Skip to content
Licensed Unlicensed Requires Authentication Published by De Gruyter Oldenbourg February 11, 2020

Early Prediction of University Dropouts – A Random Forest Approach

Andreas Behr, Marco Giese, Herve D. Teguim K and Katja Theune

Abstract

We predict university dropout using random forests based on conditional inference trees and on a broad German data set covering a wide range of aspects of student life and study courses. We model the dropout decision as a binary classification (graduate or dropout) and focus on very early prediction of student dropout by stepwise modeling students’ transition from school (pre-study) over the study-decision phase (decision phase) to the first semesters at university (early study phase). We evaluate how predictive performance changes over the three models, and observe a substantially increased performance when including variables from the first study experiences, resulting in an AUC (area under the curve) of 0.86. Important predictors are the final grade at secondary school, and also determinants associated with student satisfaction and their subjective academic self-concept and self-assessment. A direct outcome of this research is the provision of information to universities wishing to implement early warning systems and more personalized counseling services to support students at risk of dropping out during an early stage of study.

JEL Classification: I23

Acknowledgements

This work was supported by the German Federal Ministry of Education and Research under Grant number 01PX16006.

References

Aggarwal, C.C. (2015), Data Mining: The Textbook, volume 1. Springer Science & Business Media, Cham.Search in Google Scholar

Aina, C. (2013), Parental Background and University Dropout in Italy. Higher Education 65 (4): 437–456.10.1007/s10734-012-9554-zSearch in Google Scholar

Assaad, R., C. Krafft, S. Yassin (2018), Comparing Retrospective and Panel Data Collection Methods to Assess Labor Market Dynamics. IZA Journal of Development and Migration 8 (17).10.1186/s40176-018-0125-7Search in Google Scholar

Baraldi, A.N., C.K. Enders (2009). An Introduction to Modern Missing Data Analyses. Journal of School Psychology 48 (1): 5–37.10.1016/j.jsp.2009.10.001Search in Google Scholar

Behr, A. (2006), Comparing Estimation Strategies for Income Equations in the Presence of Panel Attrition. Jahrbücher für Nationalökonomie und Statistik 226 (4): 361–384.10.1515/jbnst-2006-0403Search in Google Scholar

Behr, A., E. Bellgardt, U. Rendtel (2005), Extent and Determinants of Panel Attrition in the European Community Household Panel. European Sociological Review 21 (5): 489–512.10.1093/esr/jci037Search in Google Scholar

Belloc, F., A. Maruotti, L. Petrella (2010), University Drop-Out: An Italian Experience. Higher Education 60 (2): 127–138.10.1007/s10734-009-9290-1Search in Google Scholar

Berens, J., K. Schneider, S. Görtz, S. Oster, and J. Burghoff (2018), Early Detection of Students at Risk – Predicting Student Dropouts Using Administrative Student Data and Machine Learning Methods. CESifo Working Papers No. 7259.10.2139/ssrn.3275433Search in Google Scholar

Blossfeld, H.-P., H.-G. Roßbach, J. von Maurice (2011), Education as a Lifelong Process–The German National Educational Panel Study (NEPS). Zeitschrift für Erziehungswissenschaft [Special Issue] 14. https://www.neps-data.de/Datenzentrum/Daten-und-Dokumentation/Startkohorte-Studierende/Daten-und-Zitation.10.1007/s11618-011-0179-2Search in Google Scholar

Brandstätter, H., L. Grillich, A. Farthofer (2006), Prognose des Studienabbruchs. Zeitschrift für Entwicklungspsychologie und Pädagogische Psychologie 38 (3): 121–131.10.1026/0049-8637.38.3.121Search in Google Scholar

Breiman, L. (2001), Random Forests. Machine Learning 45 (1): 5–32.10.1023/A:1010933404324Search in Google Scholar

Breiman, L., A. Cutler (2004), Random Forests. http://www.math.usu.edu/~adele/forests/cc_home.htm.Search in Google Scholar

Breiman, L., J. Friedman, C.J. Stone, R.A. Olshen (1984), Classification and Regression Trees. Chapman & Hall/CRC, Boca Raton.Search in Google Scholar

Chen, J.J., C. Tsai, H. Moon, H. Ahn, J.J. Young, and C.-H. Chen (2006), Decision Threshold Adjustment in Class Prediction. SAR and QSAR in Environmental Research 17 (3): 337–352.10.1080/10659360600787700Search in Google Scholar

