Identifying the Tendency to Drop Out in Dual Studies Instrument (MISANDS)

Abstract

The Identifying the Tendency to Drop Out in Dual Studies Instrument (known in German as the Messinstrument zur Identifikation von Studienabbruchneigung im dualen Studium; MISANDS) is an 8-item psychometric assessment designed for the multidimensional measurement of attrition intention among university students enrolled in cooperative and dual study programs. Dual study programs combine tertiary higher education coursework with structured, contractual practical training within a partnering corporation or institutional training site. Developed by Ernst Deuer, Steffen Wild, and colleagues within the longitudinal multi-center research project “Studienverlauf – Weichenstellungen, Erfolgskriterien und Hürden im Verlauf des Studiums an der DHBW” at the Baden-Württemberg Cooperative State University (Duale Hochschule Baden-Württemberg; DHBW), the instrument adapts vocational drop-out screening models to capture the distinct structural duality of modern cooperative education. The MISANDS comprises two orthogonal yet correlated latent dimensions: (1) Study-Program-Related Drop-Out Tendency (studiengangsbezogene Abbruchneigung) and (2) Training-Site-Related Drop-Out Tendency (ausbildungsstättenbezogene Abbruchneigung). Each dimension is operationalized through four items evaluated on an authentic four-point forced-choice response format (1 = ja [yes], 2 = eher ja [rather yes], 3 = eher nein [rather no], 4 = nein [no]). Psychometric validation conducted on a large cohort of first-year dual bachelor students (N = 1,429) across business, engineering, and social work curricula confirmed a robust two-factor structure via exploratory factor analysis (explaining 73% of total variance) and confirmatory factor analysis (MLR estimation: χ² = 208.97, df = 15, CFI = .95, RMSEA = .09). Internal consistency estimates demonstrated good to excellent measurement precision, with Cronbach's alpha ranging from .86 to .88 and Raykov's rho composite reliability ranging from .87 to .88 across subscales. Longitudinal predictive validity was verified against official university administrative attrition records 22 months post-baseline, yielding receiver operating characteristic (ROC) area under the curve (AUC) values of .68 for the study-program dimension and .60 for the training-site dimension, accompanied by statistically significant point-biserial correlations with actual institutional drop-out (r_bis = .25 and .17, p ≤ .001). The MISANDS serves as an economical, psychometrically sound diagnostic tool suitable for institutional quality assurance, early drop-out warning systems, and individual academic counseling.

Keywords

MISANDS, dual studies, cooperative higher education, student drop-out, academic attrition, attrition intention, study satisfaction, workplace training site, institutional research, psychometrics

Authors

The MISANDS was developed and validated by researchers affiliated with the Baden-Württemberg Cooperative State University (DHBW) in Ravensburg, Germany:

• Prof. Dr. Ernst Deuer — Duale Hochschule Baden-Württemberg Ravensburg, Faculty of Business / Economics, Marktstraße 28, 88212 Ravensburg, Germany. Email: [email protected]
• Prof. Dr. Steffen Wild — Duale Hochschule Baden-Württemberg Ravensburg, Faculty of Business / Economics, Marktstraße 28, 88212 Ravensburg, Germany. Email: [email protected]
• Research Project Collaborators: Associated with the overarching DHBW research project “Studienverlauf – Weichenstellungen, Erfolgskriterien und Hürden im Verlauf des Studiums an der DHBW” (Deuer, Wild, Schäfer-Walkmann, Heide, & Walkmann, 2017).

Purpose

Academic attrition represents one of the most pressing challenges confronting modern higher education policy, institutional governance, and economic productivity. Longitudinal analyses by the German Centre for Higher Education Research and Science Studies (Deutsches Zentrum für Hochschul- und Wissenschaftsforschung; DZHW) consistently indicate that approximately 32% of bachelor's degree entrants at traditional universities and 25% at universities of applied sciences terminate their studies prematurely without obtaining a degree (Heublein & Schmelzer, 2018). Beyond the profound personal and emotional distress experienced by individual students, university drop-out generates substantial economic deadweight loss for public funding bodies and corporate training partners. Moreover, attrition rates increasingly serve as critical institutional key performance indicators (KPIs) in accreditation processes, quality audits, and performance-based resource allocation across tertiary institutions (Klein & Stocké, 2016).

