–0.03	.56	1.78
n courses	.02	0.57 (6, 156)	.750	1.59	Program satisfaction	0.01	0.27	.785	−0.04–0.05	.29	3.50
					Performance satisfaction	0.02	1.10	.272	−0.01–0.04	.77	1.29
					Course satisfaction	−0.01	−0.72	.471	−0.05–0.02	.42	2.36
					Wish to continue	−0.002	−0.13	.897	−0.04–0.04	.34	2.98
					Wish to graduate	0.02	0.87	.385	−0.03–0.06	.44	2.25
					Wish to drop out	0.001	0.04	.972	−0.03–0.03	.56	1.78
M grades	.14	3.96 (6, 141)	.001*	1.90	Program satisfaction	0.002	0.44	.658	−0.01–0.01	.34	2.97
					Performance satisfaction	−0.01	−4.24	<.001*	−0.02—−0.01	.85	1.17
					Course satisfaction	<0.001	0.002	.999	−0.01–0.01	.48	2.08
					Wish to continue	−0.01	−1.51	.134	−0.02–0.002	.40	2.53
					Wish to graduate	−0.002	−0.34	.736	−0.02–0.01	.50	1.98
					Wish to drop out	−0.01	1.05	.297	−0.01–0.004	.55	1.81
n negative	.08	2.32 (6, 156)	.035*	1.87	Program satisfaction	0.000	0.09	.928	−0.01–0.01	.29	3.50
					Performance satisfaction	−0.01	−3.17	.002*	−0.01 – −0.003	.77	1.29
					Course satisfaction	0.002	0.52	.607	−0.01–0.01	.42	2.36
					Wish to continue	−0.004	−1.27	.207	−0.01–0.002	.34	2.98
					Wish to graduate	0.01	1.53	.127	−0.002–0.01	.44	2.25
					Wish to drop out	0.001	0.39	.699	−0.004–0.01	.56	1.78

OPEN IN VIEWER

Table 5.

Multivariate Regression Models Predicting Student Performance Indicators of Students With Two or More Enrollments, for Their Prioritized Study Program.

Outcome	R2	F(df1, df2)	p	Durbin Watson	Predictors	b	t	p	CI	Tolerance	VIF
ECTS	.18	1.30 (6, 36)	.284	2.31	Program satisfaction	0.01	0.09	.926	−0.21–0.23	.41	2.46
					Performance satisfaction	0.03	0.25	.801	−0.17–0.22	.66	1.53
					Course satisfaction	0.23	2.13	.040	0.01–0.45	.52	1.91
					Wish to continue	−0.16	−1.38	.177	−0.39–0.07	.35	2.85
					Wish to graduate	0.12	0.84	.408	−0.17–0.40	.55	1.81
					Wish to drop out	−0.08	−0.95	.351	−0.24–0.09	.60	1.67
n exams	.10	0.80 (6, 43)	.578	1.80	Program satisfaction	0.03	1.17	.250	−0.02–0.09	.40	2.49
					Performance satisfaction	0.01	0.21	.835	−0.04–0.05	.70	1.44
					Course satisfaction	−0.04	−1.30	.202	−0.09–0.02	.49	2.03
					Wish to continue	−0.01	−0.34	.738	−0.07–0.05	.34	2.91
					Wish to graduate	−0.01	−0.35	.731	−0.09–0.06	.58	1.74
					Wish to drop out	−0.02	−1.09	.283	−0.07–0.02	.59	1.69
n courses	.04	0.30 (6, 43)	.934	2.09	Program satisfaction	−0.004	−0.09	.928	−0.08–0.08	.40	2.49
					Performance satisfaction	0.01	0.20	.846	−0.06–0.07	.70	1.44
					Course satisfaction	0.01	0.22	.826	−0.07–0.09	.49	2.03
					Wish to continue	0.003	0.07	.947	−0.08–0.09	.34	2.91
					Wish to graduate	−0.01	−0.14	.888	−0.11–0.10	.58	1.74
					Wish to drop out	−0.03	−0.96	.342	−0.09–0.03	.59	1.69
M grades	.27	2.14 (6, 35)	.073	1.85	Program satisfaction	0.002	0.22	.826	−0.01–0.02	.41	2.47
					Performance satisfaction	−0.02	−2.60	.014	−0.03—−0.004	.65	1.54
					Course satisfaction	−0.01	−0.81	.422	−0.02–0.01	.52	1.92
					Wish to continue	−0.01	−0.66	.516	−0.02–0.01	.35	2.82
					Wish to graduate	−0.003	−0.38	.705	−0.0–0.02	.55	1.81
					Wish to drop out	−0.004	−0.87	.391	−0.02–0.01	.60	1.67
n negative	.18	1.60 (6, 43)	.170	1.36	Program satisfaction	0.004	1.10	.277	−0.003–0.01	.40	2.49
					Performance satisfaction	−0.01	−1.70	.096	−0.01–0.001	.70	1.44
					Course satisfaction	−0.004	−1.22	.230	−0.01–0.003	.49	2.03
					Wish to continue	−0.01	−1.40	.170	−0.01–0.002	.34	2.91
					Wish to graduate	−0.003	−0.59	.556	−0.01–0.01	.58	1.74
					Wish to drop out	−0.003	−1.23	.224	−0.01–0.002	.59	1.69