Dekker, G., M. Pechenizkiy, J. Vleeshouwers (2009), Predicting Students Drop Out: A Case Study. Proceedings of the 2nd International Conference on Educational Data Mining, 41–50.Search in Google Scholar

Di Pietro, G., A. Cutillo (2008), Degree Flexibility and University Drop-Out: The Italian Experience. Economics of Education Review 27 (5): 546–555.10.1016/j.econedurev.2007.06.002Search in Google Scholar

Elkan, C. (2001), The Foundations of Cost-Sensitive Learning. Proceedings of the Seventeenth International Joint Conference on Artificial Intelligence, 973–978.Search in Google Scholar

Georg, W. (2009), Individuelle und institutionelle Faktoren der Bereitschaft zum Studienabbruch: eine Mehrebenenanalyse mit Daten des Konstanzer Studierendensurveys. Zeitschrift für Soziologie der Erziehung und Sozialisation 28 (2): 191–206.Search in Google Scholar

Ghignoni, E. (2017), Family Background and University Dropouts During the Crisis: The Case of Italy. Higher Education 73 (1): 127–151.10.1007/s10734-016-0004-1Search in Google Scholar

Han, J., M. Kamber, J. Pei (2011), Data Mining: Concepts and Techniques. Elsevier (third edition), Amsterdam.Search in Google Scholar

Hapfelmeier, A., T. Hothorn, K. Ulm, and C. Strobl (2014), A New Variable Importance Measure for Random Forests with Missing Data. Statistics and Computing 24 (1): 21–34.10.1007/s11222-012-9349-1Search in Google Scholar

Hastie, T., R. Tibshirani, J. Friedman (2009), The Elements of Statistical Learning. Springer, New York.10.1007/978-0-387-84858-7Search in Google Scholar

Hetze, P. (2011), Nachhaltige Hochschulstrategien für mehr MINT-Absolventen. Stifterverband für die Deutsche Wissenschaft.Search in Google Scholar

Heublein, U. (2014), Student Drop-Out from German Higher Education Institutions. European Journal of Education 49 (4): 497–513.10.1111/ejed.12097Search in Google Scholar

Heublein, U., J. Ebert, C. Hutzsch, S. Isleib, R. König, J. Richter, and A. Woisch (2017), Zwischen Studienerwartungen und Studienwirklichkeit. Forum Hochschule 1/2017.Search in Google Scholar

Heublein, U., C. Hutzsch, J. Schreiber, D. Sommer, G. Besuch (2010), Ursachen des Studienabbruchs in Bachelor-und in herkömmlichen Studiengängen – Ergebnisse einer bundesweiten Befragung von Exmatrikulierten des Studienjahres 2007/08. HIS: Forum Hochschule 2/2010.Search in Google Scholar

Heublein, U., J. Richter, R. Schmelzer, and D. Sommer (2014), Die Entwicklung der Studienabbruchquoten an den deutschen Hochschulen: Statistische Berechnungen auf der Basis des Absolventenjahrgangs 2012. Forum Hochschule 4/2014.Search in Google Scholar

Heublein, U., R. Schmelzer, D. Sommer, and J. Wank (2012), Die Entwicklung der Schwund- und Studienabbruchquoten an den deutschen Hochschulen. In HIS: Forum Hochschule, volume 3, 2012.Search in Google Scholar

Hoffait, A.-S., M. Schyns (2017), Early Detection of University Students with Potential Difficulties. Decision Support Systems 101: 1–11.10.1016/j.dss.2017.05.003Search in Google Scholar

Hothorn, T., K. Hornik, C. Strobl, A. Zeileis (2018), Package ‘Party’: A Laboratory for Recursive Partytioning. Package Reference Manual for Party Version 1.3-0 16.Search in Google Scholar

Hothorn, T., K. Hornik, A. Zeileis (2006), Unbiased Recursive Partitioning: A Conditional Inference Framework. Journal of Computational and Graphical Statistics 15 (3): 651–674.10.1198/106186006X133933Search in Google Scholar

Hovdhaugen, E., P.O. Aamodt (2009), Learning Environment: Relevant or Not to Students’ Decision to Leave University? Quality in Higher Education 15 (2): 177–189.10.1080/13538320902995808Search in Google Scholar