Despite the proliferation of general student drop-out models, the specialized sector of dual or cooperative higher education presents a distinct pedagogical architecture that cannot be adequately assessed using traditional, single-environment instruments. Dual study programs fundamentally integrate two separate yet complementary learning venues: an academic higher education institution and an employment-based training company or clinical/social agency (Deuer & Träger, 2015). Under this system, students hold dual status as matriculated university scholars and contractually bound corporate employees. Attrition in this context is inherently bifurcated: a student may experience acute dissatisfaction, lack of fit, or disillusionment with the academic curriculum, with the workplace socialization and practical training demands, or with both domains simultaneously.

The primary purpose of the MISANDS is to provide higher education administrators, corporate trainers, organizational psychologists, and academic counselors with a standardized, psychometrically validated, and exceptionally brief screening tool capable of identifying latent drop-out risk during the earliest stages of academic progression. Empirical evidence underscores that the vast majority of university drop-outs occur during the first academic year (Chen, 2012; Willcoxsen, Cotter, & Joy, 2011). Early identification is paramount: by differentiating whether a student's propensity to terminate their studies stems from the university classroom (e.g., theoretical complexity, curriculum mismatch) or the corporate training environment (e.g., poor supervision, unsupportive workplace culture, excessive operational workload), targeted and tailored institutional interventions can be deployed before formal administrative exmatriculation takes place.

In addition to individual diagnostic applications, the MISANDS was engineered for large-scale institutional monitoring and programmatic quality assurance. It allows university governing bodies and partner enterprises to conduct systemic course evaluations, evaluate programmatic interventions, examine differences across academic faculties (e.g., engineering vs. business vs. social work), and detect organizational friction points between academic curricula and corporate practical phases.

Psychological Construct

The MISANDS operationalizes student attrition not as a spontaneous, binary behavioral event, but as a gradual, psychological withdrawal process that manifests through escalating cognitive intentions, evaluation of behavioral alternatives, and cumulative motivational deficits. Reflecting the structural reality of dual higher education, the instrument treats drop-out tendency as a multidimensional construct split into two distinct, correlated latent dimensions:

1. Study-Program-Related Drop-Out Tendency (studiengangsbezogene Abbruchneigung)

This subscale captures the student's psychological withdrawal from the chosen tertiary degree program and academic discipline. It reflects the degree of cognitive dissonance, dissatisfaction, and explicit intention to abandon or transfer out of the academic field of study. Academic attrition cognitions emerge when the individual perceives a fundamental mismatch between personal intellectual interests, self-efficacy, and academic competencies on the one hand, and the theoretical rigor, epistemological approach, curricular structure, or examination demands of the academic faculty on the other. Within the MISANDS, this latent dimension is tapped through four operational facets:

• Overall Academic Satisfaction: A global cognitive appraisal of one's contentment with the degree course (Item 1: “Sind Sie alles in allem mit Ihrem jetzigen Studiengang zufrieden?”).
• Hypothetical Decision Reiteration: A retrospective commitment gauge evaluating whether the student, given the choice again, would choose the identical academic program (Item 3: “Würden Sie Ihren jetzigen Studiengang wiederwählen?”).
• Historical Transfer Consideration: A lifetime cognitive probe evaluating whether the student has ever seriously contemplated switching their academic program (Item 5: “Haben Sie schon einmal daran gedacht, Ihren Studiengang zu wechseln?”).
• Acute Behavioral Transfer Intention: A present-focused, proximal intention marker reflecting an active, ongoing mental plan to exit or change the study course (Item 7: “Denken Sie aktuell daran, Ihren Studiengang zu wechseln?”).

