Teacher practice under the structural challenges of academic second chance education

Figures & data

Figure 1. Pathways to the eligibility to study via first and second-chance education in Germany.

Figure 2. Teacher practice and structural constraints.

Table 1. Sample mean and latent class profiles of teacher practices (two-group solution, imputed data set, N = 176).

Item		Antinomy	M (%)^a	All	Pedagogues (62.50%)	Specialists (37.50%)
				N = 176 M (SE)	N = 110 M (SE)	N = 66 M (SE)
1	‘For me, the personal reasons for absences are not relevant.’	N-O	2.8	3.58(.07)	3.84 (.08)	3.20 (.12)
2	‘I am consistent about punishing disciplinary offences in class.’	N-O	4.6	2.73 (.07)	2.77 (.09)	2.67 (.12)
3	‘I am consistent about recording absences.’	N-O	1.1	1.16 (.04)	1.18 (.06)	1.12 (.07)
4	‘I make exceptions to the rules for attendance, e.g. for students with a child.’	N-O	3.4	2.84 (.10)	2.53 (.11)	3.35 (.16)
5	‘I don't have fixed deadlines. Students can turn in their homework when they have finished it.’	N-O	3.4	3.84 (.08)	3.57 (.11)	4.29 (.11)
6	‘I refuse to negotiate grades with students.’	N-O	2.3	2.25 (.09)	2.27 (.12)	2.22 (.15)
7	‘If students ask me, I give them additional exercises to take home, which I then correct.’	D-A	1.1	2.16 (.09)	1.68 (.08)	2.96 (.13)
8	‘I let faster students move on their own to the next task while I am still practising with weaker students.’	D-A	1.1	2.72 (.08)	2.46 (.09)	3.14 (.12)
9	‘I adapt the homework to the different achievement levels of my students.’	D-A	1.7	3.77 (.07)	3.45 (.09)	4.32 (.10)
10	‘I make sure that all the students have understood the subject matter, before I start a new topic.’	D-A	2.3	2.70 (.07)	2.40 (.08)	3.21 (.12)
11	‘I also talk to my students about their personal problems.’	P-D	1.1	2.57 (.07)	2.19 (.08)	3.21 (.10)
12	‘I know which students have personal problems and I ask them about them.’	P-D	2.3	2.63 (.07)	2.27 (.07)	3.22 (.10)
13	‘I attach a great deal of importance to a high performance level.’	ACH	0.6	2.56 (.06)	2.60 (.07)	2.50 (.10)
14	‘For weaker students, I orient my teaching towards the entrance qualification for universities of applied sciences as this is more realistic’^b	ACH	5.7	4.18 (.07)	3.99 (.10)	4.50 (.08)
15	‘I consistently base my teaching on academic requirements.’	ACH	1.7	1.97 (.06)	2.07 (.07)	1.80 (.08)
16	‘I make sure that students start the final exams with favourable grades’^c	ACH	15.9	3.48 (.09)	3.18 (.12)	3.98 (.13)

Note: N-O, negotiation vs organization; D-A, dependence vs autonomy; P-D, proximity vs distance; ACH, academic achievement level. Coding: 1 = always – 5 = never.

^a Percentage of missings.

^b The formal goal of academic SCE is the eligibility to study at a university, which is awarded after 3 years. The eligibility to study at a university of applied sciences can be awarded after 2 years.

^c Since grades prior to the final (central) exams count for the final grade on the certificate, teachers may give students better grades than justified by their achievement to enable them to graduate from SCE. The standard errors indicate whether or not there are significant differences between the groups: If adding to and subtracting from the two adjacent means leads to an overlap of these means, the observed difference is not significant.

Table 3. Latent class profiles of teachers and students of expectations on teacher practices (imputed dataset, N = 776).