Isphording, I., F. Wozny (2018), Ursachen des Studienabbruchs – eine Analyse des Nationalen Bildungspanels. IZA Research Report No. 82.Search in Google Scholar

James, G., D. Witten, T. Hastie, R. Tibshirani (2013), An Introduction to Statistical Learning. Springer, New York.10.1007/978-1-4614-7138-7Search in Google Scholar

Johnes, J. (1990), Determinants of Student Wastage in Higher Education. Studies in Higher Education 15 (1): 87–99.10.1080/03075079012331377611Search in Google Scholar

Johnes, J., J. Taylor (1989), Undergraduate Non-completion Rates: Differences between UK universities. Higher Education 18 (2): 209–225.10.1007/BF00139181Search in Google Scholar

Kemper, L., G. Vorhoff, B.U. Wigger (2019), Predicting Student Dropout: a Machine Learning Approach. European Journal of Higher Education, forthcoming.10.1080/21568235.2020.1718520Search in Google Scholar

Kovacic, Z. (2010), Early Prediction of Student Success: Mining Students’ Enrolment Data. Proceedings of Informing Science & IT Education Conference 647–665.10.28945/1281Search in Google Scholar

Larsen, M.R., H.B. Sommersel, M.S. Larsen (2013a), Evidence on Dropout Phenomena at Universities. Danish Clearinghouse for Educational Research, Copenhagen.Search in Google Scholar

Larsen, M.S., K.P. Kornbeck, R. Kristensen, M.R. Larsen, H.B. Sommersel (2013b), Dropout Phenomena at Universities: What is Dropout? Why does Dropout Occur? What Can be Done by the Universities to Prevent or Reduce it? Danish Clearinghouse for Educational Research – Research Series 15.Search in Google Scholar

Lassibille, G., L. Navarro Gómez (2008), Why Do Higher Education Students Drop Out? Evidence from Spain. Education Economics 16 (1): 89–105.10.1080/09645290701523267Search in Google Scholar

LIfBi (2017), Startkohorte 5: Studierende (SC5) – Studienübersicht Wellen 1 bis 9. Technical report, Leibniz Institut für Bildungsverläufe e.V.Search in Google Scholar

Montmarquette, C., S. Mahseredjian, R. Houle (2001). The Determinants of University Dropouts: A Bivariate Probability Model with Sample Selection. Economics of Education Review 20 (5): 475–484.10.1016/S0272-7757(00)00029-7Search in Google Scholar

Müller, S., T. Schneider. (2013), Educational Pathways and Dropout from Higher Education in Germany. Longitudinal and Life Course Studies 4 (3): 218–241.Search in Google Scholar

Plackett, R.L. (1983), Karl Pearson and the Chi-squared Test. International Statistical Review/Revue Internationale de Statistique 51: 59–72.10.2307/1402731Search in Google Scholar

Prussog-Wagner, A., T. Weiß, F. Aust, F. Turri (2016), Methodenbericht: NEPS-Startkohorte 5 – CATI-Haupterhebung Sommer 2016 B112. Technical report, Leibniz Institut für Bildungsverläufe e.V.Search in Google Scholar

R Core Team (2018), R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria.Search in Google Scholar

Rice, J.A. (2006), Mathematical Statistics and Data Analysis. Thomson Brooks/ Cole, Duxbury.Search in Google Scholar

Romero, C., S. Ventura (2010), Educational Data Mining: A Review of the State of the Art. IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews) 40 (6): 601–618.10.1109/TSMCC.2010.2053532Search in Google Scholar

Romero, C., S. Ventura (2013). Data Mining in Education. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery 3 (1): 12–27.10.1002/widm.1075Search in Google Scholar

Sarcletti, A., S. Müller (2011), Zum Stand der Studienabbruchforschung. Theoretische Perspektiven, zentrale Ergebnisse und methodische Anforderungen an künftige Studien. Zeitschrift für Bildungsforschung 1 (3): 235–248.Search in Google Scholar

Schiefele, U., L. Streblow, J. Brinkmann (2007), Aussteigen oder Durchhalten. Zeitschrift für Entwicklungspsychologie und Pädagogische Psychologie 39 (3): 127–140.10.1026/0049-8637.39.3.127Search in Google Scholar