2. Training-Site-Related Drop-Out Tendency (ausbildungsstättenbezogene Abbruchneigung)

This subscale captures the student's psychological alienation, vocational dissatisfaction, and turnover cognitions directed specifically toward their practical corporate training partner or practical agency. Because dual students operate under formal employment contracts and receive monthly stipends, their vocational experience mirrors that of corporate employees. In this domain, drop-out tendencies arise from factors such as organizational climate, perceived workplace injustice, inadequate mentoring, repetitive or non-challenging tasks, excessive overtime, or interpersonal conflicts with supervisors and colleagues. The four operational facets precisely mirror the academic subscale, ensuring structural parallelism across dimensions:

• Overall Training Site Satisfaction: A global cognitive evaluation of contentment with the host company or practical placement (Item 2: “Sind Sie alles in allem mit Ihrer jetzigen Ausbildungsstätte/Praxisstelle zufrieden?”).
• Hypothetical Employer Reiteration: A retrospective loyalty indicator evaluating whether the respondent would select the same training partner company again (Item 4: “Würden Sie Ihre jetzige Ausbildungsstätte/Praxisstelle wiederwählen?”).
• Historical Placement Transfer Consideration: A lifetime evaluation of whether the student has previously contemplated abandoning or switching their corporate training company (Item 6: “Haben Sie schon einmal daran gedacht, Ihre Ausbildungsstätte/Praxisstelle zu wechseln?”).
• Acute Placement Transfer Intention: An immediate, proximal turnover cognition reflecting active plans to terminate or change the current corporate training contract (Item 8: “Denken Sie aktuell daran, Ihre Ausbildungsstätte/Praxisstelle zu wechseln?”).

By separating academic dissatisfaction from workplace turnover cognitions, the construct allows for highly nuanced psychological profiles. For example, a student may exhibit zero drop-out intention regarding their engineering coursework but acute drop-out intention regarding an unsupportive manufacturing employer—or vice versa. This diagnostic differentiation is essential for targeted intervention, as institutional remedies for curriculum mismatch (e.g., tutoring, academic bridge courses) fundamentally differ from organizational workplace remedies (e.g., mediation with the corporate training manager, reassignment to an alternative partner company).

Theoretical Framework

The construction and psychometric validation of the MISANDS are grounded in three complementary theoretical paradigms drawn from higher education sociology, organizational psychology, and behavioral economics:

1. Interactionist Models of Student Attrition

The primary theoretical foundation rests upon established sociological and interactionist models of college student departure, most notably the longitudinal departure model articulated by Vincent Tinto (Tinto, 1975) and subsequent organizational extensions by John Bean (Bean, 1980), as well as contemporary European syntheses (Heublein & Wolter, 2011). These models posit that student retention and drop-out are not purely individual psychological traits nor entirely structural institutional artifacts; rather, they emerge from the dynamic, ongoing reciprocal interaction between the individual student (their motives, academic disposition, socio-demographic background, and psychological resilience) and the institutional environment (the academic, pedagogical, and social structures of the university).

Whereas historical higher education research has frequently investigated personal psychological predictors (e.g., conscientiousness, prior academic achievement; Richardson, Abraham, & Bond, 2012) in strict isolation from environmental learning characteristics (Schaeper & Weiß, 2016), the MISANDS adopts an explicitly interactionist framework. In a dual university ecosystem, the “institutional environment” is split into two co-equal entities: the academic classroom and the operational corporate environment. The student continuously navigates both arenas, striving for cognitive, vocational, and institutional integration.

2. Rational Action Theory and Cost-Benefit Maximization

To model the cognitive transition from diffuse dissatisfaction to an explicit decision to terminate studies, the scale developers incorporated principles of Rational Choice Theory (Becker, 1982; Harsanyi, 1976; Olson, 1968). In this framework, educational persistence is modeled as an ongoing utility-maximization calculus. Students continuously weigh the subjective costs (e.g., intellectual exertion, personal stress, foregone alternative career opportunities, financial pressures) against expected future returns (e.g., degree completion, prestigious employment, long-term earnings, career advancement).

When the subjective evaluation indicates that continuation costs decisively outweigh anticipated returns, the rational individual ceases persistence behavior and develops turnover cognitions. Items measuring historical transfer thoughts (Items 5 and 6), acute transfer intentions (Items 7 and 8), and hypothetical decision reiteration (Items 3 and 4) directly operationalize this rational reappraisal process, capturing the cognitive moment when alternative pathways become more attractive than the status quo.