OPEN IN VIEWER

Table 6.

Multivariate Regression Models Predicting Student Performance Indicators of Students With Two or More Enrollments, for Their Nonprioritized Second Study Program.

Outcome	R2	F(df1, df2)	p	Durbin Watson	Predictors	b	t	p	CI	Tolerance	VIF
ECTS	.53	3.34 (6, 18)	.022	2.08	Program satisfaction	0.02	0.35	.730	−0.11–0.15	.28	3.56
					Performance satisfaction	0.02	0.33	.748	−0.10–0.14	.46	2.20
					Course satisfaction	0.10	1.47	.160	−0.04–0.25	.35	2.87
					Wish to continue	0.02	0.39	.698	−0.09–0.13	.32	3.11
					Wish to graduate	0.09	1.61	.124	−0.03–0.22	.54	1.86
					Wish to drop out	0.06	1.53	.143	−0.02–0.13	.56	1.79
n exams	.09	0.66 (6, 40)	.682	1.45	Program satisfaction	−0.003	−0.26	.796	−0.03–0.02	.41	2.42
					Performance satisfaction	−0.01	−0.64	.525	−0.03–0.01	.69	1.46
					Course satisfaction	0.01	1.01	.319	−0.01–0.04	.51	1.95
					Wish to continue	−0.01	−0.71	.480	−0.04–0.02	.35	2.86
					Wish to graduate	0.02	1.20	.236	−0.01–0.05	.57	1.74
					Wish to drop out	0.01	0.69	.496	−0.01–0.03	.59	1.69
n courses	.14	1.10 (6, 40)	.378	1.59	Program satisfaction	−0.004	−0.28	.780	−0.03–0.02	.41	2.42
					Performance satisfaction	−0.004	−0.39	.702	−0.03–0.02	.69	1.46
					Course satisfaction	0.02	1.26	.216	−0.01–0.05	.51	1.95
					Wish to continue	<0.001	−0.002	.999	−0.03–0.03	.35	2.86
					Wish to graduate	0.02	1.01	.319	−0.02–0.05	.57	1.74
					Wish to drop out	0.02	1.59	.119	−0.004–0.04	.59	1.69
M grades	.55	3.65 (6, 40)	.015*	2.26	Program satisfaction	−0.02	−1.30	.212	−0.05–0.01	.28	3.56
					Performance satisfaction	0.01	0.65	.526	−0.02–0.04	.46	2.20
					Course satisfaction	0.03	2.13	.047*	0.000–0.07	.35	2.87
					Wish to continue	−0.03	−2.38	.029*	−0.06—−0.003	.32	3.11
					Wish to graduate	−0.02	−1.26	.224	−0.05–0.01	.54	1.86
					Wish to drop out	−0.01	−1.47	.159	−0.04–0.01	.56	1.79
n negative	.08	2.94 (6, 206)	.009*	1.76	Program satisfaction	<0.001	−0.10	.922	−0.01–0.01	.41	2.42
					Performance satisfaction	0.001	0.34	.735	−0.004–0.01	.69	1.46
					Course satisfaction	0.002	0.62	.537	−0.004–0.01	.51	1.95
					Wish to continue	−0.01	−2.47	.018*	−0.01—−0.001	.35	2.86
					Wish to graduate	−0.01	−1.36	.182	−0.01–0.002	.57	1.74
					Wish to drop out	−0.001	−0.72	.475	−0.01–0.003	.59	1.69