Severiens, S., G. Ten Dam (2012), Leaving College: A Gender Comparison in Male and Female-Dominated Programs. Research in Higher Education 53 (4): 453–470.10.1007/s11162-011-9237-0Search in Google Scholar

Siri, A. (2015), Predicting Students’ Dropout at University Using Artificial Neural Networks. Italian Journal of Sociology of Education 7 (2): 225–247.Search in Google Scholar

Smith, J.P., R.A. Naylor (2001), Dropping Out of University: A Statistical Analysis of the Probability of Withdrawal for UK University Students. Journal of the Royal Statistical Society: Series A (Statistics in Society) 164 (2): 389–405.10.1111/1467-985X.00209Search in Google Scholar

Stinebrickner, R., T. Stinebrickner (2014), Academic Performance and College Dropout: Using Longitudinal Expectations Data to Estimate a Learning Model. Journal of Labor Economics 32 (3): 601–644.10.1086/675308Search in Google Scholar

Strobl, C., A.-L. Boulesteix, T. Kneib, T. Augustin, and A. Zeileis (2008). Conditional Variable Importance for Random Forests. BMC Bioinformatics 9 (307).10.1186/1471-2105-9-307Search in Google Scholar

Strobl, C., A.-L. Boulesteix, A. Zeileis, T. Hothorn (2007). Bias in Random Forest Variable Importance Measures: Illustrations, Sources and a Solution. BMC Bioinformatics 8 (25).10.1186/1471-2105-8-25Search in Google Scholar

Suhre, C.J., E.P. Jasen, E.G. Harskamp (2007). Impact of Degree Program Satisfaction on the Persistence of College Students. Higher Education 54 (2): 207–226.10.1007/s10734-005-2376-5Search in Google Scholar

Svetnik, V., A. Liaw, C. Tong, J.C. Culberson, R.P. Sheridan, B.P. Feuston (2003), Random Forest: A Classification and Regression Tool for Compound Classification and QSAR Modeling. Journal of Chemical Information and Computer Sciences 43 (6): 1947–1958.10.1021/ci034160gSearch in Google Scholar

Tinto, V. (1975), Dropout from Higher Education: A Theoretical Synthesis of Recent Research. Review of Educational Research 45 (1): 89–125.10.3102/00346543045001089Search in Google Scholar

Tinto, V. (1993), Leaving College: Rethinking the Causes and Cures of Student Attrition. Chicago: Chicago University Press (second edition).10.7208/chicago/9780226922461.001.0001Search in Google Scholar

Twala, B. (2009), An Empirical Comparison of Techniques for Handling Incomplete Data Using Decision Trees. Applied Artificial Intelligence 23 (5): 373–405.10.1080/08839510902872223Search in Google Scholar

Ulriksen, L., L.M. Madsen, H.T. Holmegaard (2010), What Do We Know About Explanations for Drop Out/Opt Out Among Young People from STM Higher Education Programmes? Studies in Science Education 46 (2): 209–244.10.1080/03057267.2010.504549Search in Google Scholar

Van Bragt, C.A., A.W. Bakx, T.C. Bergen, M.A. Croon (2011a). Looking for Students Personal Characteristics Predicting Study Outcome. Higher Education 61 (1): 59–75.10.1007/s10734-010-9325-7Search in Google Scholar

Van Bragt, C.A., A.W. Bakx, P.J. Teune, T.C. Bergen, M.A. Croon (2011b). Why Students Withdraw or Continue Their Educational Careers: A Closer Look at Differences in Study Approaches and Personal Reasons. Journal of Vocational Education and Training 63 (2): 217–233.10.1080/13636820.2011.567463Search in Google Scholar

Vandamme, J.-P., N. Meskens, J.-F. Superby (2007), Predicting Academic Performance by Data Mining Methods. Education Economics 15 (4): 405–419.10.1080/09645290701409939Search in Google Scholar

Vogler-Ludwig, K., N. Düll, B. Kriechel (2016), Arbeitsmarkt 2030 – Wirtschaft und Arbeitsmarkt im digitalen Zeitalter. W. Bertelsmann Verlag, Bielefeld.Search in Google Scholar