3. Organizational Psychology and Job Satisfaction Models

Because dual students function as contracted corporate employees during their practical phases, the MISANDS incorporates foundational concepts from work and organizational psychology. Extensive organizational research confirms that job satisfaction exerts a profound causal influence on work performance (Judge, Thoresen, Bono, & Patton, 2001), somatic health (Fischer & Sousa-Poza, 2009), sickness absenteeism (Ybema, Smulders, & Bongers, 2010), and organizational turnover intention (Wright & Bonett, 2007). Work by Westermann and Heise (2018) demonstrates that academic study satisfaction shares substantial conceptual and empirical overlap with occupational job satisfaction.

Finally, the MISANDS builds directly upon earlier vocational training research conducted by Ernst Deuer (2006), who developed an early detection screening tool for apprenticeship attrition across the twin dimensions of occupation (Beruf) and enterprise (Betrieb). The MISANDS conceptually elevates this vocational apprenticeship model to the tertiary level, recognizing that dual university students encounter the exact same structural hurdles as apprentices—such as strict schedule adherence, dual institutional reporting, professional examination pressure, continuous workplace socialization, and rapid alternation between theoretical and practical environments (Deuer & Träger, 2015).

Validity

The validity of the MISANDS has been extensively examined across multiple empirical criteria using data collected from large student cohorts at the Baden-Württemberg Cooperative State University (DHBW):

Content Validity and Expert Review

Content validity was established through a multi-stage iterative development process. The original conceptual foundation was derived from Deuer's (2006) validated apprenticeship drop-out screener (N = 594). In 2016, during the first panel wave of the DHBW student cohort study (N = 5,838), an initial adaptation was piloted with three distinct structural modifications: re-labeling the dimensions from “enterprise/occupation” to “training site/degree program”, introducing a fourth item assessing lifetime transfer cognitions, and testing a 5-point satisfaction scale. Prior to the second panel wave (March 2017), an expert panel consisting of higher education researchers, psychometricians, and corporate training directors conducted a formal content review. To eliminate cognitive friction, reduce item difficulty, and minimize respondent fatigue (Porst, 2014), the panel harmonized all items onto an identical four-point response scale. The expert panel concluded that the eight final items thoroughly and exhaustively sampled the cognitive and evaluative universe of student attrition intentions across both educational venues.

Factorial and Construct Validity

Factorial validity was established in a sample of N = 1,429 first-year dual students through complementary exploratory and confirmatory factor analyses. Principal Component Analysis (PCA) with oblique (Oblimin) rotation unambiguously revealed two underlying dimensions matching the theoretical framework, with all items exhibiting high primary factor pattern loadings (≥ .77) and no problematic cross-loadings. Confirmatory factor analysis (CFA) executed via robust maximum likelihood (MLR) estimation verified this two-dimensional structure, yielding acceptable goodness-of-fit parameters (χ² = 208.97, df = 15, p < .001; CFI = .95; RMSEA = .09 [90% CI: .08–.11]). Latent factor correlations between the study-program and training-site dimensions were moderate and statistically significant (Ψ = .37, p < .001), demonstrating that while the two venues share common variance within the broader dual study experience, they remain empirically distinct constructs that must not be collapsed into a single unidimensional score.

Predictive and Criterion-Related Validity

A major strength of the MISANDS is that its criterion-related validity was not evaluated solely against self-reported, subjective proxy variables, but against actual administrative drop-out records retrieved from the university's central administration database (DUALIS) 22 months post-baseline (January 2019). Drop-out was officially defined as any formal exmatriculation without successful degree completion (Deuer et al., 2017).

To evaluate diagnostic accuracy across varying clinical cut-offs, Receiver Operating Characteristic (ROC) analyses were performed (Zhou, Obuchowski, & McClish, 2011). The Area Under the Curve (AUC) reached .68 for the Study-Program-Related Drop-Out Tendency subscale and .60 for the Training-Site-Related Drop-Out Tendency subscale. In psychometric and epidemiological screening contexts, an AUC of .68 over a nearly two-year predictive time horizon represents an acceptable, statistically meaningful diagnostic signal, especially considering that actual academic departure is subject to complex external contingencies (e.g., sudden financial crises, family events, private relocations) occurring long after the initial psychometric measurement.