Additional Results

Multiple one-factorial univariate ANOVAs were calculated on the long-format dataset for the full dataset and the filtered dataset including parallel inscriptions. The major outcome variables were used as dependent measures and the independent measure was the priority of study programs. No models have been calculated for the dataset of students with only one inscription, since there are no groups in the independent variable that could be used for comparison. The results are listed in Table 7.

OPEN IN VIEWER

Table 7.

Results of Univariate ANOVA Models for the Major Outcome Variables Dependent on Prioritization of Parallel Study Programs.

Full dataset (n = 213)
Model nr	Dependent	F	df	p	η p 2	Levene's p
1	ECTS	3.26	2, 189	.033	.03	.200
2	n exams	19.92	2, 265	<.001	.13	<.001
3	n courses	12.10	2, 265	<.001	.08	.094
4	M grades	1.00	2, 189	.371	.01	.668
5	n negative	0.57	2, 265	.567	.004	.085
Post hoc comparison						p
Model 1: ECTS
Same (M = 7.22, SD = 3.93) × main (M = 10.24, SD = 6.30)						.152
Same (M = 7.22, SD = 3.93) × side (M = 7.64, SD = 6.51)						.975
Main (M = 10.24, SD = 6.30) × side (M = 7.64, SD = 6.51)						.127
Model 2: n exams
Same (M = 1.09, SD = 1.42) × main (M = 3.02, SD = 2.50)						<.001*
Same (M = 1.09, SD = 1.42) × side (M = 1.18, SD = 1.94)						.982
Main (M = 3.02, SD = 2.50) × side (M = 1.18, SD = 1.94)						<.001*
Model 3: n courses
Same (M = 1.39, SD = 1.71) × main (M = 3.19, SD = 2.64)						<.001*
Same (M = 1,39, SD = 1.71) × side (M = 1.60, SD = 2.93)						.932
Main (M = 3.19, SD = 2.64) × side (M = 1.60, SD = 2.93)						<.001*
Model 4: M grades
Same (M = 2.39, SD = 1.09) × main (M = 2.05, SD = 0.94)						.341
Same (M = 2.39, SD = 1.09) × side (M = 2.05, SD = 1.26)						.512
Main (M = 2.05, SD = 0.94) × side (M = 2.05, SD = 1.26)						.999
Model 5: n negative
Same (M = 0.09, SD = 0.29) × main (M = 0.14, SD = 0.42)						.745
Same (M = 0.09, SD = 0.29) × side (M = 0.09, SD = 0.29)						1.00
Main (M = 0.14, SD = 0.42) × side (M = 0.09, SD = 0.29)						.641
Parallel study programs (n = 47)
Model nr	Dependent	F	df	p	η p 2	Levene's p
1	ECTS	0.15	2, 47	.866	.62	.016
2	n exams	0.26	2, 97	.776	.01	.511
3	n courses	0.12	2, 97	.886	.002	.117
4	M grades	1.20	2, 49	.310	.05	.528
5	n negative	0.58	2, 97	.561	.01	.061

ANOVA: analysis of variance.

Satisfaction and Performance

This study aimed toward exploratively obtaining evidence about associations between student performance and student satisfaction in the context of being able to simultaneously enroll in and study multiple programs at once. Over all models and analysis levels, no variance could be explained for the outcome variables ECTS credits, the number of exams or the number of courses by any of the predictors. Analyses on the unfiltered dataset showed that performance satisfaction, but no other variable, could negatively explain variance of students’ grades. The same predictor could also negatively explain variance of the number of negative exams, with no other predictors having a significant influence. The same results were obtained for students with only one enrollment. This means that the higher the performance satisfaction, the better (lower) the grade average and the lower the number of failed exams in the general dataset and for people with only one study program.