Vossensteyn, H., B. Stensaker, A. Kottmann, E. Hovdhaugen, B. Jongbloed, S. Wollscheid, F. Kaiser, L. Cremonini (2015). Dropout and Completion in Higher Education in Europe. Luxembourg: Publications Office of the European Union.Search in Google Scholar

Weerasinghe, I.S., R. Lalitha, S. Fernando (2017), Students’ Satisfaction in Higher Education Literature Review. American Journal of Educational Research 5 (5): 533–539.Search in Google Scholar

Yathongchai, W., C. Yathongchai, K. Kerdprasop, and N. Kerdprasop (2012), Factor Analysis with Data Mining Technique in Higher Educational Student Drop Out. Latest Advances in Educational Technologies, 111–116.Search in Google Scholar

Zinn, S., H.W. Steinhauer, C. Aßmann (2017), Samples, Weights, and Nonresponse: The Student Sample of the National Educational Panel Study (Wave 1 to 8) (NEPS Survey Paper No. 18). Technical report, Leibniz Institut für Bildungsverläufe e.V.Search in Google Scholar

Appendix

Table 11:

Participants, temporary leavers, last participation, and final panel leavers in the current SUF (LIfBi, 2017, and own calculations).

waveinstrumentpartic. surveytemp. leaverslast partic.final panel leavers
1stCATI (+test)17,91001,2991299
2ndCAWI12,2735,591594594
3rdCATI13,1134,560561561
4thCAWI11,2026,424638638
5thCATI (+test)12,6943,444765765
6thCAWI10,1837,0391,0411,041
7thCATI (+test)9,5477,161774138
8thCAWI8,6296,0241,156338
9thCATI10,0964,3211,99295
10thCATI9,0904,1929,0900
sum17,9105,469

Table 12:

Difference in the distribution of Y according to some covariates in the leaver and retained population. Here, categorical variables with low number of missing values are appropriate to be tested.

retained persons(C=0)/ 12,441panel leavers(C=1)/ 5,469total sample/17,910
dropout(Y=2)graduate(Y=1)dropoutgraduate dropoutgraduate
gender
male5.9%67.6%5.3%7.0%5.7%48.9%
female4.5%70.9%4.0%9.8%4.3%52.4%
difference1.4%–3.3%1.3%–2.8%1.4%–3.5%
subject field (*difference for mathematics and linguistics)
engineering7.1%73.6%6.5%7.6%6.9%52.2%
mathematics5.9%71.5%5.3%8.7%5.7%53.4%
law4.7%74.8%3.5%9.6%4.3%52.7%
linguistics4.7%65.9%4.1%8.7%4.5%49.1%
difference*1.2%5.6%1.2%0.0%1.2%4.3%
immigration
no4.9%71.1%4.5%9.1%4.8%53.1%
yes5.7%64.3%4.5%7.5%5.2%43.9%
difference–0.8%6.8%0.0%1.6%–0.4%9.2%
family life
with biol. par.4.8%70.4%4.1%9.1%4.6%52.1%
else6.5%64.5%6.7%6.1%6.7%44.5%
difference–1.7%5.9%–2.6%3.0%–2.1%7.6%
type of school attended
up. sec. educ.3.9%70.4%3.7%9.5%3.9%53.1%
other types9.2%67.1%6.5%6.7%8.2%45.6%
difference–5.3%3.3%–2.8%2.8%–4.3%7.5%
completed vocational training before study
yes8.5%71.2%6.7%8.1%8.0%49.8%
no4.1%69.2%3.7%8.9%4.0%51.4%
difference4.4%2.0%3.0%–0.8%4.0%–1.6%
dropout from training before university
yes9.5%62.9%7.5%6.7%8.7%40.4%
no4.9%69.9%4.3%8.8%4.7%51.4%
difference4.6%–7.0%3.2%–2.1%4.0%–11.0%

Table 13:

Tests on mean difference and test on independence between some determinants and the status of panel leavers. We test variables with low number of missing values.