Furthermore, point-biserial correlations (r_bis) computed between the continuous manifest subscale scores at baseline and the binary exmatriculation criterion (0 = retained student, 1 = drop-out) revealed statistically significant positive associations: r_bis = .25 (p ≤ .001) for the study-program scale and r_bis = .17 (p ≤ .001) for the training-site scale. These findings provide empirical confirmation of the scale's prognostic capacity.

Reliability

The reliability of the MISANDS was evaluated on the complete first-year validation cohort (N = 1,429) using classical internal consistency estimates as well as structural equation modeling composite reliability coefficients.

Internal Consistency

Classical Cronbach's alpha (α) demonstrated strong internal consistency across both four-item subscales:

• Study-Program-Related Drop-Out Tendency: α = .86
• Training-Site-Related Drop-Out Tendency: α = .88

Composite Reliability (Raykov's Rho)

As psychometric methodologists have rigorously demonstrated (Cortina, 1993; Raykov, 1997), Cronbach's alpha presupposes essential tau-equivalence (identical factor loadings across all items). When measurement models are tau-congeneric—meaning manifest indicators exhibit varying factor loadings (λ) and error variances—alpha can yield biased, typically conservative underestimates of true scale reliability. To address this structural reality, the authors computed Raykov's rho (ρ) within a structural equation modeling framework:

• Study-Program-Related Drop-Out Tendency: ρ = .87
• Training-Site-Related Drop-Out Tendency: ρ = .88

These composite reliability coefficients confirm that both subscales possess high measurement precision, exceeding the conventional threshold of .80 recommended for basic research and approaching the .90 benchmark required for high-stakes individual diagnostics and counseling recommendations.

Item-Total Correlations and Discriminative Power

Corrected item-total correlations (discriminative power / Trennschärfe) for all eight manifest indicators were exceptionally high, ranging from .64 to .77. On the study-program subscale, item-total correlations were .64 (Item 1), .71 (Item 3), .76 (Item 5), and .75 (Item 7). On the training-site subscale, item-total correlations were .71 (Item 2), .74 (Item 4), .77 (Item 6), and .74 (Item 8). No item detracted from the overall scale reliability; iterative deletion of any indicator resulted in a reduction of the corresponding subscale alpha, confirming that all four items per factor contribute essentially to latent trait measurement.

Factor Analysis

The factorial architecture of the MISANDS was comprehensively examined using both exploratory and confirmatory psychometric procedures on the calibration sample of N = 1,429 first-year students.

Exploratory Factor Analysis (EFA) / Principal Component Analysis (PCA)

Prior to extraction, sampling adequacy and data factorability were confirmed through formal statistical checks. The Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy reached .79, indicating substantial shared variance across indicators, while Bartlett's Test of Sphericity yielded a highly significant result (χ² reaching statistical significance at p ≤ .000), rejecting the identity matrix null hypothesis.

A Principal Component Analysis was conducted specifying oblique (Oblimin) rotation. Oblique rotation was theoretically mandated because the workplace training site and the academic study course constitute integrated, overlapping components of the overarching cooperative study experience. The scree plot clearly indicated a two-component structure, characterized by a sharp drop after the second eigenvalue. The initial eigenvalues were 3.95 for Component 1, 1.86 for Component 2, and well below 1.0 for all subsequent components. Together, these two components accounted for an impressive 73% of the total variance.

All eight items displayed high communalities (h² ≥ .62). Factor pattern loadings after Oblimin rotation demonstrated clear, simple structure, with primary loadings of a ≥ .77 and negligible cross-loadings, perfectly segregating the four study-program items onto the first factor and the four training-site items onto the second factor.

Confirmatory Factor Analysis (CFA)

To confirm the two-dimensional model, a confirmatory factor analysis was conducted using Mplus 7.1 (Muthén & Muthén, 2012). Because manifest response distributions exhibited significant deviations from multivariate normality (pronounced positive skewness and high kurtosis), parameter estimation was performed using the Robust Maximum Likelihood (MLR) estimator with Huber-White sandwich standard errors.

In accordance with advanced structural equation modeling practices for parallel multi-trait multi-method battery designs (Urban & Mayerl, 2014), residual covariance parameters (θ) were freed between identically formulated item pairs across the two domains (e.g., correlating the residual error of Item 1 [overall program satisfaction] with Item 2 [overall training site satisfaction], and Item 7 with Item 8). These residual correlations were modest, ranging between θ = .06 and θ = .08.