Unlike the first two analysis levels, students with two or more enrollments showed no associations between predictors and outcomes in their prioritized study program. However, course satisfaction and the wish to continue one's studies, but not performance satisfaction or other variables, explained variance of grades in the nonprioritized study programs with a moderate to large effect size of R² = .55. Course satisfaction showed a positive association, while the wish to continue was negatively associated with one's average grades. The higher the satisfaction with the courses in all of students’ studies, the worse (higher) the average grades and the higher the wish the continue one's studies, the better (lower) the average grades in the nonprioritized program. There was also a negative association between the wish to continue and the number of negative exams. The higher the wish to continue one's studies, the lower the number of negative exams in the nonprioritized programs. No other predictors showed significant influences in these models. Dimensions of satisfaction seem to differently contribute to student performance, depending on the priority of a program. Due to the prioritization some students define with their programs, the association between satisfaction and performance may be dependent on the reasons for prioritization. This means that a student could perform well in their main program when satisfaction with their courses is high. However, the same student might have a second program they use to register for courses of interest, which they are not allowed to take in their main program. In such a case, satisfaction with courses could be a bigger criterion for performance than the satisfaction with their general performance (measured in grades and workload), as courses are chosen due to interest instead of the structure of the curriculum in the nonprioritized program.

Prioritization of Programs

Including all students in the analyses, results showed that studies of people treated equally have a lower number of exams and courses compared to single enrolled programs and prioritized studies. Main programs or single enrolled programs have more exams and courses compared to second-priority programs. No effects in analyses concentrating solely on study programs from students with parallel programs were obtained. This means that there is an influence from people in one study program affecting the outcome of the analyses. These results may reflect the overall workload students have to invest in studying for exams and courses. Having only one program does not require distributing the workload compared to doing two or more studies, no matter if one is prioritized, while treating them equally assumes the same amount of workload for every program. However, there must be a certain amount of work even in nonprioritized studies. Due to Austrian law, students cannot ignore second-priority programs, since they have to fulfill the minimum requirement of 16 ECTS credits 2 years after enrollment for each study program separately in order to keep their enrollment valid (Federal Ministry of Education, Science and Research, 2022). Skill acquisition and enrichment of the curriculum vitae is a motivation for students to pursue a double degree program (Borsetto & Saccon, 2022, 2023). Assuming parallel enrollments are based on similar motivations and that students with multiple programs are not planning to graduate from all of them, there is still a need to fulfill these requirements to be able to take courses and exams in these studies.

The absence of associations between predictors and all of the outcomes for students in multiple programs in their prioritized programs compared to the nonprioritized programs as well as students with only one enrollment may partly be explained by the methodology. Students have been asked whether they treat their studies equally or if they have a main program they favor. In the analyses, prioritized programs have been defined by the maximal number of ECTS credits among all programs of each student, if they did not specify a main program. By mixing the two subgroups of students with a clear focus on one program with students dividing their focus on two or more studies, the data might have been skewed, not showing significant results. Additional descriptive statistics revealed that the groups of students considering their studies equal (48%) and having a main study program (52%) were almost the same size. Due to small sample sizes, a further distinction of these groups was not meaningful. Still obtaining significant associations in the nonprioritized programs may be due to the lower and worse performance indicators in these programs as a result of the ECTS-sorting process. Future studies should account for these possibly different groups among students with parallel enrollments. It should also be noted that previous research shows that student performance can also be linked to various dimensions of satisfaction such as satisfaction with academic environment, programs, and services (Karemera et al., 2003), student-centeredness, instructional effectiveness, academic counseling, registration effectiveness, service excellence, concern for the individual, campus climate (Oja, 2011), and other dimensions (e.g., Duque, 2014; Schertzer & Schertzer, 2004). This means that some of the variance not explained in this study may possibly be explained by other dimensions of student satisfaction not being included in the questionnaire.