Item		M T^a	M S^b	All	Group 1	Group 2	Group 3	Group 4	Group 5
				N = 776	N = 284 (36.55%)	N = 191 (24.58%)	N = 176 (22.65%)	N = 88 (11.33%)	N = 38 (4.89%)
		%	%	Mean (SE)	Mean (SE)	Mean (SE)	Mean (SE)	Mean (SE)	Mean (SE)
If, due to difficult personal circumstances, a student misses the first lesson, this should not be recorded. – If, due to difficult personal circumstances, a student misses the first lesson, this should of course be recorded.	N-O	1.7	5.3	3.73 (.08)	1.99 (.06)	5.99 (.08)	3.54 (.14)	4.84 (.20)	3.66 (.35)
If a student is close to failing in a school subject, his/her personal circumstances should be taken into account before deciding on the grade – If a student is close to failing in a school subject, only achievement criteria can be taken into account when deciding on a grade.	N-O	5.1	5.8	3.45 (.07)	2.53 (.09)	3.84 (.13)	3.52 (.12)	4.70 (.20)	5.13 (.30)
Teachers should purposefully help students to quickly overcome motivation problems. – Students are themselves responsible for overcoming motivation problems.	D-A	0.6	5.7	2.91 (.05)	2.26 (.07)	2.31 (.08)	3.56 (.09)	3.82 (.18)	5.76 (.16)
Teachers should pay attention to students as persons, for instance by asking them how they are – Teachers should only pay attention to a student’s achievements and progress.	P-D	1.1	5.2	3.31 (.07)	2.78 (.10)	2.48 (.11)	3.93 (.12)	4.71 (.21)	5.37 (.24)
A teacher should be professionally competent and impressive as a person – Teachers should only be professionally competent^c	P-D	0.6	5.7	2.42 (.06)	1.42 (.03)	1.45 (.04)	3.44 (.04)	6.05 (.09)	1.66 (.09)
Proportion of students (%, of all students (N = 600))				100	38.94	15.31	25.79	13.81	6.16
Proportion of students (%, of group)					82.39	48.17	88.07	94.32	97.37
Proportion of teachers (%, of all teachers, N = 176)				100	28.41	56.25	11.93	2.84	0.57
Proportion of teachers (%, of group)					17.61	51.83	11.93	5.68	2.63

Note: Teacher questionnaire. The items for students were formulated similarly. Higher values indicate an orientation towards institutional demands. N-O, negotiation vs organization; D-A, dependence vs autonomy; P-D, proximity vs distance. Values range from 1 to 7.

^a Percentage of missings, teacher sample.

^b Percentage of missings, student sample.

^c We assign this item to the antinomy ‘proximity vs distance’ because the personal qualities of a teacher could be of significance, especially if a more personal relationship with students is preferred. For the explanation of the function of SE see Table 1.

Table 4. Summary of main research findings.

	Research question	Main finding
(1)	What are the types of teacher practice that can be distinguished in academic SCE, and can a dominant practice be identified?	Two groups of teachers are identified: The larger group (‘pedagogues’, 62.5%) consists of teachers who maintain academic goals while still being person-oriented. The smaller group (‘specialist teachers’, 37.5%) consists of teachers who tend to prefer distanced relationships with students and emphasize disciplinary rules.
(2)	How is the practice of an individual teacher influenced by the student composition at their school?	The more teachers perceive students as having non-adequate cognitive prerequisites, the more likely it is that they belong to the group of specialist teachers, meaning that they are more likely to decide not to support the learning of their students.
(3)	To what extent and in what aspects do teachers make their practice fit the expectations of their students?	There is only a partial match between teachers and students. 85% of the teachers and 55% of the students have in common that they expect teachers to be oriented towards a personal relationship, learning support and the adaptation of rules to a student’s personal circumstances. However, the majority of teachers are still more discipline-oriented than most of the students expect them to be. Finally, 20% of the students expect a distanced relationship and autonomy from teachers, a view shared by almost none of the teachers.

Table 2. Latent class profiles of teacher practice and covariates (imputed dataset, N = 176).