The hypothesized two-factor CFA model yielded an acceptable global fit to the empirical data:

• Chi-Square (χ²): 208.97 (df = 15, p < .001)
• Comparative Fit Index (CFI): .95
• Root Mean Square Error of Approximation (RMSEA): .09 (90% Confidence Interval: .08–.11)
• Standardized Factor Loadings (λ): All manifest path coefficients were large and statistically significant, with λ ≥ .69 across all indicators.
• Standardized Residual Variances (ε): All item measurement residuals remained small (ε ≤ .53).
• Latent Factor Intercorrelation (Ψ): The latent correlation between the two drop-out tendency dimensions was moderate and significant (Ψ = .37, p < .001).

Instrument / Measurement Tool

The operational characteristics and administrative guidelines for the MISANDS are structured as follows:

• Test Type: Standardized self-report psychometric rating scale / behavioral intention inventory.
• Application Format: Optimized for computerized/online administrative surveys, institutional student panels, and paper-and-pencil evaluations.
• Target Population: Students enrolled in dual, cooperative, work-integrated, or apprenticeship-based higher education programs. (Note: The four-item Study-Program-Related Drop-Out Tendency subscale can be administered independently to students at traditional universities and universities of applied sciences).
• Administration Duration: Highly economical, requiring approximately 2 to 3 minutes for completion by the student and under 5 minutes for automated scoring and interpretation.
• Number of Items: 8 items organized into two parallel 4-item subscales.
• Subscale Architecture:
  • Study-Program-Related Drop-Out Tendency (studiengangsbezogene Abbruchneigung): Items 1, 3, 5, and 7.
  • Training-Site-Related Drop-Out Tendency (ausbildungsstättenbezogene Abbruchneigung): Items 2, 4, 6, and 8.
• Authentic Response Scale: A mandatory four-point forced-choice Likert-type response format: 1 = ja (yes), 2 = eher ja (rather yes), 3 = eher nein (rather no), 4 = nein (no). No neutral midpoint or “undecided” option is provided.
• Item Polarity and Recoding Rules:
  • Positively Polared Items (Items 1, 2, 3, 4): These items measure contentment and hypothetical re-selection. To ensure that higher numerical values reflect higher drop-out tendency, these items are scored: 1 = ja → 1 point, 2 = eher ja → 2 points, 3 = eher nein → 3 points, 4 = nein → 4 points. (Answering “nein” reflects maximum dissatisfaction, hence maximum attrition propensity).
  • Negatively Polared Items (Items 5, 6, 7, 8): These items directly assess past or current transfer intentions. To align their direction with drop-out propensity, they are reversed: 1 = ja → 4 points, 2 = eher ja → 3 points, 3 = eher nein → 2 points, 4 = nein → 1 point. (Answering “ja” reflects immediate transfer intention, hence maximum attrition propensity).
• Score Calculation: Subscales must be scored and analyzed separately. For each four-item subscale, compute the unweighted arithmetic mean of the four recoded item scores. The resulting subscale index ranges continuously from 1.00 (lowest possible drop-out tendency / complete institutional commitment) to 4.00 (highest possible drop-out tendency / acute attrition risk).
• Missing Data Handling: Due to potential non-random attrition (students with severe drop-out tendencies are disproportionately prone to survey non-response, indicating Missing Not at Random [MNAR] mechanisms), imputation is only recommended if missing values do not exceed 30% of scale indicators under MAR/MCAR assumptions (Lüdtke et al., 2007; Rost, 2013). In institutional online surveys, configuring all 8 items as mandatory response fields is strongly advised.
• Descriptive Norms (Calibration Cohort, N = 1,429):
  • Study-Program Subscale: Mean = 1.57 (SD = 0.66), Skewness = +1.46, Kurtosis = +1.82.
  • Training-Site Subscale: Mean = 1.50 (SD = 0.70), Skewness = +1.63, Kurtosis = +2.10.
  • Both subscales display pronounced right-skewed, leptokurtic distributions, reflecting that the healthy majority of students report low drop-out intentions, whereas elevated scores identify a critical minority requiring administrative or psychological intervention.