Associations Between Predictors and Performance Parameters

The results of this study suggest that there are differences depending on the performance parameters and the predictors. ECTS credits, the number of exams and the number of courses showed no associations with the predictor variables of satisfaction. Especially workload as the number of accumulated ECTS credits is a major performance criterion for students in the Austrian university system (Federal Ministry of Education, Science and Research, 2002). A previous study found differences in satisfaction depending on both ECTS workload and grade point average. However, satisfaction was operationalized for teaching–learning processes, not study programs in general (Fuente et al., 2011). The missing effects being attributed to the operationalization of the items used to measure student satisfaction is possible, but rather unlikely, since significant effects on other performance indicators could be obtained. Due to the cross-sectional character of the study and correlative analyses, at least some predictors and outcomes may have influences in both directions. The outcome of performance satisfaction explaining variance of grades, could be of bidirectional nature. This means that bad grades lower the satisfaction of students. A longitudinal study showed that better grades increase life satisfaction of college students (Slavinski, Bjelica, Pavlović & Vukmirović, 2021). Therefore, one explanation is that some of the effects that were obtained in this study could have bidirectional connections to each other. The nonsignificant models may not be dependent on the outcomes influencing the satisfaction of students. ECTS credits as a measure of workload, the number of exams and courses in a semester may have less influence on one's satisfaction of studying in general than grades and failed exams.

This explanation is supported by the models showing that students’ perception of their own performance can explain variance of average grades and the number of failed exams in the overall models and for people in one study program. However, for students in multiple study programs the wish to continue studying had an influence on grades and negative exams, promoting better performance in the nonprioritized study programs. Students’ grades in nonprioritized programs were negatively affected by their overall satisfaction with the courses, meaning that a higher overall satisfaction is associated with worse grades. Since no effects were obtained for the prioritized programs, these results add to the notion that enrolling into additional study programs can be an escape from other programs for some students. Instead of dropping out from university altogether, they may first choose to try out another field of study. The obtained influence from this predictor may represent students’ focus on a main study program and being satisfied with the courses there. If satisfaction with the courses is high in their prioritized program, less focus may be laid on the other programs, leading to worse grades. This study did not have information on the chronological order of enrollments and there was also no information about students changing their perceived main study program. To support this theory of enrolling into a new program as an alternative to dropping out, future studies with a focus on these changes over the student lifecycle are needed.

Sample Size and Statistical Power

Post hoc power analyses for multiple regression analyses on 47 students with more than one enrollment, six predictors, and effect sizes between R² = .08 and R² = .55 suggest power estimations of .22 and .96, respectively. Although future studies will have to consider bigger samples of students to reveal smaller effects, it can be argued that moderate to strong effects are of higher interest when it comes to gaining knowledge for implications for university management and policy makers. Given the same analyses, a priori power analyses with estimations of moderate effects, that is, R² = .30, and a power of .80, a sample size of 53 students would have been sufficient. The sample at hand (47 students) equates a power of .74 for such moderate effects and can be considered acceptable, as it was close to .80. Although statistical power was moderate to good in this study, a broader assessment on larger samples with better power is warranted to replicate the results.

Methodological Limitations

As one of the first approaches to address the Austrian university system in regards of parallel enrollments, this study concentrated on the concept of student satisfaction as main predictors for student performance. Retrieving the performance indicators of each student from the administrative database of the university leaves no room for error and missing cases. Although the overall research project concentrated on student work and parallel enrollments and was promoted as such, the sample size of students with two or more programs was rather low. Due to time constraints created by the nearing summer holidays for all students and the end of the examination and finals phase at the end of June, an expansion of the data collection seemed not meaningful. Extending the time period for recruitment would have meant that students filling in the questionnaire later, would have been faced with a very different situation. It can be assumed that especially stress levels would have been different for the students in the sample (Zunhammer et al., 2013) compared to possible additions to it. Since stress is linked to student satisfaction (Vermisli et al., 2022), this could have potentially affected the analyses.