Reference group = group 1 (Pedagogues)	M1		M2
	Coef. (SE)	p	Coef. (SE)	P
Individual characteristics
Female	−1.07 (.43)	.012	−1.22 (.46)	.008
At least one parent graduated from university	−.26 (.42)	.536	.53 (.46)	.248
Years of work experience in SCE	.02 (.02)	.412	.02 (.02)	.347
Previous work experience at an academic institution	1.32 (.51)	.009	1.45 (.55)	.008
Teaching STEM subjects	1.20 (.43)	.005	1.37 (.47)	.003
Perception of student composition
% of students who do not fulfil the academic requirements			.03 (.01)	.030
% of students with serious personal and/or family problems			−.03 (.01)	.087
−LL	−4480.589		−4477.4792
BIC	9286.919		9291.040

Note: If p ≤ .05, the effect is significant on the 5%-level. Negative coefficients decrease, and positive coefficients increase the likelihood of belonging to the group of specialist teachers instead of pedagogues. Example: Female teachers are significantly less likely (−1.07/.012) to belong to the group of specialist teachers than to that of pedagogues.

Table A1. Fit statistics and group size for different group solutions (LCA).

	BIC		Number of groups < 5%		Number of groups with N < 50		Min average LC posterior probability^a
	1	2	1	2	1	2	1	2
One group	1970.726	15,651.22	0	0	0	0	.10	.10
Two groups	1897.223	15,211.62	0	0	0	0	.92	.92
Three groups	1900.129	15,061.65	0	0	1	0	.87	.90
Four groups	1899.901	14,989.07	0	0	2	0	.84	.81
Five groups	1904.311	14,928.98	1	1	3	1	.86	.80
Six groups	1904.566	14,935.08	1	2	5	2	.79	.77
Seven groups	1931.521	14,901.93	1	1	6	1	.88	.73

^a Minimum average latent class posterior probability.

1 = LCA on teacher practices (only teachers). BIC is the lowest for a solution with two and five groups. However, two groups of the five-group solution are very small (N = 8, N = 17), and according to Weller, Bowen, and Faubert (2020) and Muthén and Muthén (2006), we chose the two-group solution (see bold number).

2 = LCA on expectations of teacher practices (teachers and students). BIC is lowest for a solution with five and seven groups (Table 1), and for each of these solutions, there is one very small group (N < 50; < 5%). However, the average latent class posterior probability for the seven-group solution is below .80 for two groups, while for the five-group solution, there is no group with a probability below .80. Weller, Bowen, and Faubert (2020, 294) state that values below .80 are unacceptable, and this as well as the good interpretability (Weller, Bowen, and Faubert (2020, 292) justifies the choice of the five-group solution (see bold number).

Table A2. Fit statistics, group size and example items, unimputed data set.

Teachers-only sample					Teacher and student sample
LCA on teacher practices	BIC	group size of preferred solution	%	Example item (Mean, SE)	LCA on expectations of teacher practices	BIC	group size of preferred solution	%	Example item (Mean, SE)
		Two-group solution		‘I also talk to my students about their personal problems.’			Five-group solution		‘Teachers should pay attention to students as persons, for instance by asking them how they are’
One group	9049.877	group 1:	65.14	2.25 (.09)	One group	14987.07	Group 1:	35.55	2.74 (.10)
Two groups	8979.049	Group 2:	34.86	3.20 (.11)	Two groups	14559.91	Group 2:	24.90	2.49 (.12)
Three groups	9010.019				Three groups	14411.49	Group 3:	23.57	3.93 (.12)
Four groups	9063.259				Four groups	14345.22	Group 4:	11.19	4.81 (.22)
Five groups	9112.896				Five groups	14286.08	Group 5:	4.79	5.31 (.25)
Six groups	8975.044				Six groups	14251.08
Seven groups	9024.654				Seven groups	14236.76

Weller, B. E., N. K. Bowen, and S. J. Faubert. 2020. “Latent Class Analysis: A Guide to Best Practice.” Journal of Black Psychology 46 (4): 287–311. https://doi.org/10.1177/0095798420930932.

 Web of Science ®Google Scholar

Muthén, B., and L. Muthén. 2006. “Integrating Person-Centered and Variable-Centered Analyses: Growth Mixture Modeling with Latent Trajectory Classes.” Alcoholism: Clinical & Experimental Research 24 (6): 882–891. https://doi.org/10.1111/j.1530-0277.2000.tb02070.x.

 Google Scholar