Permissions & Fee and Test Year

The MISANDS was developed between 2016 and 2017 as part of the Baden-Württemberg Cooperative State University research project “Studienverlauf – Weichenstellungen, Erfolgskriterien und Hürden im Verlauf des Studiums an der DHBW” (Deuer et al., 2017). Formal psychometric validation and predictive modeling using 22-month longitudinal follow-up records were published in institutional research reports in 2017 and 2018 (Deuer & Wild, 2017, 2018). The scale is made freely accessible for academic, institutional, and scientific research without licensing fees or royalties. Higher education researchers, institutional research units, and academic counselors may utilize and adapt the instrument for non-commercial educational and diagnostic purposes, provided appropriate scholarly attribution is accorded to the original authors (Ernst Deuer and Steffen Wild) and the Baden-Württemberg Cooperative State University (DHBW).

References

• Bean, J. P. (1980). Dropouts and turnover: The synthesis and test of a causal model of student attrition. Research in Higher Education, 12(2), 155–187. https://doi.org/10.1007/BF00976194
• Becker, G. S. (1982). Der ökonomische Ansatz zur Erklärung menschlichen Verhaltens. Mohr Siebeck.
• Blüthmann, I., Thiel, F., & Wolfgram, M. (2011). Studienabbruch und Studienfachwechsel an der Freien Universität Berlin. Freie Universität Berlin.
• Burger, C., & Groß, E. (2016). Studienabbruch und Studienfachwechsel: Eine explorative Studie zu den Ursachen an der Universität Wien. Zeitschrift für Hochschulentwicklung, 11(2), 221–237. https://doi.org/10.3217/zfhe-11-02/14
• Chen, R. (2012). Institutional risk factors in student dropout: Investigating the effects of student-faculty ratio and institutional expenditure. The Journal of Higher Education, 83(1), 1–30. https://doi.org/10.1080/00221546.2012.11777232
• Cortina, J. M. (1993). What is coefficient alpha? An examination of theory and applications. Journal of Applied Psychology, 78(1), 98–104. https://doi.org/10.1037/0021-9010.78.1.98
• Deuer, E. (2006). Früherkennung von Ausbildungsabbrüchen: Modell zur Identifikation gefährdeter Jugendlicher. VDM Verlag Dr. Müller.
• Deuer, E., & Träger, H. (2015). Das duale Studium: Definition, Entwicklung und Perspektiven. In E. Deuer & H. Träger (Eds.), Duales Studium: Herausforderungen, Chancen und Gestaltungsmöglichkeiten (pp. 9–32). Springer VS. https://doi.org/10.1007/978-3-658-09516-1_1
• Deuer, E., & Wild, S. (2017). Die Messung der Abbruchneigung im Rahmen der ersten Erhebungswelle des DHBW-Studierendenpanels – Begründung und Entwicklung eines Instruments zur Früherkennung von Studienabbrüchen (Forschungsbericht 2/2017). Duale Hochschule Baden-Württemberg.
• Deuer, E., & Wild, S. (2018). Validierung eines Instruments zur Erfassung der Studienabbruchneigung bei dual Studierenden (Forschungsbericht 4/2018). Duale Hochschule Baden-Württemberg.
• Deuer, E., Wild, S., Schäfer-Walkmann, S., Heide, K., & Walkmann, N. (2017). Studienverlauf – Weichenstellungen, Erfolgskriterien und Hürden im Verlauf des Studiums an der DHBW: Erster Projektbericht. Duale Hochschule Baden-Württemberg.
• Fischer, J. A. V., & Sousa-Poza, A. (2009). Does job satisfaction improve the health of workers? New evidence using panel data and objective measures of health. Health Economics, 18(1), 71–89. https://doi.org/10.1002/hec.1346
• Georg, W. (2008). Ursachen des Studienabbruchs: Eine empirische Untersuchung über Ursachen und Hintergründe des Studienabbruchs an Schweizer Universitäten. Eidgenössische Technische Hochschule Zürich.
• Harsanyi, J. C. (1976). Essays on ethics, social behavior, and scientific explanation. D. Reidel Publishing Company. https://doi.org/10.1007/978-94-010-1533-2