Students have not been asked the questions of the survey for all of their studies separately. This study does not contain information about how satisfaction applies to each of the parallel enrolled programs. A higher granularity of the data may lead to the statistical models better representing the real-world situation. This means more significant influences from satisfaction on performance may be obtainable if the predictors are operationalized for each program.

There was no possibility to control for enrollments over several universities and more than the top-two programs were not considered in the analyses. In respect to the 16 ECTS threshold as the minimum requirement for keeping an enrollment after the second year (Federal Ministry of Education, Science and Research, 2022), students in programs at different universities may have a higher workload demand than observed in this study. Four students mentioned having another enrollment at a different university. Not being able to control the true enrollments of students via access to the other university's registrar data leaves a potential for error. For instance, students with one enrollment in this study could actually have other enrollments elsewhere, not answering the according question about multiple enrollments correctly. It needs to be noted that the absence of tuition fees and the freedom of enrollment applies to all public Austrian universities (Federal Ministry of Education, Science and Research, 2002). Therefore, parallel enrollments across universities are possible without financial or structural consequences for students other than a higher self-management demand that comes with course enrollment and spatial distances of the institutions.

What also needs consideration is the representativeness of the sample. It reflects the situation at the University of Graz and at other Austrian universities. Due to different university systems worldwide, the available data suggests limited comparability to EU, OECD, and US regions of the sample. However, different sample characteristics do not mean that aspects of this study cannot be applied to other systems. Despite some possibly unique properties, such as student activity, Austria's university system has other attributes comparable to institutions worldwide. Especially in regard to the possibility of parallel enrollments, other university systems currently not offering similar opportunities for students may profit form the results of this study. This study is a first step toward evaluation and further research of possible benefits of parallel enrollments in both similar and completely different university systems. The University of Graz being representative for a mid-European university, overlap with similar institutions in Europe and the EU regions can be assumed. For instance, although German university law is different from Austria's, there is a possibility for enrolling into more than one program. There is also a similar concept to the Austrian tolerance semesters in some federal states, meaning that there are no general tuition fees until a certain number of semesters is crossed (e.g., State Parliament of Baden-Württemberg, 2005).

Future Outlook and Implications

Student satisfaction plays a role for both students with one enrollment and students with multiple enrollments. Especially nonprioritized parallel studies showed a moderate to large effect of overall course satisfaction and the wish to continue studying influencing the grades of students in their nonprioritized studies. The results suggest that giving students the freedom of simultaneously enrolling into as many programs as they want may be one way to prevent immediate student dropout and that students may use additional programs as an alternative to an existing enrollment. This points out the need for policy makers and university management to set up sufficient support measures for both groups of students to increase satisfaction. Such structures could be identified as important key measures that promote student retention and success (Suhlmann et al., 2018; Zając & Komendant-Brodowska, 2019). However, clearly more research is needed about the long-term implications of multiple enrollments along the student lifecycle on dropping out and the motivations and reasons for students to study more than one program at once. Some of the outcomes of this study could only be based on assumptions, since no research does exist for this specific setting. Motivations to begin a double degree program may be similar to multiple enrollments (Borsetto & Saccon, 2022, 2023), but due to the circumstances of the Austrian university system differences can still be expected. Predictor variables such as student satisfaction should not be operationalized in a generalized manner, but for each enrolled study program. Other predictor variables for student performance as an outcome also need to be studied, since influences could not be identified for all performance indicators used in this study.

From the perspective of Austria's university system, this study creates first evidence that student behavior, satisfaction and success is different in students with one program and those with two or more programs. Parallel enrollments have not been assessed in official statistics (Federal Ministry of Education, Science and Research, 2024) and are not given specific consideration in country-wide (Federal Ministry of Education, Science and Research, 2022b) or university-wide (University of Graz, 2021) strategy documents. The results of this study suggest that students with more than one enrollment may have different needs than students in one program, as illustrated by the different results for student satisfaction. Since a third of all students at the University of Graz had more than one enrollment over the last decade, a targeted monitoring of this group and follow-up studies are warranted to be able to establish measures that address the differences observed in this study and increase both student satisfaction and performance.