known statusunknown statusp-value
meanstd. err.meanstd. err.t-test on meanchisq.test on
differenceindependence
generation status3.610.843.530.930.012***0.111
immigration0.240.430.270.440.0950.114
repeated classes0.230.500.220.490.5430.899
birth year1988.164.261988.424.110.1160.402
gender0.370.480.410.490.1000.112
vocational training0.280.450.240.430.036***0.033***
dropout from training before study0.050.220.050.210.6090.663
at least one field change0.050.230.050.230.9861
at least one uni change0.030.160.030.160.8070.909
at least one degree change0.010.120.020.150,0630.157
subject field0.197
family life0.850.360.850.360.7600.800
school leaving qualification1.760.541.790.500.3100.09
direct costs of higher education3.421.033.361.020.1240.169
informed about study3.580.823.580.820.9830.986
opportunity costs2.981.052.991.000.7280.067
mother qualification4.612.214.682.190.4300.210
father qualification5.022.335.012.380.8930.197
mother job50.7819.8651.1719.820.6840.988
father job53.4021.9453.0122.510.6980.670
grade on school leaving qualification2.390.602.370.610.33890.082
type of high school0.740.440.750.440.9240.961
German as graduation exam0.770.420.780.410.7410.778
mathematics as graduation exam0.770.420.770.420.6950.734

  1. *** statistically significant at 5%-level

Table 14:

Attributes description.