• Heublein, U., & Schmelzer, R. (2018). Die Entwicklung der Studienabbruchquoten an den deutschen Hochschulen: Berechnungen auf Basis des Absolventenjahrgangs 2016 (DZHW Brief 04/2018). Deutsches Zentrum für Hochschul- und Wissenschaftsforschung.
• Heublein, U., & Wolter, A. (2011). Studienabbruch in Deutschland: Definition, Häufigkeit, Ursachen, Maßnahmen. Zeitschrift für Pädagogik, 57(2), 214–236.
• Judge, T. A., Thoresen, C. J., Bono, J. E., & Patton, G. K. (2001). The job satisfaction–job performance relationship: A qualitative and quantitative review. Psychological Bulletin, 127(3), 376–407. https://doi.org/10.1037/0033-2909.127.3.376
• Klein, O., & Stocké, V. (2016). Determinanten des Studienabbruchs: Individuelle und institutionelle Faktoren. Kölner Zeitschrift für Soziologie und Sozialpsychologie, 68(4), 589–615. https://doi.org/10.1007/s11577-016-0391-4
• Lüdtke, O., Robitzsch, A., Trautwein, U., & Köller, O. (2007). Umgang mit fehlenden Werten in der psychologischen und erziehungswissenschaftlichen Forschung: Ein Überblick über moderne Verfahren. Psychologische Rundschau, 58(2), 103–117. https://doi.org/10.1026/0033-3042.58.2.103
• Muthén, L. K., & Muthén, B. O. (2012). Mplus user’s guide (7th ed.). Muthén & Muthén.
• Olson, M. (1968). The logic of collective action: Public goods and the theory of groups. Harvard University Press.
• Porst, R. (2014). Fragebogen: Ein Arbeitsbuch (4th ed.). Springer VS. https://doi.org/10.1007/978-3-658-02118-4
• Raykov, T. (1997). Estimation of composite reliability for congeneric measures. Applied Psychological Measurement, 21(2), 173–184. https://doi.org/10.1177/01466216970212006
• Richardson, M., Abraham, C., & Bond, R. (2012). Psychological correlates of university students’ academic performance: A systematic review and meta-analysis. Psychological Bulletin, 138(2), 353–387. https://doi.org/10.1037/a0026838
• Rost, J. (2013). Testtheorie und Testkonstruktion (3rd ed.). Huber.
• Schaeper, H., & Weiß, T. (2016). Studieneingangsphase und Studienerfolg: Befunde aus dem DZHW-Studierendenpanel. Deutsches Zentrum für Hochschul- und Wissenschaftsforschung.
• Tinto, V. (1975). Dropout from higher education: A theoretical synthesis of recent research. Review of Educational Research, 45(1), 89–125. https://doi.org/10.3102/00346543045001089
• Urban, D., & Mayerl, J. (2014). Strukturgleichungsmodellierung: Ein Resümee zur praktischen Anwendung. Springer VS. https://doi.org/10.1007/978-3-658-01908-2
• Westermann, R., & Heise, E. (2018). Arbeits- und Studienzufriedenheit: Ein Vergleich konzeptueller Modelle und empirischer Befunde. Zeitschrift für Arbeits- und Organisationspsychologie, 62(1), 14–29. https://doi.org/10.1026/0932-4089/a000258
• Willcoxsen, L., Cotter, J., & Joy, S. (2011). Beyond the first-year experience: The impact on attrition of student experiences throughout undergraduate degree studies in six diverse faculties. Studies in Higher Education, 36(3), 331–352. https://doi.org/10.1080/03075070903581533
• Wright, T. A., & Bonett, D. G. (2007). Job satisfaction and psychological well-being as nonadditive predictors of turnover. Journal of Applied Psychology, 92(2), 541–560. https://doi.org/10.1037/0021-9010.92.2.541
• Ybema, J. F., Smulders, P. G., & Bongers, P. M. (2010). Antecedents and consequences of employee absenteeism: A longitudinal perspective on the role of job satisfaction. Social Indicators Research, 98(3), 437–452. https://doi.org/10.1007/s11205-009-9550-3
• Zhou, X. H., Obuchowski, N. A., & McClish, D. K. (2011). Statistical methods in diagnostic medicine (2nd ed.). John Wiley & Sons. https://doi.org/10.1002/9780470906514

Items of the Scale