AttributeDescription (Data type)
Pre-study
genstatGeneration status (numeric: from 1 = 1st generation to 4 = no immigration background)
Number_Dropouts = 871, Number_Graduates = 9139
immigrationDo you have an immigration background? (binary: 0 = No, 1 = Yes)
Number_Dropouts = 871, Number_Graduates = 9139
rep_classHow many class years have you ever repeated? (numeric: from 0 to 4)
Number_Dropouts = 871, Number_Graduates = 9138
ger_prepTo what extent had you acquired German knowledge and skills before starting university? (numeric: from 1 = not at all to 4 = very much)
Number_Dropouts = 487, Number_Graduates = 6564
math_prepTo what extent had you acquired maths knowledge and skills before starting university? (numeric: from 1 = not at all to 4 = very much)
Number_Dropouts = 450, Number_Graduates = 5924
familylifeWith whom did you spend most of your childhood up to the age of 14? (binary: 1 = with biological parents, 0 = else)
Number_Dropouts = 871, Number_Graduates = 9136
school_typeType of school attended (binary: 1 = upper secondary education, 0 = other types)
Number_Dropouts = 838, Number_Graduates = 8947
qualif_maxSchool-leaving qualification obtained (numeric: 2 = general university entrance qualification, 1 = university of applied science entrance qualification, 0 = other degrees)
Number_Dropouts = 870, Number_Graduates = 9136
grade_schoolApproximate overall grade awarded in the school-leaving certificate (numeric: from 1 to 5)
Number_Dropouts = 842, Number_Graduates = 8976
exam_germanWas German an examination subject for your school-leaving qualification? (binary: 0 = No, 1 = Yes)
Number_Dropouts = 745, Number_Graduates = 8633
exam_adv_germanGerman as first examination subject for your school-leaving qualification (binary: 0 = No, 1 = Yes)
Number_Dropouts = 752, Number_Graduates = 8669
exam_mathsWas maths an examination subject for your school-leaving qualification? (binary: 0 = No, 1 = Yes)
Number_Dropouts = 743, Number_Graduates = 8646
exam_adv_mathsMaths as first examination subject for your school-leaving qualification (binary: 0 = No, 1 = Yes)
Number_Dropouts = 752, Number_Graduates = 8673
genderGender of the person (binary: 1 = Male or 0 = Female)
Number_Dropouts = 871, Number_Graduates = 9139
birthyearYear of birth of the person (numeric: from 1950 to 1994)
Number_Dropouts = 871, Number_Graduates = 9139
mother_qualifHighest mother’s general school-leaving qualification (numeric: from 0 = No school leaving qualification to 8 = Highest tertiary education)
Number_Dropouts = 862, Number_Graduates = 9088
mother_jobMother’s occupation (ISEI-08) (numeric: from 11.74 to 88.96)
Number_Dropouts = 638, Number_Graduates = 6733
father_qualifHighest father’s general school-leaving qualification (numeric: from 0 = No school leaving qualification to 8 = Highest tertiary education)
Number_Dropouts = 833, Number_Graduates = 8953
father_jobFather occupation (ISEI-08) (numeric: from 11.74 to 88.96)
Number_Dropouts = 677, Number_Graduates = 7184
voctrainCompleted vocational training before university (binary: 0 = No, 1 = Yes)
Number_Dropouts = 871, Number_Graduates = 9139
fail_prestudyHave you ever dropped out from training before university? (binary: 0 = No, 1 = Yes)
Number_Dropouts = 871, Number_Graduates = 9139
Decision phase
fieldofchoiceEnrolled in the subject of first choice (binary: 0 = No, 1 = Yes)
Number_Dropouts = 621, Number_Graduates = 7072
institofchoiceTake up the degree at the institute of higher education of choice (binary: 0 = No, 1 = Yes)
Number_Dropouts = 649, Number_Graduates = 7438
study_alternativeWould you rather have started something else instead of a degree? (binary: 0 = No, 1 = Yes)
Number_Dropouts = 648, Number_Graduates = 7426
study_judge_parentWhat do your parents think about the fact that you are studying? (numeric: from 1 = does not apply at all to 5 = applies completely)
Number_Dropouts = 649, Number_Graduates = 7446
study_judge_friendWhat do your friends think about the fact that you are studying? (numeric: from 1 = does not apply at all to 5 = applies completely)
Number_Dropouts = 652, Number_Graduates = 7448
info_useful_...Usefulness of information received from parents, friends, current university students, school teachers, professionals employed in the field of interest, media, university counseling, literature, school events, sneak peak at university, job agencies, companies etc. (numeric: from 0 = not used to 4 = very helpful)
Number_Dropouts = 470, Number_Graduates = 6180
study_restrictIs the study subject to admission restrictions or a selection procedure? (binary: 0 = No, 1 = Yes)
Number_Dropouts = 762, Number_Graduates = 7976
Early study phase
satisf_enjoyReally enjoy the studied subject (numeric: from 0 = does not apply to 10 = applies completely)
Number_Dropouts = 399, Number_Graduates = 7866
satisf_conditionsWish better study conditions (numeric: from 0 = does not apply to 10 = applies completely)
Number_Dropouts = 398, Number_Graduates = 7865
satisf_matchDegree course and other obligations hard to match (numeric: from 0 = does not apply to 10 = applies completely)
Number_Dropouts = 399, Number_Graduates = 7865
satisf_wholeOn the whole, satisfied with actual studies (numeric: from 0 = does not apply to 10 = applies completely)
Number_Dropouts = 399, Number_Graduates = 7866
satisf_frustratingExternal circumstances of study are frustrating (numeric: from 0 = does not apply to 10 = applies completely)
Number_Dropouts = 397, Number_Graduates = 7846
satisf_killDegree course is killing me (numeric: from 0 = does not apply to 10 = applies completely)
Number_Dropouts = 399, Number_Graduates = 7864
satisf_interestingDegree course is really interesting (numeric: from 0 = does not apply to 10 = applies completely)
Number_Dropouts = 399, Number_Graduates = 7864
satisf_concernsConcerns of students are not taken into account sufficiently (numeric: from 0 = does not apply to 10 = applies completely)
Number_Dropouts = 397, Number_Graduates = 7849
satisf_tiredDegree course often makes feel tired and exhausted (numeric: from 0 = does not apply to 10 = applies completely)
Number_Dropouts = 399, Number_Graduates = 7865
partic_peopleParticipation in university events aimed at getting to know people (binary: 0 = No, 1 = Yes)
Number_Dropouts = 753, Number_Graduates = 7937
partic_orgaParticipation in university events on study organization (binary: 0 = No, 1 = Yes)
Number_Dropouts = 746, Number_Graduates = 7865
partic_facilParticipation in university events on the use of central facilities (binary: 0 = No, 1 = Yes)
Number_Dropouts = 740, Number_Graduates = 7789
partic_courseParticipation in university events on bridging courses (binary: 0 = No, 1 = Yes)
Number_Dropouts = 748, Number_Graduates = 7867
partic_acadskillsParticipation in university events on academic skills (binary: 0 = No, 1 = Yes)
Number_Dropouts = 744, Number_Graduates = 7769
preparationHow can you rate your preparation at the start of the university in work techniques, fundamental academic methods etc.? (numeric: from 0 = bad to 4 = good)
Number_Dropouts = 548, Number_Graduates = 7132
skills_prepNecessary knowledge acquired in maths, German, English and computer science before university (numeric: from 1 = not at all to 4 = very much)
Number_Dropouts = 544, Number_Graduates = 7096
workload_matchStudy progress (number of courses, credits earned) match to the curriculum plan (numeric: from 1 = much less to 5 = many more)
Number_Dropouts = 388, Number_Graduates = 6944
performance_evalSatisfaction with the academic performances till yet (numeric: from 1 = does not apply at all to 4 = applies completely)
Number_Dropouts = 426, Number_Graduates = 6999
probsuccessYour opinion on the probability that you will graduate (numeric: from 1 = very unlikely to 5 = very likely)
Number_Dropouts = 425, Number_Graduates = 6978
selfconceptPerception of your talent for studying (numeric: from 1 = low to 7 = high)
Number_Dropouts = 423, Number_Graduates = 6921
study_informedHow well you are informed about the possibilities, limitations etc for your degree course? (numeric: from 1 = very poor to 1 = very good)
Number_Dropouts = 856, Number_Graduates = 9124
socint_instructorsAcceptance by instructors and getting along well with them (numeric: from 1 = does not apply to 4 = applies completely)
Number_Dropouts = 426, Number_Graduates = 6998
socint_studentsSuccessful in establishing contacts and getting along well with classmates (numeric: from 1 = does not apply to 4 = applies completely)
Number_Dropouts = 425, Number_Graduates = 6986
commit_necessaryCommitment to degree course: Do no more than necessary (numeric: from 1 = does not apply to 5 = applies completely)
Number_Dropouts = 424, Number_Graduates = 6977
commit_enjoyCommitment to degree course: enjoyment of degree program (numeric: from 1 = does not apply to 5 = applies completely)
Number_Dropouts = 424, Number_Graduates = 6962
commit_demandsCommitment to degree course: High demands on self (numeric: from 1 = does not apply to 5 = applies completely)
Number_Dropouts = 423, Number_Graduates = 6954
commit_identificatCommitment to degree course: Identification with degree program (numeric: from 1 = does not apply to 5 = applies completely)
Number_Dropouts = 421, Number_Graduates = 6938
helplessnessYou think you will never get better grades (numeric: from 1 = does not apply to 5 = applies completely)
Number_Dropouts = 420, Number_Graduates = 6921
job_semesterNumber of hours spent in a week during semester time for employment (numeric: from 0 to 60)
Number_Dropouts = 434, Number_Graduates = 7064
study_semesterNumber of hours spent in a week during semester time for study-oriented activities (numeric: from 0 to 60)
Number_Dropouts = 434, Number_Graduates = 7069
job_breakNumber of hours spent in a week during semester break for employment (numeric: from 0 to 60)
Number_Dropouts = 434, Number_Graduates = 7061
study_breakNumber of hours spent in a week during semester break for study-oriented activities (numeric: from 0 to 60)
Number_Dropouts = 434, Number_Graduates = 7059
costs_directHow difficult is it to pay for direct costs of higher education? (numeric: from 1 = very difficult to 5= very easy)
Number_Dropouts = 858, Number_Graduates = 9119
costs_opportunityLimitation of the possibilities to earn own money and supporting yourself up until graduation (numeric: from 1 = not at all to 1 = a lot)
Number_Dropouts = 857, Number_Graduates = 9110
financialaid_bafoegCurrently receive student financial aid (BAföG)? (binary: 0 = No, 1 = Yes)
Number_Dropouts = 201, Number_Graduates = 3093
fundingAmount of money at your disposal on average each month in Euros (numeric: from 0 to 10900)
Number_Dropouts = 387, Number_Graduates = 6899
change_fieldHave you ever changed the study field at least once in the past? (binary: 0 = No, 1 = Yes)
Number_Dropouts = 871, Number_Graduates = 9139
change_uniHave you ever changed the university type at least once in the past? (binary: 0 = No, 1 = Yes)
Number_Dropouts = 871, Number_Graduates = 9139
change_degreeHave you ever changed the type of your degree at least once in the past? (binary: 0 = No, 1 = Yes)
Number_Dropouts = 871, Number_Graduates = 9139

Supplemental Material

The online version of this article offers supplementary material (DOI:https://doi.org/10.1515/jbnst-2019-0006).


Received: 2019-01-31
Revised: 2019-11-15
Accepted: 2019-11-20
Published Online: 2020-02-11

© 2020 Oldenbourg Wissenschaftsverlag GmbH, Published by De Gruyter Oldenbourg, Berlin/Boston

Scroll Up Arrow